REKINDLING REFORM Ia Analysis for Californ PACE Education 6 0 O 2 N nlu T A c U D E M N R O rlr II.— A C m cho 55’: 6%»? £00!) IN CALIFORNIA EDUCATION 2006: REKINDLING REFORM INS I ITUTE CF QQv: DTI JMENTAL .“T |"‘” ' 5 ‘~ ’ 7 :RV VINHOJHVQ dOAJISEjEAiNQ ‘3 ‘3 E 83:3 UNIVERSITY CJI" CALIFORNIA A} H ,M :‘--I I03 ILj,’ IIS TOO IO BifliIiSNI IACE Education . A A A Q... . \I'A Q." I Rekindling Reform — Crucial Issues Facing California’s Schools Educators and policymakers have accomplished much over the past decade, aided by parents and taxpayers. Cali- . fornia has set a high bar for what all children are expected to learn, from kindergarten forward. Instructional materi- als and tests are now closely aligned, so that teachers know how to gauge their students’ progress. And achievement growth within thousands of elementary schools has been strong since these accountability tools were forged in 1999. Yet this is no time to be complacent. Earlier gains in test scores are leveling off in some grades. Achievement gaps between children from poor and middle—class families have failed to narrow. California students continue to learn at slower rates than students in other states, like New York or Texas, where family demographics are similar. We have constructed a high—standards education system, with high-stakes testing for students, but with a finance system that fails to match standards and accountability demands. Every few years PACE commissions analyses by top scholars of crucial issues facing California’s public schools. This 2006 edition comes at a pivotal time. A new legislature is arriving in Sacramento. The governor is starting a four-year term. Next year in Washington, the Congress will review and adjust the No Child Left Behind Act. These scholars, after illuminating the problems, put forward a variety of policy options. They are not shy in suggesting how educators might advance improvements locally, as well. Indeed, one crucial issue is whether Sacramento policy makers will continue to centralize authority within the state capital, or whether the capacity of local districts will be advanced. California can rekindle school reform by building from recent success and squarely facing new challenges. We conclude this volume by suggesting four policy shifts — l Reorient school accountability and finance to raise the achievement of low—performing students. I Pursue a coherent strategy for boosting the performance of English learners. I Make school finance simple, transparent, and adequate for a high—standards public school system. I Focus the state’s role on tracking district and school performance, then deregulate authority over resources down to the school level. The analyses contained in this volume support two key messages. First, a school reform agenda that fails to close achievement gaps is simply not working. Second, the current policy approach is long on rules and short on resources, an order that’s unlikely to motivate long—term gains inside schools. ’6 This volume results from a major effort to better inform California’s education debates with empirical evidence and sound analysis. This series of studies, accompanied by seminars with state policymakers, education associations, and leading educators, was cooperatively supported by the William and Flora Hewlett Foundation, Bill and Melinda Gates Foundation, James Irvine Foundation, and Noyce Foundation. Editorial production of Crucial Issues was led by Haleh Hatami, aided by Elisabeth Woody who directed the over— all project. Several leading analysts served as reviewers of individual chapters, including Michal Kurlaender, Martin Carnoy, Gloria Rodriguez, John Yun, Heather Rose, and Russ Rumberger. We also want to express deep gratitude to Jennifer Kuhn and Paul Warren at the Legislative Analyst’s Office in sharing data. Former PACE Codirector Mike Kirst served as a steady coach and reviewer for the volume. We thank John Mockler and Rick Simpson for their criti— cal feedback. Any errors of omission or interpretation belong to the authors, not our reviewers. Many thanks to Bob Hass for editing the chapters and to Joanne Klein for her artistic sense and design of the volume. Additional PACE staff played essential roles: Aimee Scribner made sure we were up to date on developments in Sacramento. Doug Kearney provided logistical support. Joseph Wright pulled together data and bird-dogged countless facts. And many thanks to Mike Smith and Kristi Kimball at the Hewlett Foundation for their unflagging support. Policy Analysis for California Education PREFACE AND ACKNOWLEDGMENTS i CHAPTER 1 - Rules and Resources: The Evolving Context for School Reform 1 Haleh Hatami, PACE PART I — EDUCATION FINANCE IN CALIFORNIA CHAPTER 2 — Funding California’s Schools, Part |: Past, Present, and Future? 15 LAWRENCE O. Picus, University of Southern California CHAPTER 3 — Funding California’s Schools, Part II: Resource Adequacy and Efficiency ................................... 27 Jay G. Chambers and Jesse D. Levin, American Institutes for Research (AIR) PART II - PUBLIC SCHOOL ACCOUNTABILITY TODAY CHAPTER 4 - California’s Accountability System 51 Catherine Bitter and Jennifer O’Day, American Institutes for Research (AIR) CHAPTER 5 — Evaluating State Intervention: The High Priority Schools Grant Program .................................. 75 Thomas Timar, University of California, Davis CHAPTER 6 — Achievement and Attainment: The Comprehensive High School and the Problem of Reform 91 Leslie Santee Siskin — New York University PART III — MEETING THE NEEDS OF CALIFORNIA’S STUDENT POPULATION CHAPTER 7 — Crucial Issues in Preparing Teachers of English Learners 103 Julie Maxwell—Jolly and Patricia Gandara, University of California, Davis CHAPTER 8 — Addressing the Needs of Low-Income Students 119 Elisabeth Woody and Melissa Henne, PACE CHAPTER 9 — Expanding and Improving Preschool 133 Margaret Bridges, PACE CHAPTER 10 — Rekindling Reform 145 Bruce Fuller, PACE CONTRIBUTORS PAGE 151 PACE is an independent research center that aims to enrich education policy debates with sound analysis and hard evidence. From preschool, to K—12, to higher education we are dedicated to carefully defining issues and assessing the effectiveness of policies and programs. PACE provides analysis and assistance to California policymakers, educators, and the public, as well as issues of national significance. Policy Analysis for California Education University of California Berkeley and Davis Stanford University www.pace.berkeley.edu is for California E cation . ALIFORNIA’S TEACHERS AND SCHOOL LEADERS HAVE MADE LONG STRIDES IN RAISING YOUNG CHILDREN’S BASIC SKILLS. THE SHARE OF FOURTH—GRADERS MEETING THE STATES DEMANDING PROFICIENCY STANDARD IN MATHEMATICS HAS CLIMBED FROM 36 PERCENT TO 50 PERCENT SINCE SACRAMENTO POLICYMAKERS ENACTED AGGRESSIVE ACCOUNTABILITY REFORMS IN 1999. Progress in boosting reading skills has proven more difficult to achieve. Still, tens of thousands of additional students—those in the lower elementary grades—display stronger reading skills than before. Educators have worked hard to realize these gains—even as California continues to invest less in per pupil funding than other states. The Public Schools Accountability Act (PSAA) of 1999, crafted by former Governor Gray Davis and the Legislature, has delivered clearer learning aims, grade by grade, along with tighter alignment between the intended curriculum and how students are tested. Sacramento has targeted fresh dollars to low performing schools with the goal of closing achievement gaps among diverse students. Policymakers at first offered incen- tives for schools showing marked progress, although these carrots disappeared as the state’s budget deficit worsened. Adding to Sacramento’s accountability program, President Bush signed the No Child Left Behind (NCLB) Act in 2002, complicating—some would say undercutting—the state’s own efforts. While making remarkable progress on several fronts, California’s reform initiative may be losing steam. Several signs indicate that the initial motivat— ing force of standards—based accountability may be waning. When gauged by the federal standard for proficiency in math and reading, the performance of our elementary school students has largely leveled—off. Figure 1 shows both commendable gains in achieve— ment in past years, along with signs that test scores have climbed up to a plateau.1 Classroom teachers have responded to a grow— ing array of new demands and challenges crafted in Sacramento. But significant challenges remain. Cali— fornia is now implementing among the highest cur— ricular standards in the nation, while our schools are propped—up by one of the most rickety school finance systems. We have thousands of young teachers who enter the profession each year, only to leave within five years, often burned out and holding little affection The Serrano school finance decisions moved the Legislature to achieve equity in per pupil spending by the late 19703, just as taxpayers revolted against what they saw as climbing levies on their homes and small businesses. The passage of Proposition 13 in 1978 continues to constrain the overall tax base on which public support of its schools depends. The notion of school reform, until the mid- 19803, centered on equalizing school spending across rich and poor districts (which does not equate with rich and poor communities) and expanding dollars for categorical aid programs. But national gauges of student progress flattened out in the 19803, despite the fact that real spending on education had almost tripled since the late 19505. The national discourse over school reform moved from buying more “inputs” to focusing on how to generate stronger performance from what many saw as an inef— ficient school system. Motivated by the state’s dismal results on the 1994 National Assessment of Education Progress (NAEP), policymakers began building on previous accountability efforts. These included the California Assessment Program (CAP) in the 19705 and early ‘80s, and later the California Learning Assessment System (CLAS), put in place by then state superinten- dent Bill Honig. Having placed next to last nationally in NAEP reading scores in 1994 (above Mississippi), California faced formidable challenges. How did our students find themselves at the low end of the performance scale? Analysts have pointed to the state’s demograph- ics, falling per pupil spending post—Proposition 13 compared with other states, and the lack of coherent curricular goals. Curricular experiments in math and reading were, according to critics, “. . .at best a distrac- tion and at worst quite damaging” (Kirst, 2005). And California’s English learner population was far out- pacing the local schools’ capacity to address language acquisition needs. The net effect of these forces was a system in decline. Two years following California’s poor showing on the NAEP, Wilson introduced the Standardized Test— ing and Reporting System (STAR). Thus began the heavy work of developing rigorous standards in major subject areas for each grade level. Moving toward the ideal of an aligned standards—based system, lawmakers planned for assessments to measure how well students had mastered the standards. Instead, pressure to pro- vide the public with test results led the state to adopt the Sanford 9, an off—the-shelf national test which was not aligned with California’s curricular standards. It wasn’t until 2001 that the California Standards Test was incorporated into state examinations. Governor Gray Davis continued to build on the vision of an aligned system. In 1999 the Public Schools Accountability Act was passed which created an academic performance index (API), measuring each FIGURE 3: The Ideal ofan Aligned Standards—Based Education System describe what students need to learn and be able to do. aligns classroom learning with the standards. Assessments align with standards, and accurately and fairly measure student and school performance, identifying where extra help is needed. Accountability programs use multiple measures to provide incentives for success and interventions where necessary. Teacher preparation & f FINANCIAL professional development \ / RESOURCES assures teachers can teach to the standards. . adequately support change and improvement, with flexibility to meet local needs. Source: EdSource, Aligning California’s Education Reform (2001) CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 school’s performance and growth, a new interven- tion program for so—called underperforming schools, monetary rewards for schools meeting growth targets, and a high school exit exam. Davis supported efforts to create high performance standards, beginning in elementary schools, and high stakes for high school students. The steady centralization of California’s school system is now complete. The bulk of funding now comes from Sacramento, not from locally set prop— erty taxes (which go to the state to be reallocated to schools); the state sets curricular goals and allows one of two curricular packages in many grade levels; and Sacramento attempts to shape targeted programs aimed at closing achievement gaps. All told, combined centrally controlled state and federal aid amounts to more than 80 percent of funds that schools receive. Implicit in this tough-love strategy has been distrust in local decision~making, as veteran policy analyst, Michael Kirst, has emphasized. Nor has much attention focused on how to bolster the capacity of local districts in raising student achievement, since local educators, too often, are viewed as part of the problem, not the solution. Then, on top of Sacra— mento’s accountability system now rest Washington’s NCLB provisions. California has set the bar high when it comes to student performance standards. Earlier this year the Fordham Foundation again ranked California’s curricular standards among the most rigorous in the country. According to a new survey, 73% of Califor— nians polled believe that high school students should pass a high school exit exam in order to graduate (Baldassare, 2006). So, California has not shied away from raising expectations of all students. The ques- tion is whether the intensity of our hopes and rules are matched by the adequacy and efficient use of re- sources. Within two years of Governor Davis signing the Public Schools Accountability Act of 1999, the Con— gress was ironing out the details of what became the most forceful and centralized education initiative ever advanced in the nation’s capital. The No Child Left Behind (NCLB) Act created a parallel accountability system—even a separate way of determining which California schools are meritorious or deemed fail— ing. California would be penalized by Washington for a high bar defining student proficiency—since our Rules and Resources: ' The, Evolving Cantext for Schoiil Refonn :g students now have a much longer staircase to climb before reaching universal proficiency in reading and math by 2013, the deadline set by NCLB. For the first time ever, not only individual schools, but entire districts are held accountable under federal law. By spring 2006, 152 California dis— tricts had fallen under Program Improvement status for failing to meet proficiency targets. These districts face federal sanctions. But the long—term question is, what’s the state’s capacity for large—scale interven- tion from a fiscal and instructional perspective? Can Sacramento offer a clearer reform strategy than what districts have already devised? Policymakers will be weighing the political implications and technical ef- ficacy of a wide scale intervention across school dis— tricts, perhaps involving state takeovers, reconstituting the staff of hundreds of schools, or (as encouraged by NCLB) creating many new charter schools. NCLB has come with significant increases in Title I funding for local schools. But the allocation of these funds is not necessarily linked to performance. Local educators report experiencing harsh sanctions and public embarrassment if their students don’t show stronger performance. But when they do, few rewards flow from NCLB. When any student subgroup—de— fined by family income, ethnicity, or special educa- tion status—doesn’t meet growth targets, the entire schools is deemed “failing” under NCLB. An earlier PACE analysis showed how this tends to hit racially integrated schools especially hard, since they suffer from a much higher probability of hitting one of the trip—wires, bringing federal sanctions. Segregated schools, in contrast, enjoy lower probabilities of being deemed failing (Novak 81 Fuller, 2003). DEMOGRAPHIC SHIFTS California’s public schools educate the largest and most diverse student population in the nation. One quarter of these students are English language learn— ers. The population boom of the 19705 and ‘805 is over, yet California’s student enrollment will continue to grow over the coming decade. As of 2006, Califor— nia’s student population includes 47% Latino stu- dents, 31% White, 11% Asian, and 8% Black students. Table l and Figure 4 reflect both a 2006 demographic snapshot as well as trends over time. PACE W 5 California has the most diverse student population in the world, with more than100 languages spoken in the homes of those students. Today, our student population is ”majority—minority.” Forty- one percent of our students speak a language other than English at home, and a quarter of all California public school students are struggling to learn the English language in school. —State Superintendent of Public Instruction, Jack O’Connell State of Education Address, February 2006 Both state and federal ac- countability programs focus on student subgroups that often lag behind, defined by ethnic- ity, economic status, and special education status. This strategy— requiring schools and districts to track achievement progress by subgroup—has yielded little success in closing achievement gaps. It may have avoided a situ- ation where standards—based ac- countability best served children of middle—class and affluent families. Still, knowing in greater detail that children from poor or non—English speaking families fall behind others is proving in— sufficient in closing gaps. California continues to display one of the highest child poverty rates the country. Chap- ters 7, 8, and 9 provide a closer look at the state’s collective student profile and those poli- cies aimed at serving diverse children. Our authors show that taking the state’s demographic context into consideration means going beyond “being aware” to taking an active stance that incorporates their needs into policy. New analyses show a slowing trend in overall growth statewide. Total enrollment is projected to increase from 6,264,661 students in 2004 to 6,484,243 students in 2014, an increase of 3.5 percent. A down— ward trend in the enrollment growth level is expected until 2009 when the state’s enrollment is projected to grow by about 0.1 percent. Enrollment growth is then expected to slowly increase during the remainder of the projection period (Department of Fi- nance, 2005). A major problem for California, however, is that enroll— ments are shrinking in several urban districts, yet growing sharply in inner-ring suburbs, the Inland Empire, and parts of the Central Valley. So, Sacramento is faced with essentially reallocating a fixed pie from one dis- trict to another. And pressure may build, once enroll- ments decline, to capture budget savings to reduce the deficit or support other public institutions and programs. HIGH STANDARDS, HIGH STAKES ACCOUNTABILITY California policymakers, since 1999, have created high performance standards for local educators and students—then laid down an array of regulations for how schools reach these intended learning outcomes. Many voters and policymakers believe that local edu— cators should be held more strictly accountable for stronger performance, with or without new resources. Others argue that it’s unrealistic to expect that teach- ers and school leaders can meet the high standards and high stakes set by Sacramento without more ad— equate resources. Either way, future school improve— ment efforts will be built from the contemporary rules of accountability and the old finance system. Stiff accountability measures certainly won’t go away as long as the performance of California’s stu— dents remains low, even when compared against other states, like Texas or New York, with similar family demographics. Californians remain dissatisfied with the quality of public education, although many feel good about their own local school (Baldassare, 2006). Despite gains in student scores, California’s public education system ranks at the bottom of the nation on a number of indicators (see Table 2— How Cali— fornia Compares). Yet previous “lows” in California’s education rankings, as we chronicled in Crucial Issues 2000, have fueled innovative reforms, including the current “results—based” accountability system. When TABLE 1: The Numbers — From Quality Counts, 2006 (EdWeek) Public Schools Public School Teachers Students (PreK — 12) Annual PreK — 12 Expenditures 9,222 Children in Poverty 304,31 1 Minority Students 6,413,862 Students With Disabilities $48 bill. English Language Learners CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 RAND analysts recently ranked states according to student achievement levels, taking into account the social-class and language backgrounds of students, California ranked dead last. Other states with student profiles similar to California show much higher per— formance levels. The tandem accountability programs—Sacra— mento’s PSAA and Washington’s NCLB—define the landscape in which future reforms must be situated and pose various challenges for policymakers — I High standards — Sacramento has banked on the admirable goals of pushing all students to meet high standards, while struggling for an effective initiative to close achievement gaps. However, high fixed proficiency targets as defined by NCLB provisions, may simply dash the hopes of educa— tors and students if capacity locally remains con— strained by inadequate resources. I Curriculum and instruction — After devising de— manding learning aims and curricular standards, Sacramento has mandated two acceptable cur— ricular programs in elementary reading. This advances the tight alignment of standards and classroom materials, while narrowing the discre— tion of local school boards and teachers. FIGURE 4: California Public K—12 Enrollment* Rules and Resources: The Evolving Context for School Reform I The California high school exit exam (CAHSEE) — This high-stakes capstone to California’s account- ability system is proving controversial. Account— ability only works if consequences are felt locally. But should students, primarily from low—income families, feel the brunt of the consequences, after receiving 12 years of low-quality schooling? The public heavily favors the exit exam, while it is re- ceiving mixed reviews in the courts.3 I Intervention system — The state’s interventions for low—performing schools show disappointing results for many schools. For the first time ever, entire districts face consequences under NCLB. Districts are increasingly sharing a role in inter— vention. I Academic Performance Index (API) — State and federal accountability systems yield different results. To date, one thousand California schools are meeting Sacramento’s growth standards, but are deemed failing by Washington. This creates confusion for parents and policymakers alike. Last year the Bush Administration announced that it would allow states to track and reward growth in student performance. But the state de— partment of education lacks the technical capac- ity and data system to follow children over time, necessary to tracking achievement growth. 3,500,000 Hispanic 3,000,000 2,500,000 — — — — — — — _ ~ 2,000,000 ~ "' — ._ ‘u . ' ""' White 1,500,000 1,000,000 , Asian/Filipino/Pacific IlllllllIIIIIIIIHIHIIIIIIIIIIIIIIIIll!IIIIIIIIIIIIIIIIIIHIIIIIIIIllIHIIIIIIIHHHHIHIHIIII [Slander 500,000 I“nnllIIIIII‘iu‘H'I" —————————————— Black “WWW American Indian (23‘ O’ '\ 03" °3 <5“ 0? ,0K 6’3 0"; $4 903‘ [,3 ,5 Q) Cb 0) Q)’ Q)’ 0’ (I; V (0’ Q), Q fl; Q) Q) C5 Cb 0) 0) Q Q O No) No3 No) no) No) No, r9 9 Source: California Department of Finance * Multiple race and nonresponses are allocated to the known categories. Actual school enrollment through 2002-03 was provided by the California Department of Education, California Basic Educational Data System, and projected enrollment, by the Department of Finance, Demographic Research Unit. California Youth Authority and State Special Schools are excluded. PACE m 7 California’s accountability system has witnessed positive outcomes on two fronts. As we mentioned earlier, on the whole, student performance on state tests has been steadily rising, at least in the early e1- ementary grades. Also, what is being tested is better aligned with what’s being taught (state curricular standards). Researchers, though, are beginning to ask questions about the plateau in test scores and what this spells for the model’s sustainability. Chapter 4 will delve into results-based accountability and the assumptions behind the current model, and Chapter 3 will address how different funding approaches can assess cost implications. Chapter 5 offers a focused look at school spending and improvement efforts in low—performing schools. UNSTABLE AND INADEQUATE SCHOOL FINANCING The state’s method for financing schools is based on a maze of funding channels and represents an accretion of three generations of court decisions, ballot propo— sitions, and the pursuit of silver bullets for school reform. School finance in California now depends on TABLE 2: How California Compares 1St in nation, student population, 2005. California has 35% more students than US average. 1" in nation, English learner population, 2004 43rd in nation on NAEP Proficiency, 2005 (National Assessment of Educational Progress, ranking reflects scores 4th and 8‘h grade math and reading) (EdWeek) 29‘" in nation graduation rate, 2002 (using Cumulative Promotion index)* 31 st in nation, per pupil spending, 2001 ~02 (latest statistics) * Editorial Projects in Education (EPE) Research reports 71% graduation rate, state reports that 87% of students are graduating, Manhattan Institute reports 65% in 2003. See Chapter 6 for challenges in accurate calculation. Source: National Education Association, Rankings and Estimates, 2004—05 unstable state revenue sources, like capital gains taxes which have fluctuated over the past 15 years. So, as local educators’ feet are being held to fire that grows hotter, they must work with revenues that have become unstable. This is a non—system that only a few insiders understand and which surely erodes the morale of local school boards, principals, and class- room teachers. When the California economy recently declined, with taxes from capital gains going into a temporary tailspin, Governor Davis was forced in 2002 to cut teacher training efforts, eliminate all incentive pay- ments to schools and teachers whose students showed performance gains, and to scale back the state’s target- ed initiative to raise the performance of low-perform— ing schools. An accountability program that began with a mix carrots and sticks suddenly lost its positive incentives. Arnold Schwarzenegger ran in 2003 on a plat— form of fiscal conservatism and budget reform. His fiscal philosophy created animosity among educa— tion advocates when he suspended the Proposition 98 funding guarantee for the schools to narrow the 2004—05 budget deficit (see Chapter 2). He promised lawmakers to make it up the next year, but reneged on the promise. Adding salt to the wound, he backed Proposition 76 in 2005 that would give him the power to make “across the board” cuts in state spend— ing when there is a budget deficit, including cuts to Proposition 98 guarantees. Voters rejected it and every other ballot measure that appeared on that special election slate. By early 2006 the economic skies had cleared somewhat. Capital gains taxes were up, along with state income tax revenues. This unexpected windfall added $7.5 billion to the state general fund, allowing the governor to boost K—12 spending by more than 11 percent in the current fiscal year. Schwarzenegger also settled a court challenge over his ignoring the Prop 98 funding guarantee, agreeing to spend $2.9 billion over the next several years, tentatively set to reduce class sizes in the upper elementary and middle—school grades. The new budget also increases funding to vo— cational education, arts, and physical education, in a move to counter the state accountability system which has been criticized for narrowing curricular focus by testing demands. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Despite the funding reprieve, in 2006, the state’s economy is still trying to catch up to the population booms and economic changes of the last thirty years. California still faces a $3 to $4 billion structural defi- cit, where spending obligations outstrip anticipated revenues. A major class action suit settled in 2004 drew public attention to the disparity between funding and demands for high achievement. Williams v. State of California raised questions about how the state could hold schools accountable to high standards without providing adequate inputs to support schools in reach- ing proficiency goals. Plaintiffs from 18 school districts argued that lack of basic resources like sound facilities, sufficient materials, and qualified teachers, left stu— dents in low—income schools at an unfair disadvantage. Williams serves as a reminder that new policies aimed at improving student achievement must consider the context in which learning takes place. The Williams settlement is not a structural remedy to the school finance system. To understand “how much is adequate,” researchers and policymak— ers are looking more closely at factors that affect learning. In an unprecedented effort, a large—scale research project commissioned by four major founda- tions, will attempt to clarify the “parts” of California’s school finance system.4 The “Getting Down to the Facts” investigation is tracking the relationship be— tween money, efficiency, and student performance in California. These studies are due out in early 2007, sponsored by the Governor’s Advisory Committee on Education Excellence. FIGURE 5: California Expenditures Per K-12 Pupil (ADA) Californians, still worried about the state of the public schools, are willing to pay some new taxes. They are less likely to approve increases in property tax (24%) or sales tax (36%). But they were more likely to accept income tax increase on the state’s highest earn- ers (64%). All told, however, a vast majority (81%) prefer more efficient use of existing funds to higher taxes to improve the schools (Baldassare, 2006). GOVERNANCE Finally, the evolving context of governance plays a central role in policy development and implementa- tion. The combined state and federal “results-based” school accountability systems place most decisions far from the schoolhouse. The state, strapped with more regulatory power than it has the capacity to enforce, seeks ways to share the burden with districts. Policymakers are faced with a persistent governance dilemma—balancing oversight with trust in local de- cision—making. The Center on Education Policy lists the ways in which the arms of the federal law radically alter gov- ernance dynamics: NCLB affects a range of state and local decisions, both small and large—when and how students take tests, which textbook series districts adopt, which children receive extra attention and how they are grouped, how states and districts spend their own money, how teachers are trained, and $9,000 8229- .8118 1997-98 1998-99 1999-00 2000—01 2001-02 2002—03 2005-04 2004-05 2005-06 2006—07 Source: Legislative Analyst’s Office, September 2006. Values are adjusted to 2006—07 dollars PACE Only one in eight Californians (12%) say they trust state government the most to make spending decisions for local schools. Residents see a host of local authorities—their local school district (36%), teachers (30%), and principals (13%)— as more trustworthy. (Baldassare, 2006) where principals and teachers are assigned to work, to cite just some examples (CEP, 2005). The way California finances its schools remains unrelated to the standards and accountability system. Outside of intervention funds, how a school meets, or fails to meet, standards (mea— sured by proficiency targets for the school and for subgroups within the school), has little connection to how that school receives funds. This fundamen- tal discord makes for conflict in governance at best, inefficiencies and negative student outcomes at worst. Faced with NCLB pressures for specific outcomes within a fixed timeframe, local discretion continues to shrink. Added to this, the school finance system is equally if not more restric— tive, leaving local governing bodies little funding flex— ibility. In the positive column, however, accountability pressures from above have produced more candid talk at the local level about how individual students and groups of students perform, leading to greater collaboration and strategic targeting among many educators. State—driven accountability, in theory, does not preempt local control over schools. The systems en— visioned by PSAA and NCLB were intended as moni— toring and motivational mechanisms for educational quality. In theory, outcomes (test scores and subse— quent sanctions or rewards) would motivate schools to improve, but approaches to improvement would be left up to local control. The same theory held for entire districts as they became accountable under NCLB (see Chapter 4 for further discussion). In practice, though, schools and districts argue that the system ties their hands. Both districts and schools seek more flexibility in how they use funds and instruction to meet the specific needs of their students. Educators point to district use of uniform curricula, and pacing schedules at the elementary level in particular, as a way that the “how” of teaching and learning is being determined by state policy, not local context. At a time when more and more of the “what and how” of teaching is decided at the state and federal house, teachers are being required to adjust. Many find little room for innovation and creativity, work- place qualities that drew them to a career in teaching. Local context, resources, and their own preparation, determine how educators respond to accountability rules. How can policies support improvement without limiting creativity? Voices from the classroom, includ- ing the expertise of the California Teacher Advisory Council (CALTAC), offer insight into the ways policy affects instruction, teacher preparation to serve the needs of language learners and low-income students, as well as workplace conditions (see Chapters 7 and 8 for further discussion). As entire districts are now being held account— able, the state walks the line between the need to intervene and the need for local solutions. How can the state build district capacity to improve schools, so that it can itself build its schools’ capacities, and ulti— mately, meet the immediate needs of failing students? Successful district practice holds promising answers to these questions, and more research is needed on how effective districts operate. Decisions at the district level have the poten— tial to leverage large-scale improvement. The new District Collaborative initiative, funded by several private foundations and coordinated by the Ameri- can Institutes for Research with PACE involvement, hopes to uncover and disseminate what works in districts across the state. A new PACE report details how three diverse districts are effectively closing achievement gaps, crafting quite different strategies (Woody, et al., 2006). In the debate over governance and whose rules count, we ask policymakers, as we did in 2000, wheth— er it’s time for a break. Regulation implies law, and the solution to imperfect rules often means more rules. How can “more rules” be designed and piloted so that they are less onerous to local actors? California’s edu— cation system has a strong foundation built through bipartisanship. Efforts to tackle the state’s pressing education issues are unfolding through the work of two education commissions appointed in 2005. Each committee includes representatives of California’s multiple governance bodies as well as researchers, parents, business and community stakeholders. As with the California Master Plan for Education committees, 10 CRUCIALISSUESIN CALIFORNIA EDUCATION 2006 Rules and Resources: The Evolving Context for School Reform ‘ under the volume’s broader framework of “rules and resources” and appear across the chapters. Our intent is to capture the debate for an au- dience interested not only in the education issues but the policy play that shapes how issues develop, succeed, fail, and affect future direction. The policy recommendations under each issue are rooted in a historical and structural foundation. To the extent possible, we look at how the rhetoric from the state, the academy, and the community play out at the class- it is hoped that by convening a myriad of policy “play- ers” and stakeholders, clarity, transparency, and more efficient and just policy will result. WHAT’S INSIDE CRUCIAL ISSUES 2006 Crucial Issues 2006 delves into a range of issues, most pertaining to how we might rethink and improve school accountability and finance systems. These problems are situated in California’s evolving demo— graphic, economic, and policy contexts, as I have just detailed. From this framing, you may read the volume sequentially. Or, you may select the chapters that best room level. Each chapter will address the following: fit your own interests. The volume is designed so that I What are the historical, persisting issues and how have policymakers tried to address them in the past? I What are the most pressing issues currently and what do we know about their dimensions empiri- each contributor’s chapter stands on its own. We often hear from veteran policymakers that history matters and is too often ignored in the race to fix the state’s education challenges. In this book we take a look at conditions prior to 2000, keeping the state’s political—economy in mind, noting progress, and highlighting policy leverage points. Topics con— cerning teachers and teaching, curriculum, school cally, and alternative policy remedies? I How to address these perennial issues in the long run, more structural long-term remedies? choice, professional development, and governance fall Layers of Governance in California Education Policy The State Board of Education is direct- ed ”to study the educational conditions and needs of the state” and to ”make public plans for the improvement of the administration and efficiency of the public schools of the state.” In designing the State Board of Education, the legis- lature intended to make the Board the ultimate governing and policy making body for the Department of Education, its officers and employees. The Superintendent of Public Instruction is an elected constitutional officer whose responsibility is to su— perintend the schools of the state. The Superintendent is also responsible for implementing Board policies. The State Department of Education is responsible for administering and enforcing the laws pertaining to educa- tion in the state. The CDE also houses, analyzes, and reports on education data. Like other state agencies, the CDE is dependent upon the legislature and the governor for its budget. One consequence of heightened gubernatorial interest has been estab- lishment of the office of Secretary of Education by Governor Wilson in 1991. Prior governors had education advi- sors, but the creation of a cabinet level position indicated a new, more visible and central role for the governor in education. Legislature: Controls school funding in California and passes laws on an array of school policy. County offices of education: Operate schools,juvenile halls, regional oc- cupation centers providing job—related training, special education classes and schools for handicapped students, and environmental education schools. In addition, county offices provide ad- ministrative and supportive services to small local school districts (per State Guide to Government). City Government: In some cases city governments havejurisdiction over local schools. Recent debate over Los Angeles Mayor Antonio Villaraigosa’s attempt to gain control of the local schools revived the contention over who should control classroom curricu— lum and accountability practices. LEA: Local Education Agency. The local school authority. For example, a school district’s administration and board of education. School Site: Run by principal (site leader) and school site councils. Source: Timar (2002) PACE % ll growing minority population. Authors present policy considerations, including fiscal and governance impli- cations, of legislating universal preschool. CHAPTER 10, Rekindling Reform, closes the book with a four-point agenda for tackling the most crucial question facing the California schools: whether the current state reform strategy will sustain achievement growth in the elementary grades, close gaps, and show improvements at the high school level. Concluding thoughts on future policies include addressing the pressing tasks of retooling school finance and offering a more motivating approach to school accountability. ENDNOTES Analysts from the National Center for Research on Evaluation, Standards, and Student Testing (CRESST/ UCLA), comparing 2006 against earlier years, concluded that “the longer an accountability system is in place, the smaller the gains. California may have reached that pla— teau.” PACE’s own analysis of test score trends across 12 states, including California, similarly found that while state officials continue to report gains on their own tests, federal results from the National Assessment of Educational Progress (NAEP) show flattening trend lines since 2001 at the fourth and eighth grade. to This section draws from Recent Evolution of California’s Accountability Policy, by Michael W. Kirst (2005). Alameda County Superior Court judge ruled the exit exam unconstitutional in May 2005. In June, at State Superintendent‘s request, the State Supreme Court over- turned the ruling. Valenzuela vs. O’Connell has gone to the state Court of Appeal. As of press time, 12% of high school seniors failed the CAHSEE. 4 The William and Flora Hewlett Foundation, The Bill and Melinda Gates Foundation, The Iames Irvine Foundation, and the Stuart Foundation. REFERENCES Baldassare, M. (2006). PPIC statewide survey: Special survey on education. San Francisco, CA: Public Policy Institute of California. Baldassare, M., & Hanak, E. (2005), CA 2025 It’s Your Choice, PPIC. San Francisco, CA: Public Policy Institute of California. Bohrnstedt, G. W., 8( Stecher, B. M. (Eds.), CSR Research Consortium. (2002). Capstone report: What we have learned about class size reduction in California. Sacramento: California Department of Education. Cain, B. E., & Kousser, T. (2004). Have term limits affected the California Legislature? PPIC Research Brief, 94. San Francisco, CA: Public Policy Institute of California. Carroll, S.I., Krop, C., Arkes, 1., Morrison, P.A., & Flanagan, A. (2005). California’s K-12 public schools: How are they doing? Santa Monica, CA: RAND. Center on Education Policy, (March 2005). From the capi- tol to the classroom: Year 4 of the national No Child Left Behind Act. Fuller, B., Gesicki, K., Kang, E., 8( Wright, I. (2006). Is the No Child Left Behind Act working? The reliability of how states track achievement. PACE Working Paper Series No. 06-1. Berkeley, CA: Policy Analysis for California Education (PACE). Goldschmidt, P, Boscardin, C . K., & Linn, R. (2006). CRESST analysis of California accountability progress report. National Center for Research on Evaluation, Standards, and Student Testing (CRESST). Greene, 1., & Winters, M. (2006). Leaving boys behind: Public high school graduation rates. Civic Report No.48. New York, NY: Manhattan Institute for Policy Research. Kirst, M.W. (March 2005). Recent evolution of California’s accountability policy, paper for PACE Legislative Seminar. Novak, 1., 8( Fuller, B. (2003). Penalizing diverse schools? PACE Policy Brief. Berkeley, CA: Policy Analysis for California Education (PACE). Policy Analysis for California Education, Crucial issues in California Education 2000: Are the reform piecesfitting together? Berkeley, CA: Policy Analysis for California Education (PACE). State of California. Department of Finance. California pub- lic K-12 enrollment and high school graduate projections by county, 2005 series. Sacramento, CA. Updated December 2005. Timar, T. (2002). You can’t always get whatyou want: School governance in California. Williams Watch Series: Investigating the Claims of Williams v. State of California. (University of California, Los Angeles), 65. Venezia, A., et al. (2005). The governance divide: A report on a four-state study on improving college readiness and suc- cess. San lose, CA: The National Center for Public Policy and Higher Education. Woody, E., Bae, S., Park, 8., 8t Russell, 1. (2006). Snapshots of reform: District efforts to raise student achievement across diverse communities. Berkeley, CA: Policy Analysis for California Education (PACE). CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Lawrence 0. Picus University of Southern California Rossier School of Education HE 20(I6~O7 BUDGET ACT SIGNED BY GOVERNOR ARNOLD SCIIW’A RZEN ECCER INCLUDES $48 BILLION IN FUNDING FOR I<~I2 EDUCATION. ACCORDII‘IC TO THE DEPARTMENT OI} FINANCE (.2006), THIS AMOUNTS TO FBI LEM FOR IbXCH STUDENT IN EVE {AGE DAI L‘i” J A'I"I‘I?T\’Il~\NCE (i-«XDAI ACROSS Tm i L SI'XTE, VIH EN INCLUDING FEDERAL, LO CA L, AN D OTHER SOURCES SWITXTIV“: I‘LTQ‘DING IS THE PRIIYIARY MITCIIANISIVI FOR DISTRIBUTING I{I1\7I{I\l LIES TO LOCAL SCHUOI, DI fl. RI CTS. The magnitude of both the total commitment and the per pupil level of resources devoted to the education of our State’s school children is substantial. Despite this, there is a general feeling across the state that funding levels for K—12 education are too low, especially considering that California’s schools include the largest and most diverse student population among the 50 states. Regardless of the level of funding, there is also a growing concern that California’s system of school finance is increasingly complex and disjointed. Because California has over six million school age children attending public schools, any decisions about the level or type of funding for schools has major implications for the funding of other governmental services, as well as for the level of taxation needed to fund schools for those children. Before making these hard choices, it is helpful to consider how California compares to the rest of the nation in school funding, to review how our state found itself in the current funding predicament, and to consider what options are available to policymakers wrestling with this issue. To accomplish that, this chapter is divided into three sections. The first describes how California compares to other states in funding its schools, focus- ing on both the fiscal resources available to schools, and on the resources those dollars are able to pur— chase in the California economy. The second section briefly traces the history of California school finance in recent years, focusing on issues of equity, pro~ ductivity, and adequacy. Finally, the third section of this chapter discusses the major policy issues facing the state today as it struggles to balance its budget, provide services to all deserving California residents, and meet the future educational needs of our school children. l5 CALIFORNIA SCHOOL FINANCE: HOW DO WE COMPARE? Comparing school finance systems across states is a complex undertaking. Under our Federal system of government, each state is responsible for education, and each has developed its own, unique approach to funding public schools. The result is that it is hard to make direct comparisons across the 50 states. There are three ways to approach the question: 1) How much do we spend? 2) What resources do the dollars buy? 3) How hard are we trying? Each is considered below. How Much Do We Spend on Education? When considering school funding issues, it is helpful to understand the relative size of the public education sector in the United States. Table 1 com— pares estimated K—12 pubic education expenditures for 2005—06 with some of the Fortune 500 companies to help put this into perspective. As the table shows, total K-12 expenditures for that year are estimated to be $415.3 billion or approximately 22% more than the $339.9 billion in revenue earned by the number one company on the Fortune 500 list, Exxon~Mobile. Moreover, the nearly $60 billion in revenue that year for California’s K—12 schools was more than that of State Farm Insurance, Company, the 22nd largest company on the Fortune 500. Table 1 also includes as estimate of the revenue received in 2005—06 by the Los Angeles Unified School District (LAUSD)—the largest district in California and the second largest in the United States. With revenue of $8.2 billion, that TABLE 1: Education and the Fortune 500 Rank Or anization 2005—06 Revenue or g Expenditures (3 Billion) us. Public K—12 Ed. 415-3“ I WaI-Mart 339.9 Calif. Public K—1 2 Ed. 596* 19 State Farm Insurance 59.2 LAUSD 8.2* 230 Reynolds American 3.3 *All Funds **Expenditures 16 district compares favorably with Reynolds Ameri- can ($8.3 billion), which would rank it number 280 among Fortune 500 companies. In terms of total dollars allocated to K—12 Edu— cation, California ranks at the top of all states in the nation. This is simply a result of our size—more than 6 million of the nation’s 48 million plus school children reside in California. On a per pupil basis, however, our spending is somewhat below the na- tional average. In 2003—04, according to the National Education Association (NBA, 2005), California’s per pupil spending was $7,584 per pupil, approximately 92% of the national average of $8,248. That year our spending was the 29‘h highest in the nation. Although this is important information and places our state’s spending in context, perhaps more important is what we purchase with those dollars. What Do We Buy? The single largest expenditure for any school system is for personnel, and the largest single expen- diture for personnel is for teachers. In fact, according to the NBA (2005), the average salary of California teachers in 2003—04 was $56,444, the second highest in the nation. Even when adjusted for California’s rela— tively high cost of living, California‘s teacher salaries still ranked near the top of the states. This fact signifies recognition among Californians of the importance of teachers to educational achievement. However, be— cause overall per pupil spending is relatively low, Cali- fornia also has the third highest pupil—teacher ratio in the nation. Only Utah and Arizona had higher pupil teacher ratios in 2003—4. Another way to look at what we buy in Califor— nia is to compare staffing patterns to those in other states. Table 2 displays the number of staff per 1,000 pupils by staffing category and compares it to aver— age staffing patterns for the entire country. The table clearly shows that for all staff categories, California‘s school children receive substantially less support than children in most other states. There are only 90.0 total staff per 1,000 students in California compared to the United States average of 123.0 staff per 1,000 students. The number of teachers per 1,000 students at 48.3 is only 77% of the national average of 63.1, and our ratio of counselors and librarians per 1,000 students is a fraction of the national average. Regardless of the staffing category considered, California’s children CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 have access to fewer adults each day at school than do children in almost every other state in the nation. How Hard Do We Try? The third way to compare California school spending to the rest of the United States is to look at the level of effort in terms of public expenditures we focus on K—12 education. California’s personal income per capita (a measure of our overall wealth) was $32,845 in 2002, some 6.6% above the national average. That suggests we have the capacity to spend more than average for public services in our state. And in fact we do. According to the NEA’s analysis of Census data and data from the Bureau of Eco- nomic Analysis, California’s state and local govern— ment expenditures per $1,000 of personal income in 2001—02 (the most recent year for which data is available) amounted to $205, or $10 more per 1,000 of personal income than the United States average of $195. Looked at another way, California spent $6,732 per capita for state and local governments in 2001—02 compared to the national average of $6,010—some 12% more than the national average. But, despite the fact that we make a greater than average effort, a higher proportion of California resi— dents are school—age children—27% compared to the national average of 25.3% (EdSource, 2005)—making our effort for K—12 education slightly below the na— tional average at $40 per $1,000 of personal income compared to $41 nationally.‘ All of this is compounded by the characteristics of our school children. We have 8% more pupils per capita to educate, which, when combined with higher salaries and generally greater student needs (48% of California’s school children qualify for free and re- duced price lunches and 25% are English Language Learners), requires that each dollar stretch further. THE HISTORY OF SCHOOL FINANCE: FOCUS ON CALIFORNIA National Trends The history of school finance in the United States during the 20‘h century can be thought of as having three distinct foci. The first is equity, the second pro- ductivity, and the third and most recent, adequacy. The focus is on how these factors have played out in California. FUNDlNG CALIFORNlA’S SCHOOLS, PART l: PAST, PRESENT, . AND-FUTURE? TABLE 2: Staff per 1,000 Pupils: 2003—04 _ *- , Avg. flank Total Staff 123.0 90.9 48 Total District Staff 5.9 5.2 31 Officials and Admin. 1.3 0.4 48 School Staff 89.9 68.4 50 76% Certified School Staff 69.7 51.7 49 74% Principals and APs 3.4 2.1 50 62% Teachers 63.1 48.3 49 77% Guidance Counselors 2.1 1.1 50 52% Librarians 1.1 0.2 51 18% Source: EdSource (2005) Equity From the 1950’s onward, equity became a focus and goal of school finance. Equity required designing state funding systems that mitigated the impact of differential property wealth per pupil across school districts. Designing school finance mechanisms that provided state aid in inverse relationship to the prop— erty wealth of school districts helped level the play- ing field, and enabled property poor districts to have more money than would otherwise be available. Productivity In the 19903 considerable emphasis was placed on understanding the relationship between money and student performance. Unfortunately, economists and statisticians have not been able to consistently identify the nature of that relationship and quantify it so that policymakers can appropriate funds in ways that will ensure improved student learning. The rea— sons for this are as complex as the equations used to estimate the relationship. But essentially they include a lack of clarity about the goals of education, and in- sufficient precision in the tools used to gather data on school finance. While today most would agree that the goals of school are to improve student performance, measur— ing that solely through standardized tests is contro- versial. Today’s tests don’t always do a good job of measuring student reasoning and problem solving PACE V1 17 FIGURE 1: Public K-12 Expenditures per $1,000 of Personal Income $45 CA I US Avg. Source: National Education Association (1996—97 est.) Latest data available 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01 2001-02 FIGURE 2: California Expenditures per K-12 Pupil (ADA) ’IOOOO US Avg. M 9000 ‘T— 9229 9125 8925 _...._. CA 8740 8000 ' 8212 Source: US Census . 7845 7867 Bureau, 2006 7000 7280 7555 7237 V 7 0" 6892 6000 5000 4000 I I I I I I I I I I I 03" 9)" ,0?) 93° 9“ 93’ .03” 9° 9" 9” 6’ 9” FIGURE 3: Public K-12 Revenue from Local, State and Federal Sources California New York Florida Texas Source: National Education Associa- . , tion, Rankings of the ||IIn0Is States 0 ’IO 20 30 40 50 60 7O K-12 revenue (in billions of dollars). 2005-2004 18 CRUCIALISSUESIN CALIFORNIA EDUCATION 2006 skills, and the multiple-choice nature of most tests makes it difficult to asses how well children can com— municate. Attempts to quantify these more complex schooling outcomes have not been very successful. Moreover, measures of self esteem and good citizenship (also potentially important outcomes of schooling) are harder to measure at the individual student level. Moreover, 28 states (California included) only collect finance data at the school district level. While the other 21 collect school level finance data, I have argued elsewhere that until we are able to sort out expenditures on an individual student basis, it is un— likely that we will be able to measure the impact of additional resources on student performance (Picus and Robillard, 2000). Even then, the ability to make accurate estimates of the effect of money on perfor— mance may be limited by the fact that we generally spend more money on those children with the great— est educational needs. Careful controls for previous ability and for the characteristics of individual chil— dren will be needed to understand the productivity issue. However, with the more recent emergence of adequacy as a school finance goal, alternative ap- proaches to determining how much money is needed have been developed. Adequacy Another school finance strategy emerged in the 19905. School finance adequacy became the most effective approach for challenging state school fund— ing systems following the Kentucky Supreme Court’s ruling in 1989 that the Kentucky funding system (and the entire education system) was unconstitutional."1 The Kentucky court ruled that all children should be able to meet certain minimum standards, and that resources were inadequate to ensure that was possible. In response, the State Legislature appropriated an ad— ditional one billion dollars a year for education and established one of the nation’s most extensive testing systems. Widely studied, results suggest that the work in Kentucky has led to improved student performance in the last decade. The adequacy movement asks a simple question: How much money is needed to ensure that all chil- dren—or almost all children—can meet a state’s per— formance standards? The problem with this approach lies in determining what that amount of money is. Today there are four approaches for estimating school finance adequacy. They are: FUNDTNG CALIFORNIA’S SCHOOLS, PART l: PAST, PRESENT, AND FUTURE? Successful districts: This approach finds school districts that currently meet state standards and uses their costs as an estimate of adequacy. Cost Functions: Using advanced statistical tech- niques, analysts estimate the resources required for students to reach a given performance level on a stan— dardized test, controlling for student characteristics such as family income and home language. Professional Judgement: Panels of educators are brought together to describe the resources they would need in a school to have some assurance that all children could meet the state’s performance stan— dards. Once specified, the costs of these resources are estimated to arrive at an estimate of the costs of adequacy. Evidence Based: This approach relies on current educational research on what works in schools to estimate the resources needed to reach state perfor— mance standards and then estimates the costs of those resources. A fifth approach is currently being tried in California. As part of an extensive school finance re- search project across the state, economists are using what might best be called a constrained optimiza— tion model to estimate adequacy. This approach asks school officials to design schools that they think will enable students to meet California student perfor— mance standards within varying fiscal constraints, and then asks them to assess the probability that students will attain the standards under that funding model. Although the results of this study were not complete as this was written, a pilot study found that educational professionals organized schools differ— ently as the per pupil funding levels increased. Not surprisingly, the studies also found that the estimated probability of students meeting the standards also increased with funding (Rose, Sonstiele, and Richard— son, 2004). Adequacy studies are used to estimate how much is needed to provide the children of a state with an adequate education. Studies of this type have been conducted in 30 states, and are summarized in the January 2005 Quality Counts issue of Education Week (Education Week, 2005). In every instance, the studies have found that current funding levels are inadequate to enable all children to meet the state’s educational standards. While all of these studies offer different methods for estimating how much money is needed, none PACE V] 19 There are three major even t5 that Shape the appear perfect and all are subject current structure of to criticism (see for example, Hanushek, 2005). This suggests California schoolfinance, that such studies may not pro- the Serrano decision, vide definitive answers to how much is needed, and the high PTOPOSithH 13, and cost estimates found in many Proposition 98. states may not lead to improved student performance unless 20 they are combined with a clearer focus on how the money should be used. Adequacy has been used as the basis for legal challenges to the school funding system in many states, and in all instances has been successful in get- ting the courts to rule that current funding levels are inadequate. California has its own adequacy lawsuit, Williams v. California. The suit in California was unique in that it seemed to focus mostly on the lack of decent school facilities for many school children, and sought more state oversight into the management of school districts.3 The Williams suit was settled in 2004, with an agreement by the state to spend something on the order of one billion dollars to improve school facili— ties in the districts with the most severe facility prob— lems, and to provide additional funding in some set— tings. The funding for facilities is not “new” money.4 While this seems a great deal of money, it compares poorly to New York where adequacy studies have rec— ommended spending increases of six to nine billion dollars. These increases are recommended for a state with half as many children as California, and one that currently spends nearly 50% more per pupil. Adequa- cy studies in other states have recommended funding increases from 10% to over 35%, suggesting that the one billion dollars in additional funds that Williams will provide for California schools (less than 2% of the more than 60 billion in the Governor’s 2006—07 budget) may not be the final answer to the question of adequacy in California. California’s School Funding Story As the discussion above implies, there is sub— stantial evidence that current funding levels for California’s K—12 education system are inadequate. This section provides a brief history of California school finance, and outlines why it has become so complex and unwieldy. It also provides a sense of why changing the system has been—and likely will con— tinue to be—so hard to do. There are three major events that shape the cur— rent structure of California school finance, the Serrano decision, Proposition 13, and Proposition 98. The op- tions facing state policymakers today are limited sub- stantially by these past events. How each affects today’s school finance environment is described below. The Serrano Decision Prior to the early 19705, California relied on a foundation approach to school finance. In general this funding mechanism guarantees each school district a fixed amount of revenue per pupil in exchange for levying a certain tax rate. Districts that can’t ’raise that amount through local taxes receive state aid to make up the difference. While this is a sound approach to school finance (see Odden and Picus, 2004), if the foundation level does not keep up with the financial needs of schools, local districts are forced to rely on their own property tax base to raise the balance. In California, there were (and continue to be) dramatic differences in the property wealth per pupil across school districts, leading to considerable variation in the ability of school districts to raise additional rev— enues. This led to substantial differences in per pupil spending across the state. The SerranoS lawsuit, filed in 1968 and litigated into the 19703, was the first step in transforming California’s school finance structure. Serrano required that all wealth-related spending differences between school districts be eliminated, or reduced to no more than $100 per pupil.“ Today, approximately 97% of all California public school children reside in school dis— tricts that fall within this narrow spending band when the size and type of district are considered.” However, in the three decades since this system was put in place, a growing proportion of state funding for education has been provided through categorical programs which are outside of the Serrano requirement to reduce wealth—related spending differences.8 Today, something on the order of one third of state revenues for schools is distributed through these categorical programs (nearly $12 billion in the 2006—07 budget request). There is evidence that this approach has led to a different, but equally detrimental, set of spending differences. Sonstelie, Brunner, and Ardon (2000) showed that this funding system has resulted CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 FIGURE 4: Revenue Sources K4 2, All Budget 2005—06* Local debt service and misc. (8%) Federal Funds (12%) Lottery fund (2%) General Fund (58%) Local property taxes (20%) “ Source: California Department of Finance * Includes funds for California Department of Education state op- erations, state special schools, state school facilities bond repay- ments, contributions to the State Teachers Retirement System, State Library, and Commission on Teacher Credentialing. in substantial inequities in the level of resources avail- able to children across school districts. Betts, Ruben, and Danenberg (2000) further show there are consid- erable variations in the resources (e.g., teachers, in— structional materials, etc.) available to children across the state. The result today is a confusing system where there is often little relationship between identified student needs and the targeting of revenues. Categorical grants are not a bad thing. In fact the largest categorical programs are for special educa— tion and class size reduction. Many others are used to ensure that funds are directed to students with special needs so that they receive the required services. For instance, Rose et al. (2003) have shown that in many instances these funds increase spending in the schools with higher percentages of high poverty children. However, California has developed a system of more than 100 categorical programs, many of them small, and highly focused on narrow student populations— or often on specific sets of school districts with little regard to student needs. In many instances, districts qualify for funding simply by Virtue of having re— ceived funds from a particular program in the past. In addition, some programs don’t give districts or schools a revenue projection until well into the school year, and then require the funds to be expended in that same year. This forces districts to make poorly thought out and poorly planned expenditures in order not to lose the money. While there have been a number of efforts to sim- plify the categorical programs, most of the so-called block grants that have been established have come with their own sets of complex rules and regulations. For this reason, they have had little impact on the general view in schools that the rules are burdensome and overly complex. This high level of state control over funding is largely the result of Proposition 13. Proposition 13 Passed in 1978 by an overwhelming majority of voters, Proposition 13 dramatically changed the fiscal relationship between schools and the state. By limiting property taxes to 1% of assessed value, and by defining what assessed value is and how much it can grow, Proposition 13 not only reduced the state’s FIGURE 5: School Expenditure K—1 2, by Type* Percent U‘l Ci Middle Elem. Equipment Services and operating exp. Materials and suppies Classified compensation Other certificated compensation I D i i Teacher compensation High Source: Rose, H., Sonstelie, J., & Reinhard, R. (2006) *Based on schools in study for 2001-02 school year PACE $1 21 The heart of the problem finance today is that we f0?” California 5Ch001 revenue capacity for govern- mental services, it also placed control of all property taxes don’t have a clear picture 0f squarely with the Legislature. how much money we need. As a result, local school districts today have no revenue raising 22 authority to speak of, and any additional revenue generated by property taxes becomes an offset to the state’s general funding of schools, known as revenue limit funding? The funding system that was developed in re- sponse to Proposition 13 remains the basis for school funding today. California uses what is generally thought of as a foundation program. Each district has a revenue limit—a historically—based figure that has been adjusted upward over time based on cost of living adjustments and the number of students in a district. The revenue limit is funded by a combination of property taxes and state funds, which make up the difference between the revenue limit guarantee and the property tax collections. Categorical programs, as described above, make up the balance of a school district’s state resources. Proposition 98 In 1988, the voters passed an initiative designed to guarantee that school funding keep pace with changes in enrollment and the cost of living (per capita income). Proposition 98 also includes a pro— vision (test) that schools receive a minimum share of the state’s budget each year. Known by its ballot designation of Proposition 98, that measure contin— ues to impact California fiscal policy today. While it guarantees approximately 40% of the state’s general fund budget to K—14 schools (K—12 and community colleges), it has a number of complex requirements that impact how new state resources can be spent, and establishes floors in funding that can make it harder to made reductions in education spending when rev— enues are low. It was even suspended once in 2003—04 due to the poor fiscal condition of the state. Although Proposition 98 provides a theoretical floor for educa— tion spending, it also has limited legislative flexibility in budget decisions, and as a result has in some in— stances also served as a ceiling for education spending. Today the focus of California’s school reform ef— forts is standards-based accountability. California has been a leader in establishing performance standards for students. The use of the Academic Performance Index (API)—-—and its wide publication across the state—has spurred schools to improve student learning as deter- mined by the measures that constitute the API (test scores, attendance, and dropout rates). This provides a clear description of what students need to know and be able to do. What is missing is a clear system of funding that is tied to strategies that will lead to that improved level of performance. In fact, when all of the many fac- tors that impact funding for schools in California today are combined, there appears to be much confusion, and a general agreement that to meet our current per— formance standards, schools need more money. In the section that follows, an attempt is made to discuss the major policy issues facing the state today, how they are impacted by the past, and the adequacy options available today. CALIFORNIA SCHOOL FINANCE TODAY Resources for Schools The heart of the problem for California school finance today is that we don’t have a clear picture of how much money we need. While the governor and other policymakers are understandably reluctant to determine what that amount is——since it is likely to be more than we currently spend—absent a target to strive for, the level of school funding will continue to be determined through political compromises emerg— ing from an increasingly unstable and under—funded state revenue system. This problem can only be solved by determining how much is needed to adequately fund California’s schools. TABLE 3: State Owes Proposition 98 Settle—up for Past Years (Dollars in Millions) 1995—96 $0.00 1996-97 $0.00 2002-03 $491 .60 2003-04 $617.60 2004—05 $1,620.90 2005-06 $1,298.90 2006—07 $0.00 $4,029.00 Source: Legislative Analyst’s Office, September 2006 (nominal dollars) CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 FUNDlNG CALlFORNlA’S SCHOOLS, PART l: PAST, PRESENT, AND FUTURE? TABLE 4: Year-to-Year Changes in Proposition 98 Funding (Dollars in Millions) ¥ _ . Changes ' Budget Act 2005-06 Budget Act 2006-07 Amount ‘ Percent K4 2 $44,644 $49,1 13 $4,469 10.0% Community colleges $5,217 $5,894 $677 13.0% Other $107 $114 $7 6.5% Totals $49,968 $55,121 $5.1 53 1 0.3% General Fund $36,591 $41,295 $4,704 12.9% Local property tax $13,377 $13,827 $450 3.4% K—1 2 attendance $6,031,404 $5,957,368 -$74,036 -1.2% K—1 2 per pupil spending $7,402 $8,244 $842 1 1.4% Source: Legislative Analyst’s Office, September 2006 (nominal dollars) TABLE 5: Increase Funding for Declining Enrollment Adjustment MAY ESTIMATE ' ENACTED ‘ 2001 -02 2002-03 2003-04 2004-05 ‘ 2 00 5 -0 6 . ‘ 2 00 6— 07 . Statewide growth rate 2.10% 1.70% 0.90% 0.40% —0.22% —0.27% Districts receiving adjustments 327 375 412 438 * * "Phantom" ADA** funded*** 16,000 20,000 29,000 49,000 69,000 73,000 Costs of declining enrollment $74 $93 $137 $242 $357 $398 (in millions) * Unknown ** Average daily attendance *** Difference between level at which districts are funded (based on prior year) and the numbers of students they are actually serving. Source: Legislative Analyst's Office, September 2006 (nominal dollars) POLICY IMPLICATIONS No Child Left Behind recognized that states needed a dozen or more years to establish systems to enable all children to meet perfor— mance standards. Therefore it is unrealistic to expect that California can find all the money it needs in one year. But it is essential to know what we need and have a plan to get there. Even with that, it is likely that state revenues will fluctuate over time. Manage Resource Flow Over Time Any funding plan needs to accommodate the long—term growth in revenue and provide for dips and spikes over time. Figure 6 provides a simplified “cash flow” analysis of how the state could manage its resources toward the goal of adequate funding. In years when rev— enues exceed needs, it would be wise to bank funds for future years when state revenues are below identified needs. Similar to many states’ “rainy day funds,” this concept probably re— quires substantially more spending restraint on the part of the Legislature or return of tax receipts to taxpayers in good years than they have in the past. It also requires the education community to allow the funds to be banked, rather than diverted to uses not part of a long-term strategic plan. It requires PACE E2 23 24 FIGURE 6: Managing the Flow of Resources OverTime ’1000 800 600 400 200 ,4" a; ,//.r I .r g r44 fl / "// I .1 1/ >/’ a dramatically different approach to the al- location and use of tax revenues than we have seen in California in recent history. But where does that money come from? Cali— fornia is one of the wealthiest states in the nation. Our average per capita income exceeds the national average by nearly 7% and ranks 12th among the states. Yet our spending on education is similar to states near the bottom of the income rankings. The question is, who should pay for our schools? There are two options, neither popular, but both with the potential to resolve this funding issue. Consider Income Tax Increase On Two Highest Tax Brackets It is estimated that these citizens (with Cali— fornia taxable incomes exceeding $200,000 for individual returns and over $400,000 for joint returns) will receive nearly $12 billion in tax breaks from the federal tax cuts. Tapping these tax benefits could go a long way toward funding schools without increasing the total tax payments of the state’s wealthiest citizens. Review and Modify Proposition 13 This approach is even less popular than the previous suggestion. Proposition 13 has hamstrung state and local government for years by reducing the revenue potential of all governments. Moreover, it has created substantial inequities, not only between hom— eowners in similar homes, but across classes of property with more of the tax burden being shifted to residential property. (More on Proposition 13 in Chapter 3). While some argue that Proposition 13 is needed to protect business, in reality it only protects existing businesses, and makes it hard for new firms to build the production facilities they need and to compete with ex- isting firms. Under those circumstances, not only does governmental revenue suffer, but the lack of competition hurts all consumers. It is possible to find a fair and reasonable way to increase the revenue potential of property taxes while ensuring that state residents don’t get taxed out of their homes or businesses, provided we are willing to make the sacrifices Governor Schwarzenegger has called for. Some have argued that the property tax rolls should be split, allowing business property to be assessed at full market value (and possibly subject to a higher tax rate), while residential property continues to be treated as it is cur- rently and only reassessed when it is sold. Since business property changes hands less frequently than residential property, theoreti- cally this would stem the shift of property taxes to residential property. Another option might be to assess all property at market value, establish tax rates on that new assess— ment that collect the same revenue (rates should be lower given the assessment growth caps in Proposition 13) and allow schools to seek approval for higher property taxes up to the existing 1% rate cap. Other potential options exist as well, although the political climate in our state suggests such changes are unlikely in the near future. Other Options There are also additional options for ensur— ing students receive an adequate education. Analyses of adequacy often point out that children have needs that go beyond the public school system’s capabilities and responsibili- ties. Access to good prenatal care, high quality CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 medical and dental facilities, and good pre— schools can lead to improved school perfor- mance for many children, particularly those from low—income homes. A recent analysis of the public, and of nonprofit services available for children and their families in the area sur— rounding the University of Southern Califor- nia, revealed that there is as much as $12,500 per child available. These resources provide assistance to families, food stamps, recreation opportunities, early childhood education programs, and cultural activities, to name just a few. Combined with a similar amount through the public schools in that area,10 there is nearly $25,000 available per child to provide educational and other social services. But often the problem lies in making sure that these resources reach their intended target, and that the agencies responsible for provid— ing those services coordinate their efforts. While schools have typically been organized from the bottom up and most other social services from the top down, California’s highly controlled school funding system may be an ideal place to begin breaking down the barriers between agencies toward the creation of coordinated educational and social services for all children. This could be accomplished at little or no additional cost. CONCLUSION California faces a number of major challenges in pro— viding adequate funding for its public schools. The student population includes large numbers of chil— dren who require additional services to succeed and a school revenue structure that is entirely dependent on the condition of the state budget. These condi— tions have produced a high demand for educational resources coupled with a fluctuating capacity to meet that need. While the state has established high stan- dards for what our children need to know and be able to do, it has not yet developed a clear strategy to fund the educational services needed to meet those needs. FUNDING CALIFORNIA’S SCHOOLS, ‘ PART I: PAST, PRESENT, AND FUTURE? ENDNOTES National Education Association Rankings and Statistics 2004-05, with 2001—02 data which is latest availale. 2 Rose v. Councilfor Better Education, 790 SW. 2d 186 (Kent, 1989). California is also the only state where the defendants (the State) countersued, claiming that the problem was not inadequate funding, but rather mismanagement by local district officials who had access to the same level of fund- ing as other, more successful school districts. Williams facility funding comes from the Proposition 98 Revision Account, money that has already been appropri— ated for Prop 98 but not spent. 5 Serrano v. Priest, 5 Cal. 3d 584, 487 P.2d 1241, 96 Cal. Rptr. 601 (1971) (Serrano I); Serrano v. Priest, 18 Cal. 3d 728, 557 P.2d 929, 135 Cal. Rptr. 345, (Serrano II) (1976), reh. denied, Ian. 27, 1977; as modified Feb. 1, 1977 cert. denied, 432 US. 907 (1977). This figure has been adjusted for inflation and today is over $300 per pupil. California school districts are organized into elementary (K—8), high school (9—12) and unified (K—12 districts, and further divided into small (less than 101 students for elementary, less than 301 for high school, and less than 1,501 students for unified districts) and large districts. The assessment of Serrano compliance is determined in these six groups of districts. Wealth—related in terms of assessed property values with- in bounds of a school district but not defined by income. Such “high—wealth” districts included San Francisco, Oakland, Richmond, Berkeley, Los Angeles where the majority of children in poverty lived. There are a few districts that are an exception to this. Known as basic aid districts, they are able to generate more property tax revenues than their revenue limit calls for—and they are allowed to keep the difference. Voter approved property taxes for local general obligation bonds are also an exception. While an important issue, it has relatively little bearing on the general discussion that is the focus of this chapter. 10 When the total “all funds” budget of the LAUSD is di— vided by its average daily attendance (ADA), the resulting calculation approaches $12,500 per student. REFERENCES Betts, 1., Rueben, K., 8: Danenberg, A. (2000). Equal resourc- es, equal outcomes? The distribution ofschool resources and student achievement in California. San Francisco, CA: Public Policy Institute of California. EdSource. (2005). How California ranks: A nationalperspec- rive. Palo Alto, CA: EdSource. PACE % 25 26 Education Week. (2005). Quality counts: No small change, targeting money toward student performance. Bethesda, MD: Editorial Projects in Education. Fortune 500. (2006). Fortune 500: Full list, Retrieved May 11, 2006 from http://money.cnn.com/magazines/fortune/ fortuneSOO/full_list Hanushek, EA. (2005). The alchemy of ‘costing out’ an adequate education. Retrieved May 11, 2006, from http:// edpro.stanford.edu/hanushek/admin/pages/files/uploads/ alchemy.revision.pdf Legislative Analyst. (2006). 2006-07 Budget analysis, educa- tion chapter. Sacramento, CA: Legislative Analyst’s Office. (2006, March 10). Retrieved May 11, 2006, from http:// www.1ao.ca.gov/analysis.aspx?year=2006andchap=0and toc=4 National Center for Education Statistics. (2006). Projections of education statistics to 2014. Washington, DC: National Center for Education Statistics, Table 33. Retrieved May 11, 2006, from http://nces.ed.gov/programs/projections/ TableDisplay.asp?id=tab_33.aspandazelemsec NBA (2005). Rankings and estimates. Washington DC: National Education Association (June). Odden, A., & Picus, L. (2004). Schoolfinance: A policy per- spective (3rd ed.). New York: McGraw Hill. Picus, L., & Robillard, E. (2000). The collection and use of student level data: Implications for school finance re- search. Educational Considerations. 28 (1), 26—31. Rose, H., Sonstelie, 1., Reinhard, R., & Heng, S. (2003). High expectations, modest means: The challenge facing California’s public schools. San Francisco, CA: Public Policy Institute of California. Rose, H., Sonstiele, 1., & Richardson, P. (2004). School bud- gets and student achievement in California: The principal’s perspective. San Francisco, CA: Public Policy Institute of California. Sonstelie, I., Brunner, E. & Ardon, K. (2000). For better orfor worse? Schoolfinance reform in California. San Francisco, CA: Public Policy Institute of California. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Jay G. Chambers American Institutes foi" "' ROVIDING AN ADEQUATE SYSTEM OF SCHOOL FUNDING, ONE THAT AEFORDS ALL STUDENTS THE OPPORTUNITY TO SUCCEED ACADEMICALLY, HAS BECOME A NATIONAL MOVEMENT. WHILE THE ANTECEDENTS OE THE ADEOUACY MOVEMENT HAVE BEEN AROUND FOR A LONG TIME, IT IS IRONIC THAT IT HAS BEEN GIVEN A SIGNIFICANT BOOST SINCE 2000 BY THE ADVENT OF THE NO CHILD LEFT BEHIND (NCLB) l,.A\’\C This law builds on the standards-based reform movement of the 19905 and requires that by 2014 all students reach a certain level of proficiency as defined by their state education system. This requirement has helped define the obligation of the states to meet the needs of the diversity of students served within each state public education system, and hence, push educational adequacy to the forefront of debate in school finance reform. The purpose of this paper is to outline the im— plications and challenges put forth by the adequacy movement for the state of California. The chapter is organized into three sections. Section 1 includes a brief historical perspective on school funding for the nation as a whole and in the context of California. The second section describes the concept of cost— based funding and how it addresses several common problems in traditional school funding systems. The final section describes the strengths and challenges of various approaches to addressing adequacy and what implications these may have for developing a new K— 12 school funding system in California. HISTORICAL TRENDS IN LOCAL, STATE AND FEDERAL SHARES OE SCHOOL FUNDING During the first half of the 20‘h century, schools were for the most part funded by local taxes. In 1920 well over 80% of K—12 revenues were derived from local sources, while less than 20% came from the state and a negligible amount from the federal government (see Figure 1). Up through the 19705, the dramatic declin— ing share of aggregate educational revenue made up by local sources was largely offset by a roughly equal increase in state revenue. In addition, the federal role expanded, growing to a maximum of approximately resources simply required ensuring equal access to educational dollars across local jurisdictions. In turn, state governments faced two questions: I How much can we afford to spend? I How do we best distribute these dollars among local communities? Indeed, it was through Serrano v. Priest, the property tax limits imposed by Proposition 13, and the subsequent centralization of school finance at the state level that California was pushed into asking and answering the second of these questions. Unfor— tunately, an unintended consequence of the Court’s I How much can we afford to spend? I How do we best distribute these dollars among local communities? I How much differently do we need to treat children with varied needs? I How differently do we need to treat the various types of communities in which these children are served? ruling was that funding was to be equalized down where, rather than providing general aid to poorer districts at the level enjoyed by more affluent districts, spending limits were imposed at a level between the two. That is, while poorer dis— tricts had their levels of general aid increased, many less needy districts witnessed dramatic decreases in the amount of general aid they received. As Iohn Mockler, former execu— tive director of the California State Board of Education, ex— plained, “This was like having a crippled arm and a good arm and making things equitable by crippling both arms.”1 To compensate for the decline in general aid to dis- tricts, the following years saw California’s school finance for— mula becoming progressively more complex, with a growing amount of categorical dollars being used to address special need populations. As men— tioned in the previous chapter, the state of California cur— rently has approximately 100 categorical programs that make up its school funding formula.2 However, it should be noted that there have been ef- forts made to consolidate overlapping categorical aid programs with duplicative goals.3 As the use of categorical aid is targeted for spe- cific purposes or student populations, the shift from general to categorical funds meant that local educa- tion authorities lost significant control over how they could allocate their resources. This inflexibility in the use of categorical funding has been cited as a major factor in districts’ inability to target resources to their best use. To make matters worse, recent literature (Timar, 2004) suggests that the relationship between student need and the distribution of funds from cat- egorical aid programs is weak at best and regressive at worst (i.e., in some cases the least needy students are receiving the highest level of these funds). Shifting from Equity to Adequacy Many would agree that the current school finance debate in many states could be described as tumultu— ous. State governments are being sued for failing to provide an “adequate” education, suits which often results in a lengthy appeals process. While definitions of what constitutes an “adequate” education are at least as numerous as the court cases being tried, for the pur— poses of this chapter we define this concept as follows: A public education may be perceived as “adequate” when the goods, services, and delivery systems under which they are pro— vided are capable of offering all students, regardless of need or circumstance, the op— portunity to achieve a prespecified level of achievement that has been agreed upon as desirable by consensus of a significant con— stituency of stakeholders. But exactly how did the debate shift from a (hori— zontal) equity to adequacy? As it became commonly recognized that children and the communities in which they were served varied widely with respect to their need and wealth, the school finance literature was pushed toward addressing the concept of vertical equity.4 Under legislation that embraced this concept, state funding sys- tems were developed that treated children and taxpayers of varying need and wealth in systematically different ways. For taxpayers, it meant recognizing differences in ability to pay for public education and the purchasing 3O CRUCIALISSUESIN CALIFORNIA EDUCATION 2006 power of the educational dollar in their geographic area. For children it meant recognizing differential student needs of economically and disadvantaged children, and eventually students with disabilities and those for whom English was not their first language. It is within this context of vertical equity that we find the movement towards the concept of adequacy where the following questions are asked: I How much differently do we need to treat chil— dren with varied educational needs? I How differently do we need to treat the various types of communities in which these children are served? The first question begs for some basis on which to define how children differ with respect to their needs. For example, we could simply define this in terms of different types and intensities of services, but then we need to consider to what end (i.e., what outcomes) we direct those services. That is, what is the educational goal we wish to achieve? The second question leads one to think about purchasing power of the educational dollar and requires some decision on the part of the state to make appropriate adjust— ments to equalize access to resources and services. As the conceptual shift from equity over the past 25 years manifested itself into legal action taken by individuals, school districts, and other interest groups, we have seen a majority of states forced to address the issue of K—12 school funding from an adequacy per— spective. Most state constitutions contain clauses that describe their obligations to the citizens and children of the state with respect to education. Increasingly, as states have come under litigious fire for failing to deliver on these constitutional promises, their obliga— tions have become interpreted as a charge to provide all students an adequate education, with “adequate” being defined as achieving (or an opportunity to achieve) at some prespecified level of outcome(s). Thus far, 45 states have been taken to court for failing to fulfill their commitments.5 Since 1989, 21 states have been told by their own Supreme Courts that their school funding systems are inadequate to provide what their constitutions promise to the chil— dren of their state" Feeling the pressure, several other states have undertaken to address the questions of adequacy on their own. ‘ FUNDING CALIFORNIA’S SCHOOLS, PART II: RESOURCE ADEQUACY AND EFFICIENCY Federal policy also contributed to the push for educational adequacy through the encouragement of better performance of state education systems. The No Child Left Behind (NCLB) law requires greater accountability and even encourages improvements in school funding systems through the newly enacted Education Finance Incentive Grants under Title I of NCLB. It should be noted that California had the components of a strong accountability system in place prior to NCLB, unlike many other states. Through this legislation, the federal government strayed from its traditional “hands off” position with regard to the states’ unfettered provision of education. These forces have also helped to fuel a new push for state public school systems .. . The [Williams] case . called into question the to prov1de an adequate educa- tion for all children, irrespective of their need and circumstance.7 unbalanced relationship between the state’s School systems must pay at— tention not just to the average allocation 0f 7'650117’C65 student, but to students with . . . . . and the standards—based disabllities and non—natlve Eng— lish speakers, as well as those accountability system, , _ whose economic circumstances have not afforded them the same educational opportunities as their wealthier counterparts. Clearly, school finance systems designed to promote horizontal equity have not been able to achieve this objective. In sum, the state’s responsibility as educational provider has shifted from ensuring that resource dollars are distributed equitably among chil— dren, to allocating resources so that there is equity in terms of the opportunity each child has to reach some designated level of educational outcomes. In the context of California, the adequacy move- ment is still well in its infancy. Thus far, the largest step towards adequacy has been the Williams v. California school finance case, where the state was sued for not providing the bare essential resources necessary for children to learn. As Oakes et a1. (2003) explain:8 The problems highlighted in Williams v. Cali— fornia—students forced to learn in schools with too few qualified teachers, insufficient textbooks and instructional materials, and overcrowded, unsafe buildings—are simply not addressed in today’s “results-based” accountability systems. PACE ’2 31 In short, the Williams plaintiffs seek to hold state officials accountable for the condi- tions under which California students are expected to learn, particularly for the often shocking conditions in schools most likely to be attended by poor children and children of color. What is perhaps most interesting was that the case called into question the unbalanced relationship between the state’s allocation of resources and the standards-based accountability system, where schools are expected to perform at a minimally acceptable level. The state was not being held accountable to pro— vide the level of resources that is arguably necessary to achieve at this level. In August 2004 a settlement between the plaintiff and defendant was reached and later in March 2005, was approved by the court. The settlement included the establishment of new, more balanced accountability standards to ensure: Each and every student has a right to “suf— ficient textbooks,” a school in “good repair,” and a qualified teacher. Districts must per— form self—evaluations to ensure compliance with the textbook and facilities standards, and then share the results of their evalua— tions and teacher misassignment and va— cancy reviews with the public in their annual School Accountability Report Cards.9 The resources provided per the settlement included: I $138 million for new instructional materials for students attending the lowest performing schools (in 2004—05 budget) I $25 million for a one—time comprehensive assess- ment of school facilities (in 2004—05 budget) I $100 million per year, beginning in the 2005— 2006 school year, up to a maximum of $800 mil— lion, to be put into a newly established School Facilities Emergency Repair Account to be used by districts to fix facilities at low—performing schools. As mentioned in Chapter 2, no new funds were used to settle this suit. Three Fundamental Adequacy Issues With the new era of adequacy in place, state poli- cymakers have realized the need to approach the issue of school funding in an entirely different way. Invari- ably, they have been forced to consider the following three fundamental issues associated with educational adequacy: I Goals — What are the goals of the school system? More specifically, how do states op- erationalize the constitutional obligations they have set out for the children they serve? I Cost — What are the types, quantities, and subsequent costs of various resources re~ quired by school systems to achieve those goals? Or more simply, what does it cost to provide an adequate education? I Distribution of Resources — What kind of funding formula must be devised to ensure all children have access to adequate edu- cational opportunities, as dictated by the system’s goals? Often the clauses found in state constitutions provide vague descriptions as to the exact definition of an adequate education. For instance, the term “thorough and efficient” was introduced in 1857 in Minnesota, in 1872 in West Virginia, and in 1947 in New Jersey to describe the state obligations in provid— ing public education to its citizenry.10 The interpreta— tion of such constitutional obligations will drive the determination of what resources are necessary to provide educational adequacy. Therefore, before one can begin to address the issue of cost, it is essential to have a well—defined objective of the public education system that includes measurable outcomes that must be attained. While formal concrete statements stemming from constitutional adequacy clauses are not read— ily available across all states, two factors have helped push the establishment of these goals statements. First, the recent wave of court cases has revealed the need for states to operationalize their goals for public education. For example, much of what the DeRolph case in Ohio was about was defining the concept of “thorough and efficient” and how it relates to CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 adequacy in school funding.11 The recent court case in New York sought to define the concept of a “sound basic education” and to determine the cost of pro- viding this opportunity to all children in New York public schools.12 Second, the federal NCLB act has forced all states to design standards-based accountability systems that provide a foundation for appropriate achieve~ ment goals. However, it must be noted that academic achievement is not the sole basis upon which the per— formance of school systems should be evaluated. Once an educational goal is established, the next step is to determine the cost of an adequate education and develop a system of resource distribution that ensures all districts can provide adequate educational services to their children. Concurrent with the con- ceptual shift away from equity was a new strand of research and literature that focused on how best to address these issues of cost and distribution. The fol- lowing section documents the emergence of this liter— ature surrounding cost—based funding, and describes the various models that have been used to determine the cost of achieving educational adequacy. COST-BASED FUNDING During the 19703 there were a number of researchers who began to develop approaches to addressing geo— graphic differences in the cost of education.‘3 These studies focused, for the most part, on factors affecting the supply and costs of comparable teachers across local jurisdictions. The purpose of these studies was to consider ways of adjusting state school funding for variations in the purchasing power of the educational dollar across local educational agencies (LEAs). How— ever, these variations only scratch the surface of cost differences. In addition to these cost adjustments, there was a growing recognition that additional state aid would be required to meet the needs of diverse student populations. The growth of federal support for special student populations (e.g., through Title I of the El— ementary Secondary Education Act in the 19605 and the Education for All Handicapped Children’s Act of the 19705) increased state recognition that additional state funding would be necessary. FUNDING CALIFORNIA’S SCHOOLS, PART II: RESOURCE ADEQUACY AND EFFICIENCY Elements of Cost-Based Funding In order to break down the concept of cost-based funding more precisely, let us consider three factors that underlie the variations in the costs of educational services. These can be summarized in three words: price, need, and scale. School funding systems need to take these three factors into account when designing the foundation for state aid distribution. Price refers to the cost of acquiring comparable inputs in different geographic locations. State aid needs to account for variations in the factors that affect the cost of comparable inputs. For the most part, this results from differences in the supply of school personnel in different regions of a state. Variations in the cost of living and the attractiveness of differ- ent regions of the state affect the willingness of indi- viduals to live and work in certain areas. This is just as . . . there was a growing recognition that additional state aid would be required to meet the needs of diverse student populations. true for auto mechanics, engineers, store clerks, and lawyers as it is for school teachers. Even if you were to set a single salary schedule throughout the state, the reality of cost differences would exist because people with higher/lower qualifications may cluster to differ- ent regional labor markets. Ask yourself whether you would be indifferent about working in San Francisco, Stockton, Fresno, and Lassen County if you were of— fered the same compensation in all of these locations. Need refers to the differential resources required to address the unique socioeconomic or educational circumstances of the students being served (e.g., stu- dents living in poverty, with disabilities, and coming from homes where English is not the first language). This component seems to be the most intuitive to the public and policymakers. Students from economically disadvantaged homes simply don’t have the same access to educational experiences outside of school as students from wealthier families, not to mention health care, nutrition, and other environmental fac- tors that affect a student’s ability to take advantage of schooling.14 English learners will most certainly need specialized educational approaches and the support of professional staff with bilingual capabilities. (See Chapter 7). Finally, students with disabilities require more intensive support and services to accomplish PACE 91’ 33 the same goals as students without disabilities and to compensate for specific mental, physical, and emo— tional challenges. Scale refers to the differences in the cost of of- fering and providing the necessary outcomes in smaller and more remote com- munities, as opposed to larger HOW does one go about more urbanized ones. Schools in remote, rural regions of the designing an adequate state will not be able to take schoolfunding system? For advantage of the economies of large-scale operation that are decades State policymakers available to the more urbanized have sought advice on how school systems.15 Geographic barriers (mountains and rivers) t0 do just that. While most and long distances between recognize that price, need, communities that make home— to—school transportation infea- and scale are COSt factors, sible, prevent consolidation and how should these be taken require additional educational resources. The big issue here into account . . . ? is to determine the extent to which scale is a matter of choice 34 versus a necessity. The notion is we want to compensate only for those differences in cost that are, for all intents and purposes, outside local control. Any differences in spending across local educa— tion authorities (LEAs) that remain after accounting for price, need, and scale can be attributed to choices to make larger or smaller investments in education by state and local policymakers. What Does Research Tells Us About Educational Adequacy? How does one go about designing an adequate school funding system? For decades state policymak— ers have sought advice on how to do just that. While most recognize that price, need, and scale are cost factors, how should these be taken into account when developing a school funding formula? To address this issue, it is necessary to understand the relationship between school spending and student outcomes, and to note adjustments in state aid that reflect these cost factors. Educational researchers have struggled for decades trying to ascertain what works in education. The literature has sought to determine the impact on student outcomes of differences in: class and school size (Finn and Achilles, 1999; Grissmer, 1999), teacher training and experience (e.g., Dar- ling-Hammond, 2004), access to preschool programs (Barnett, 1995, 1996, 2000), parental involvement (Henderson, 1998), availability and utilization of technology (Wenglinksy, 1998), and access to various pupil support personnel (Wehlage and Stone, 1996). While researchers have struggled to determine unambiguously what works in education, they have also recognized that some of the difficulty in doing so results from the complexity of separating out the multitude of family background factors and com- munity characteristics from the effects of schools on student outcomes. The US. Department of Education (DOE) has established the Institute for Education Sciences (IES) to promote randomized field trials to encourage more scientific studies of what works in education. Moreover, the DOE has funded research to conduct comprehensive reviews of the literature to summarize what works.16 The research to date is far from conclusive and so cannot yet be used to develop school funding formu— las. However, it is possible to draw some conclusions about cost differences from the available data, which can at least serve as a starting point for deliberations. Furthermore, much of the literature focuses on a narrow set of schooling outcomes. Many of the stud- ies focus on student achievement test scores, which in many instances are based on multiple—choice tests of children’s learning in math and language arts. In some instances, these tests are extended to the natu— ral sciences, history, and other social sciences. What these tests often don’t measure are students’ creativ- ity, the ability to work in groups to solve problems, their knowledge of music and the arts, their physical condition, and their ability to learn in the future. As Richard Rothstein comments, “To judge schools exclusively by their test results is, therefore, to miss much of what matters in education.” ‘7 He goes on to suggest the following: Obsessed with test scores, educators have de— voted almost no effort to identifying or mea— suring noncognitive outcomes...While public opinion surveys consistently include higher test scores as a school goal, they are not the only goal and apparently not the most crucial. In one such survey, conducted in CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 1994, over two—thirds of Americans said that teaching values was a more important role of public schools than teaching academic subjects; the top rated “value” was teaching students to solve problems without violence. In another recent survey, Americans were asked to rank school purposes; respondents gave their highest rank to preparing respon- sible citizens; helping students to become economically self—sufficient was second. In yet another pool conducted recently, 80% of respondents said that whether graduates practice good citizenship was a “very impor— tant” way to measure schools, but only 50% said test scores were a “very important” way. Thomas Iefferson called upon public schools to provide not only “academic proficiency” but also to “understand duties to neighbors and country, and to observe with intelligence and faithfulness all social relations18 Studies by a research team led by Christopher Iencks (1979) found that adult economic success was affected more by years of schooling completed, and noncognitive traits such as leadership ability, than by test scores. Indeed, the Iencks study showed that even children who had completed a greater number of years of school (either high school or college) were more successful in economic life than those with less, even if they had comparable test scores. If these noncognitive outcomes were linked to cognitive (academic) outcomes and the test scores that presumably measure them, then measuring aca- demic outcomes alone might be sufficient to assess schools. But the very emphasis on the more easily measured academic outcomes resulting from NCLB and the standards-based reform movement may have reduced the correlation between these two outcomes. Some studies have extended the list of outcomes to include attendance rates, dropout rates, and gradu— ation rates.19 Others have focused on student partici— pation rates in athletic or extra curricular programs, and yet other studies have measured student satisfac— tion, self esteem, and sense of self—efficacy.20 Indeed, colleges use dimensions other than SAT scores in the admissions process, and there has been significant research on some of the problems with SAT scores as opposed to high school grades and FUNDING CALIFORNIA’S SCHOOLS, PART II RESOURCE ADEQUACY AND EFFICIENCY measures of participation in extracurricular activi- ties, student government, autobiographic narratives, and statements of ambitions as measures of potential success in college. Some colleges have even considered dropping SAT scores altogether in favor of alternative measures in the admissions process.21 Thus, student test scores are only a part of what schools are about or should be about. However, some psychologists have even questioned whether the test scores themselves are a good measure of what stu— dents learn in school. As Professor James Popham of UCLA points out: Traditionally constructed standardized achievement tests measure a bit of what’s taught in school. But by and large they mea— sure what children bring to school, not what they learn there. To me, one of the most frightening things about the preoccupation of raising test scores is the message it sends to children about what’s important in school. Rather than trying to make the classroom a learning environment where exciting new things are required, the classroom becomes a drill fac— tory, where relentless pressure and practice on test items may raise test scores—but may end up having children hate school.22 Education research does provide some insights into factors that contribute to success in schools. For instance, at least one reasonably well—designed ex— perimental study has suggested the positive benefits of smaller class sizes (see Krueger, 1997; Krueger and Whitmore, 2001). In addition, there is some evidence that there are benefits to smaller schools, but this is not universal (Barker and Gump, 1964; Lee and Smith, 1997; Raywid, 1997/1998; and, Ready et al., 2004). There is also quite a bit of evidence that dem- onstrates the value of preschool programs, such as the recent report by Karoly and Bigelow (2005). Teachers do make a difference, but researchers are not entirely sure what it is about them that makes a difference (Hanushek, 2002). There has been some evidence that additional investment in special education can have benefits in terms of student achievement for students with PACE $2 35 n e legislative representatives, representatives of the gov- ernor’s office, and members of the business commu— nity. Their charge was to review the work and provide feedback directly to the professional judgment panels (PIPs). The PIPs are told up front that they will be asked to defend their work and respond to feedback from the stakeholder panel. By insisting that the PIPs must defend and justify their recommendations and specifications, the stakeholder panel provides some- what of a check on the system. The stakeholder panel requires the educators to become accountable within the professional judgment process. The most recent application of the professional judgment approach (Rose et al., 2004 and Sonstelie, 2005) does not use panels of educators per se, but rather samples of individual educators who are asked to complete a spreadsheet listing various school re- sources and their prices.28 This approach takes the educators through exercises that specify the compo- sition of pupil needs, the budget of the school, the types of staffing and resource 0n the impact ofvarioas performance (e. g, teacher There is a great deal of inputs with which they have to work, and then asks them to specify the quantities of resourc— COHfliCtiflg evidence es within the budget constraint to maximize student outcomes (e.g., the Academic Performance resources on student Index in California). By varying the budget and composition of students, they can predict the certification, experience, relationship between predicted and educational preparation ), as well as the impact of service outcome levels by the educators, the costs, and the pupil needs. This new approach was devised to address two weak— nesses of the original profes— delivery models. sional judgment approach. First, previous studies simply asked 38 professionals what they thought schools needed, regardless of prices or budget. Second, previous studies “tend to involve groups of professionals working together to come up with the resources an ideal school would contain” and “these group sessions obscure the varia— tion among individuals in the resources they think are needed to reach some target.” 29 Successful Schools Model Almost a decade after the early studies of school finance adequacy by Chambers and Parrish (1982, 1984), Augenblick (1997) developed the successful schools approach to estimate the costs of an “ad- equate” education. Based on this model, Augenblick has worked with states to identify schools and dis- tricts that, based on certain specific criteria, were found to be highly successful in regard to student per- formance. The cost of adequacy was then estimated based on the expenditures of these high performing districts and schools. Unfortunately the manner in which the Suc- cessful Schools Model is generally applied does not account for variation in the demographic and need characteristics of students. Consequently, the high performing districts and schools were generally those with access to more resources. Also, this approach is not well suited to provide estimates of what is re— quired to serve different populations of students with respect to poverty, English language proficiency, and disability characteristics. In effect, the model provided no inherent mechanism for determining the marginal or incremental costs of meeting the needs of special populations of students. Nevertheless, this approach did provide an initial benchmark for cost estimation. A more recent modification of this model in— volves the use of a beating-the-odds (BTO) analysis. This approach uses econometric techniques to iden— tify schools that are performing better than expected given the populations of students they serve. Thus, it uses statistical methods to compare schools serv— ing comparable populations and then identifies the best of each group of schools. One can then obtain detailed information about the resources and in— structional practices of the best schools in order to estimate costs. There are a number of limitations to the BTO model. First, it is difficult to estimate the cost of achieving very high standards because few, if any, of the schools serving the most challenging student pop— ulations are achieving at these levels. Second, in those rare instances where such schools are achieving at high levels, there are often idiosyncratic resources that cannot be easily replicated, or the student populations attending the school are systematically different than those at similar schools. In the first case, the school could have an exceptional principal with extraordinary CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 leadership skills and charisma that motivates the staff. These kinds of factors are inherently difficult to mea- sure and likely have any easily estimable cost implica— tions. In the second case, one often finds that these schools are magnet programs or the students are self— selected through some mechanism that distinguishes them from students at the other schools. Third, the BTO model is limited by the use of readily available measures of school performance, which are for the most part test scores. Evidence-Based Approach In recent years, an approach called the Evi— dence-Based (EB) Approach has been put forth uses models drawn from the literature in education research on effective schooling practices to estimate the cost of achieving adequacy in school funding. While there is much to be said for the concept of an evidence-based approach to cost estimation, educa- tional research is unclear regarding effectiveness. As suggested above, there is a great deal of conflicting evidence on the impact of various resources on stu— dent performance (e.g., teacher certification, experi— ence, and educational preparation), as well as the impact of service delivery models (e.g., small versus large class sizes or schools). For this reason, this ap— proach is very sensitive to the selection of literature chosen to serve as a foundation for cost analysis. This model tends to promote a “one size fits all” ap— proach and is not very sensitive to the differential objectives (outcome levels) that might be specified by state law or policymakers. Nevertheless, the notion of using what evidence is available as a starting point for further deliberations and cost analysis has merit. But this means that for the approach to be functional, it must be combined with some sort of professional judgment application that offers local educators an opportunity to make ap- propriate adjustments to meet their needs. Moreover, because the literature on what works is so thin, the EB approach does not naturally lead to any direct esti— mates of the marginal costs of serving various special needs student populations in the way the professional judgment approach permits. Cost Function Model Some economists have contributed to the school finance adequacy literature by using econometric FUNDING CALIFORNIA’S SCHOOLS, PART II: RESOURCE ADEQUACY AND EFFICIENCY models to estimate what they refer to as cost func— tions.}0 These econometric models are simply adapta- tions of the types of cost models that economists have used for decades to analyze data from manufacturing and other production processes in private industry. The models relate total production costs to outcomes, input prices (e.g., teacher salaries), and demographic characteristics that reflect pupil needs and assume optimization behavior on the part of the key decision makers (e.g., output maximization or cost minimiza- tion decision—making processes). This approach pro- vides a systematic way to explore the relationship be— tween educational costs and student outcomes, while controlling for input prices and other factors affecting educational production. That being said, there are several drawbacks to the cost function model. It relies on a limited set of outcome measures (e.g., usually one index of aver— age district achievement). Furthermore, the approach does not provide a very transparent perspective on how the inputs are organized and utilized, and is gen- erally difficult to explain to non-researchers such as legislators and policymakers. Also, cost functions pro- vide little information to help policymakers under- stand the factors underlying the differences in costs. The cost function model requires fairly large samples of districts to estimate stable statistics, and this requirement makes it difficult to apply in many states with limited numbers of districts. Often the models become complex, requiring the estimation of numerous statistics to determine the cost impact of attaining various outcome measures (e.g., test scores, attendance rates, graduation rates). In turn, the data requirements associated with running these models are quite large. For instance, to be complete the models should ideally control for the various categories of students served (e.g., percent in poverty, English learners, or with various disabilities), school input prices (e.g., for teachers, school administra— tors, aides), measures of district size (e.g., enroll- ment and enrollment squared), and environmental factors (e.g., density of student populations). Com- plex functional forms can also substantially increase the number of independent variables used and reduce the ability to interpret the statistics directly. To estimate these parameters generally requires a state with a few hundred school districts and data spanning a period of three to five years. PACE m 39 40 An additional criticism of the econometric ap- proach is that it often must resort to cost projec- tions based on outcomes and student demographic combinations that are well beyond the sample from which the model was estimated econometrically. For example, statistical models can be used to project the expected cost of producing a given high level of achievement for a student population facing extreme economic disadvantage or other characteristic needs (i.e., incidence of special education or English learner students). In reality, however, schools with such ex— treme need characteristics that also exhibit such high levels of achievement might not exist! In this case, much faith must be put into the ability of the statisti— cal model being used to reliably predict what it will take to produce achievement well beyond what is currently observed in schools. This point is especially salient, as the outcome objectives established by the states for adequacy studies are generally well beyond the observed performance of most (or all) of the school districts within a state. Given that schools are a public enterprise operating outside the context of a market to motivate behavior, some economists might question the applicability of the econometric approach. The notion here is that there is no motivation for school district decision makers to maximize outcomes for any given budget constraint or minimize the budget necessary to achieve a certain out— come. The advent of NCLB may have altered this moti- vation somewhat through the public embarrassment of school decision makers for failing schools. While using this model can provide some in— teresting comparisons, the complex assumptions on which it is based, the limited outcomes usually con— sidered, and the significant data requirements neces— sary to implement the model make it, for the most part, an impractical approach to conducting work on adequacy. Where this approach would seem most useful is in combination with other approaches to check for consistency of cost estimates. Multiple Model Studies and Hybrid Approaches More recent adequacy studies by the various research teams have presented adequacy estimates stemming from multiple methodologies or have used combinations of the above approaches to come to an estimate. For example, Augenblick and Meyers (2001, 2002, 2003) use both the Successful Schools and Professional Judgment approaches to obtain separate estimates of the cost of an adequate education in the states of Kansas, Maryland, and Missouri. Odden et al. (2003b) have begun studies using the matrix of resources based on the Evidence-Based approach, but then have appointed a professional judgment to review the matrix and make adjustments. Chambers et al. (2004, 2006) report on one of the most comprehensive studies of adequacy conducted in New York State by a team led by the American In- stitutes for Research (AIR) and Management Analysis and Planning (MAP). This model included an elabo- rate public engagement component implemented by the Campaign for Fiscal Equity in New York, in which input from a wide variety of constituencies was col— lected about the desired goals and types of resources necessary to achieve the desired results. In addition to public engagement, the New York study combined elements of Evidence—Based and beating—the—odds approaches. Beating-the-odds analyses were conducted for two purposes: 1) to identify schools from which some of the participants on the professional judgments were selected, and 2) to develop comparative information on the staff- ing and resource allocation patterns that existed in high— versus average—performing schools. The results of the analyses were to be presented to a review team selected from the professional judgment panels. The Evidenced-Based approach was utilized by developing a brief, well—balanced review of the exist— ing education literature on “what works.” This review identified educational practices that had been shown by some of the literature to have positive effects on student outcomes, but also presented (when available) alterna— tive views on these effects. The purpose was to provide something for the members of the professional judg- ment panels to consider prior to their deliberations. The New York study included eight independent panels from four different categories of school dis— tricts to ensure that the analysis did not rely on the specifications of any one panel.“1 A second stage panel was also selected to review the synthesis of the analy— sis conducted by the research team, to help interpret the data, and to make any necessary revisions. Concluding Remarks on Cost-Based Funding The value of these approaches lies not so much in offering a final answer to the question of adequacy. In CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 fact, most research teams, regardless of what approach they use, don’t present just one estimate to their cli- ents. They often run simulations that offer a range of numbers based on varying assumptions. In the New York study, for example, different cost estimates were presented at the different stages of the process. These included estimates based on initial specifications of the panels and those after they had made some modi- fications, as well as differential assumptions about how central administration, maintenance and opera— tions, and other cost components were accounted for in the analysis. These types of analyses may provide a starting point for discussion and debate over how best to fund schools. They present a systematic set of numbers on which policymakers can base state aid allocations. They offer a rational basis for resources to be distributed in such a way as to promote vertical equity, rather than just allocating whatever dollars may be available after money is first allocated to other social services. No matter which approach is selected, the numbers that are derived are based on some analysis of existing spending or resource allocation patterns, and/or determinations of what resources may be necessary to achieve the de— sired results for the public education system. Some states have conducted multiple adequacy studies, in some cases by completely different research teams, and there is something to be said for that ap- proach. It provides some options for consideration by the legislature and allows comparisons across models to see how close the numbers are. But it is critical that these studies proceed to some degree outside the reach of the political process. While the use of public engagement and stakeholder panels seek widespread input into the analysis, the results need to be insulated from any attempt to limit the findings to what is po- litically acceptable. There is also a need to be honest about setting the goals of public education and in designing a structure to meet those goals, even if the cost estimate derived may not be politically feasible at any given point in time. We can at least establish an objective system— atic picture of what it will take to achieve the desired outcomes, and then work toward making the funding available. In establishing this dialogue with the public, it is necessary to be honest with the citizenry of the state about what it will take to achieve adequacy as defined by a consensus—based educational goal. FUNDING CALIFORNIA’S SCHOOLS, PART ll RESOURCE ADEQUACY AND EFFICIENCY POLICY mucmom, j; Implementation Phase-In It is very important to implement the kinds of potentially major changes in school fund- ing that may be recommended as part of an adequacy study in a very deliberate way. It is extremely difficult to accomplish this kind of change in a short period of time. School districts and the state must go through a sig- nificant planning process with the end goal in mind. New teachers may have to be trained by the higher education system. Incentives must be provided to help reduce turnover among existing staff, and higher salaries may be re- quired to induce additional people into the market for these jobs. New classroom space may be required to meet programmatic needs, and this will in- volve spending additional dollars on facilities. For example, if the adequacy estimates are based on smaller class sizes or smaller pupil staffing ratios, additional facilities will be necessary to house all of the additional class— rooms and offices for professional staff. The state needs to work in concert with local school district decision makers to make this process as smooth as possible. For instance, if money is to be added to the system in order to increase staffing, it must be done in a deliberate way to ensure that certification standards for new personnel can be satisfied as the instructional programs are expanded. It has been suggested that such a lack of foresight and planning severely limited the effectiveness of California’s recent, well—in— tended, large—scale reform, statewide class size reduction. Although resources were made available for hiring additional instructional staff, school districts were unable to recruit and employ enough qualified teachers in the short period of time they were given. As a result, the level of teacher quality suffered, PACE a 41 CRUCIAL ISSUES Given the costs, it would seem that there may well be a role for some federal assistance in conducting this kind of research. The No Child Left Behind (NCLB) Act requires all states to establish standards and accountabil- ity systems against which to measure the per- formance of public schools. Moreover, NCLB has established a timeline which requires that all students achieve proficiency on the estab- lished state standards by the 2013—14 school year. Given that states, school districts, and schools across the nation are expected to meet the requirements of NCLB in the near future, it is essential that they have available to them estimates of the cost of reaching these goals. This begs the question of what level of invest- ments of federal, state, and local resources will be needed to provide a fair opportunity for all public school students to meet these proficiency goals by 2013—14. It is evident that good answers to these ques- tions require a range of studies in addition to the costing—out research studies that many states are funding. While it is tempting to consider ways in which states could share re— sources and conduct joint studies, this would require establishing some areas of common ground where cross state studies could serve the purposes of many states. In fact, many of the studies that have already been carried out on adequacy related to school governance, analyses of geographic cost adjustments, analyses of the cost of school facilities, and home~to~school transportation could be car— ried out across multiple states. A possible avenue for the proliferation of this work nationwide would be for the federal government to establish regional research centers for the investigation of adequacy and equity in school funding. Such centers could have at least two benefits. First, the centers could be designed to encourage consortiums of researchers currently working in this arena to collaborate in their efforts across states to improve the efficiency with which these studies could be carried out. And second, the centers would facilitate an increase in the comprehensiveness of the types of studies supporting school finance reform across the nation. IN CALIFORNIA EDUCATION 2006 . FUNDING CALIFORNIA’S SCHOOLS, PART II: t, RESOURCEADEQUACY AND EFFlClENCY APPENDIX A: Litigations Challenging Constitutionality of K—1 2 Funding in the 50 States In Process (23) No Current Lawsuit (22) Never Had a Lawsuit (5) Alaska Nebraska Alabama Ohio Delaware Arizona New Hampshire California Oklahoma Hawaii Arkansas New Jersey Florida Oregon Mississippi Colorado New Mexico Illinois Pennsylvania Nevada Connecticut New York Indiana Rhode Island Utah Georgia North Carolina Iowa South Dakota Idaho North Dakota Louisiana Vermont Kansas South Carolina Maine Virginia Kentucky Tennessee Massachusetts Washington Maryland Texas Michigan West Virginia Missouri Wyoming Minnesota Wisconsin Montana Source: Molly A. Hunter, Litigations Challenging Constitutionality of K—1 2 Funding in the 50 States (Campaign for Fiscal Equity, Inc. 2005) APPENDIX B: School Funding ”Adequacy” Case Decisions Since 1989 Plaintiff Victory (21) suit/13$???“ Pending (9) Alaska New Hampshire Alabama Alaska* Arizona New Jersey Florida Arizona* Arkansas New Mexico Illinois Colorado Georgia" New York Massachusetts (2005)’r Connecticut Idaho* North Carolina Pennsylvania Kentucky* Kansas Ohio Rhode Island Louisiana Kentucky South Carolina* Texas (2005)1' Missouri* Maryland Texas (I 989)t Nebraska Massachusetts (l993)t Vermont North Dakota Missouri Wyoming Montana This chart reflects the highest state court’s decisions or un—appealed lower court decision. *Alaska and Arizona plaintiffs won capital funding cases; operational funding cases are pending. Kentucky and Missouri plaintiffs won new funding systems years ago; new suits allege that those systems have changed and become unconstitutional. Georgia’s high court denied state’s motion to dismiss and remanded for trial. tMassachusetts and Texas have each had two high court adequacy cases. Source: Molly A. Hunter, Litigations Challenging Constitutionality of K—1 2 Funding in the 50 States (Campaign for Fiscal Equity, Inc. 2005) PACE P2 45 46 E N U1 C ENDNOTES Quote taken from February 2005 article, “California’s school funding went from first to among the worst. What went wrong?” California Educator 9(5): 6—21. Actually, the study by Timar (2004) states that in 2000—01 there were 124 categorical programs, which provided $113 billion of educational funding. However, the author is careful to point out, “The exact number of categorical programs is debatable due to differences in the definition of categorical programs.” Funding for programs outside K-12 like preschool, childcare, and Adult Education are categoricals. An example of this is the passing of Assembly Bill 825 in 2004, which consolidated over 20 existing categorical education programs into six categorical block grants. For an example, see the study by Clune (1994). Appendix A.1 includes a list of states involved in litiga- tion over the constitutionality of their school finance systems. Appendix A.2 contains a list of school funding adequacy case decisions since 1989. It must be noted that, while the incentive structure in- stituted by NCLB’s accountability system focuses on en— suring that all students achieve at state-mandated levels of proficiency, the act offers little assistance to states in the way of technical direction or additional funding to achieve these achievement goals. See http://www.edsource.org/pdf/Oakes_et_al_ Accountabilitypdf, pages 1,3. Quote taken from Allen (2005), page 10. See http://home.att.net/~pcbworks/Tand1.html for a dis— cussion. See McKinley (2005a, b) for a full discussion. See Supreme Court Of The State Of New York, County Of New York, Campaign For Fiscal Equity, Inc., et al., Plaintiffs, —against- THE STATE OF NEW YORK, et al., Defendants. Index No. 11 1070/93,m Justice DeGrasse, J. For instance, see Brazer (1974), Grubb and Hyman (1975), and Chambers (1978, 1980, 1981a, b). For a more detailed examination of the impact of socio- economic status on schools, see Rothstein (2004). Clearly, the argument can be made that there may also be diseconomies to scale (i.e., where per~pupil costs increase with size) for extremely large urban districts. However, we assume here that the number of districts currently op— erating on this scale make these cases rather exceptional. Two IFS-sponsored projects currently being conducted by the American Institutes for Research (AIR) are the What Works Clearinghouse (http://www.w—w-c.org/) and the Comprehensive School Reform Quality Center (http://www.csrq.org), which address the effectiveness of 2 2 2 2 2 2 \l on O N L» .1; U1 0‘ various types of educational interventions and compre- hensive school reforms, respectively. Rothstein (2004), page 86. ibid, 95—96. For instance, Sander and Krautmann (1995) and Evans and Schwab (1995) consider the effect of attending Catholic school on the probability of graduation and total educational attainment, while Neal (1997) also con— siders its effect on wages. For an example, see Schunk and Hanson (1987), for anal- ysis of the effect of peer modeling on student self-efficacy and achievement. See Richard C. Atkinson (2002), The Changing World of College Admissions Tests, http://www.ucop.edu/pres/ comments/wacubo.htm and California Colleges May Drop SAT Use, by Shankar Mukherji, http://www-tech. mit.edu/V121/N5.5/55$AT.5.5n.html. Quotes from James Popham of UCLA in an interview on the public broadcasting system radio show, “Frontline:” http://www.pbs.org/wgbh/pages/frontline/shows/schools/ interviews/popham.html. Quote taken from report press release, retrievable at http://www.rand.org/news/press.05/01.03.html. Though these statistics draw from different data sources, the point here is to show changes over time. See page 53 of RAND report California’s K—12 Schools: How Are They Doing (2005) which shows in figure 3.14 that the percentage of personal income in 2000 was approximate— ly 3.6%, while just under 4.5% in 1970. As noted in the report (note 32) The discrepancy between the estimate of California,s school spending as a percentage of personal income presented in Figure 3.13 and the comparable ratio presented in Figure 3.14 reflects differences in the data sources we used. The data for Figure 3.13 were taken from the US Census Bureau, which uses its own an— nual survey of government finances; the data for Figure 3.14 were taken from NCES, which uses the common core survey data. These two surveys define state public elementary and secondary school spending differently, the Census Bureau using state public school “finance amounts” and NCES using state public school “spend— ing.” The difference lies in the fact that actual spending for a year might be lower than what is initially financed. We cannot use the Census Bureau data for Figure 3.14 because they do not go back to 1970. However, because the Census data are commonly used for ranking states in terms of the ratio of public school per-pupil spending to personal income, we elected to use them for Figure 3.13. One of the most recent applications of the Professional Judgment Model (PJM) may be found in Chambers et al. (2004). For other recent applications of PJMs, see Guthrie et al. 1997; Augenblick and Meyers (2001 ); and Verstegen (2004). The report is available for download at http://www. cfequity.org/FINALCOSTINGOUT3—30—04.pdf. RUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 27 Another perspective suggests that the cost estimates may be driven largely by the level of the educational objective provided to the professional judgment panels to guide their deliberations. 38 This approach is also being applied currently to measur- ing the cost of an adequate education in California by PPIC, and the study is being led by Ion Sonstelie of UC Santa Barbara and PPIC. 19 The concern stemming from the first criticism is that the traditional model does not account for relative prices and budgets in the decision-making process. While there is some merit to this critique, the traditional approach that has been applied, for example, by Chambers et al. (2006) and others, actually does nothing more than turn the optimization problem faced by the professional judgment panels on its head. While the Rose/Sonstelie PIM chooses to maximize the outcome achieved for a given budget, the traditional application of the PIM used by Chambers et al (2006) sets the outcome level objective and asks what minimum cost combination of resources would be required to attain that goal. While explicit consideration of prices could be a useful addition to the deliberations, most of the educators who participate in these panels have a pretty good idea of the relative costs of teachers and aides, and are likely to take this into account in their deliberations. The second criticism can actually be viewed as an advan- tage of the traditional approach. Organizing a diverse group from a wide variety of educators (i.e., superinten— dents, teachers, business officers, etc.) offers diverse per- spectives on the design of the school prototypes and the associated decisions about resource allocation in relation to pupil needs. The notion here is that there is value in the synergies among the educators that might be created within the group exercises, especially given the complex- ity of the process in making judgments about what re- sources might be needed to achieve the goals and objec— tives. Moreover, the panels of educators are being asked to go beyond mere resource specifications to think about the design of the instructional programs. While there is significant uncertainty in the processes of resource al— locations that may lead to the outcomes, there is some advantage in drawing on the experience and expertise of a variety of education professionals working together. See the following for examples of cost function analyses of adequacy: Duncombe and Yinger (2000); Duncombe (2002); and Reschovsky and Imazeki (2001). w District categories were defined largely by urbanicity of district (New York City, the other four large cities, and suburban and rural districts). See, for example, evaluations of California’s class size reduction program by Iepsen and Rivkin (2002) and Wexler et al. (1998). The latter is downloadable at: http:// www.wested.org/ policy/ pubs/ full_text/ class_size/ sect 1 . htm. 33 The William and Flora Hewlett Foundation is leading the effort along with the Bill and Melinda Gates Foundation, the Stuart Foundation, and the Irvine Foundation. FUNDING CALIFORNIA’S SCHOOLS, PART ll RESOURCE ADEQUACY AND EFFICIENCY REFERENCES Allen, B. (2005). The Williams v. California settlement: The first year of implementation. ACLU Foundation of Southern California and Public Advocates, Inc. Retrieved May 12, 2006 from http://www.decentschools.org/settle— ment/WilliamsReportWebZOOS.pdf Andrews, M., Duncombe,W. &Yinger, I. 2002. Revisiting economies of size in American education: Are we any closer to a consensus? Economics of Education Review, 21, 245-262. Augenblick, I., Myers, 1., Silverstein, I. & Barkis, A. (2002). Calculation of the cost of a suitable education in Kansas in 2000-2001 using two dijj‘erent analytic approaches. Denver, CO: Augenblick and Myers, Inc. Prepared for the Kansas Legislative Coordinating Council. Augenblick, I. & Myers, I. (2001). Calculation of the cost of an adequate education in Maryland in 2000—2001 using two diflerent analytical approaches. Denver, CO: Augenblick and Myers, Inc. Prepared for Maryland Commission on Education Finance, Equity, and Excellence. Augenblick, I. & Myers, I. (2003). Calculation of the cost of an adequate education in Missouri using the professional judgment and the successful school districts approaches. Denver, CO: Augenblick and Myers, Inc. Prepared for Missouri Education Coalition for Adequacy. Barker, R. & Gump, P. (1964). Big school, small school: High school size and student behavior. Stanford, CA: Stanford University Press. Barnett, W.S. (1995). Long-term effects of early childhood programs on cognitive and school outcomes. The Future of Children: Long—Term Outcomes ofEarly Childhood Programs, 5, 25—50. Barnett, W.S. (1996). Lives in the balance: Age-27 benefit— cost analysis of the High/Scope Perry preschool program. Yspilanti, MI: High/Scope Press. Barnett, W.S. (2000). Economics of early childhood inter— vention. In Shonkoff, I. 8r Meisels, S. (Eds.), Handbook ofearly childhood intervention (2nd ed.). Cambridge: Cambridge University Press. Birman, B.F., Desimone, L., Porter, A. 8: Garet, MS (2000). Designing professional development that works. Educational Leadership, 57, 28-33. Brazer, H. E. (1974). Adjusting for difference among school districts in the costs of educational inputs: A feasibility report. In Selected Papers in School Finance. Washington, DC: Department of Health, Education, and Welfare, Office of Education. Betts, I. (1996). Is there a link between school inputs and earnings? fresh scrutiny of an old literature.” In Burtless, G. (Ed.), Does money matter? The effect ofschool resources on student achievement and adult success (pp. 141-91). Washington, DC: Brookings Institution. Card, D. 8( Krueger, A. B. (1992). Does school quality mat- ter? Returns to education and the characteristics of public PACE n 47 50 McKinley, S. K. (2005b). Part II. The journey to adequacy: The DeRolph saga.” Journal ofEducation Finance, 30 (4), 321—381. Neal, D. (1997). The effects of Catholic secondary schooling on educational achievement. Journal of Labor Economics, 15, 98-123. Oakes, J., Blasi, B. & Rogers, J. (2003). Accountability for adequate and equitable opportunities to learn. UCLA. In Sirotnik, K. (Ed.), Accountability run amok, New York: Teachers College Press. Odden, A., Fermanich, M. 8K Picus, L. (2003a). The state—of- the-art approach to school finance adequacy in Kentucky. North Hollywood, CA: Lawrence Picus and Associates. Prepared for Kentucky Department of Education. Odden, A., Fermanich, M. 8: Picus, L. (2003b). An evidence- based approach to school finance adequacy in arkansas. North Hollywood, CA: Lawrence Picus and Associates. Final Report Prepared for the Arkansas Joint Committee on Educational Adequacy. Raywid, MA. (1997/ 1998). Synthesis of research: Small schools: A reform that works. Educational Leadership, 55, 34-39. Ready, D., Lee, V. & Welner, K. (2004). Educational equity and school structure: School size, overcrowding, and schools—within-schools. Teachers College Record, 106(10), 198—20 1 . Reschovsky, A. & Imazeki, J. (2001). Achieving educational adequacy through school finance reform. Journal of Education Finance, 26(4), 373-396. Rich, D. (1985). Theforgotten factor in school success: The family. Washington, DC: Home and School Institute. Rose, H., Sonstelie, J., Richardson, P. (2004). School budgets and student achievement in California: The principal’s perspective. San Francisco, CA: Public Policy Institute of California. Rothstein, R. (2004). Class and schools: Using social, econom- ic, and educational reform to close the black-white achieve- ment gap. New York, NY: Teachers College Press. Sander, W. & Krautmann, A.C. (1995). Catholic schools, dropout rates and educational attainment. Economic Inquiry, 33, 217—233. Schunk, D.H. 8r A.R. Hanson. (1985). Peer models: Influence on children’s self—efficacy and achievement be- haviors, Journal ofEducational Psychology, 77, 313—322. Shanahan, T. (1998). On the effectiveness and limitations of tutoring in reading. Review of Research in Education, 23, 217-234. Slavin, R., Karwiet, N. & Wasik, B. (1994). Preventing early schoolfailure: Research, policy and practice. Boston, MA: Allyn and Bacon. Sonstellie, J. (2005). Afinance cost studyfor California. Proposal submitted to The William and Flora Hewlett Foundation. Taylor, L., Baker, B. 8( Vedlitz, A. (2005). Measuring educa- tional adequacy in public schools (Working Paper Number 580). College Station,Texas: Texas A & M University, Bush School of Government and Public Service. Timar, T. (2004). Categorical school finance: Who Gains, who Loses? (PACE Working Paper 04-2). Berkeley, CA: Policy Analysis for California Education. Verstegen, D. A. (2004). Calculation of the cost of an ad- equate education in Kentucky: A professional judgment approach. Education Policy Analysis Archives, 12(8), 1-36. Wendling, W. (1979). Cost-of—education indices for New York State school districts. Denver, CO: Education Commission of the States. Wenglinsky, H. (1998). Does it compute? The relationship be- tween educational technology and student achievement in mathematics. Princeton, New Jersey: Educational Testing Service, Policy Information Center, Research Division. Wenglinsky, H. (2000). How teaching matters: bringing the classroom back into discussions of teacher quality. Princeton, New Jersey: Educational Testing Service, Policy Information Center Report. Wehlage, G.G. & Stone, C. R. (1996). School-based student support services: Community and bureaucracy. Journal of Education for Students Placed at Risk, 1, 299-317. Wexler, E., Izu, J., Carlos, L., Fuller, B., Hayward, G. and Kirst, M. (1998). California’s class size reduction: Implications for equity, practice, and implementation. Joint report by Policy Analysis for California Education (PACE) and WestEd. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 CHAPTER '- ~ therine Bitter and Jennifer O’D'ay' can Institutes f . r' t :- CCOUNTABILITY FOR STUDENT OUTCOMES HAS BECOME A CENTRAL FEATURE OF EDUCATIONAL POLICY IN CALIFORNIA. WITH ROOTS IN THE CALIFORNIA ASSESSMENT PROGRAM BEGUN IN THE 19705 AND CURRICULUM PRAMEWORKS DEVELOPED IN THE 19808 AND EARLY 19908, CALIFORNIA’S CURRENT RESULTS—BASED SYSTEM GOT A BOOST PROM NATIONAL INFLUENCES AT THE END OF THE LAST CENTURY. Starting in 1994, accountability based on outcomes became the primary focus when the implementation of the Improving America’s Schools Act (IASA),l coupled with dismal scores on the 1994 National Assessment of Educational Progress (NAEP), served as the impetus to implement a comprehensive and coherent accountability system in California. Development of that system began with the establishment of standards in multiple subject areas and the Standardized Testing and Report— ing system, or STAR. But it was the passage of the Public Schools Accountability Act (PSAA) of 1999 that put a fully developed, results-based system in place. PSAA established 1) a common index called the Academic Performance Index (API), by which to measure and rank the academic performance of schools throughout California; 2) the Immediate Intervention/ Underperforming Schools Program (II/USP), a program that provided assistance and funds, and threatened potential consequences for low-performing schools; and 3) the Governor’s Per— formance Award (GPA) program to reward schools that made significant growth in student outcomes. Later the High Priority Schools Grant Program (HPSGP) was initiated under PSAA to target addi— tional funds and assistance to the lowest performing schools in the state. Implementation of the federal No Child Left Behind Act (NCLB) has further complicated Califor— nia’s accountability landscape. While the federal con— text has been present since the inception of account- ability in California and has served as a strong force in forming the current—day system, only recently has the focus on reconciling the federal system with the state initiatives come to the forefront. More rigorous test— ing requirements and outcome targets within NCLB have resulted in overlapping accountability provisions in California that draw educators’ attention in mul- tiple directions. 51 52 This chapter reviews the evolution of account- ability in California, and outlines several key issues confronting the state as it moves forward. As ac- countability in California encompasses a wide range of initiatives at the federal, state, and local levels, we will constrain our focus primarily to the evolution of PSAA and its current interaction with NCLB. In ad- dition, we will focus exclusively on school and system accountability. California has also developed pro- grams to address individual accountability, for both students and teachers. These include the California High School Exit Exam, required by high school stu— dents to graduate, as well as initiatives to ensure that students are being taught by a high quality teaching force. These aspects of accountability are addressed in other chapters of this book (6 and 7), and should be considered key parts of the full accountability land— scape in California. EVOLUTION OF RESULTS-BASED ACCOUNTABILITY IN CALIFORNIA The current approach to results—based accountability in California has undergone several major phases over the past few decades, originating in the standards— based movement and evolving into an aligned system of standards, assessments, targets for growth, and consequences. In the 19805 and early 19903, Califor— nia played a leading role in developing a systemic and aligned approach to instructional improvement, with many observers identifying 1983 as the beginning of California’s standards-based effort. In that year, which coincided with the national report, A Nation at Risk, the state legislature passed SB 813, reaffirming the need for curriculum frameworks to guide instruction, and setting standards for high school course enroll— ment. The establishment of the California Subject Matter Projects— a network of professional develop— ment institutes to support instruction aligned with the frameworks—followed shortly thereafter. The California Assessment Program (CAP) and later the California Learning Assessment System (CLAS) added aligned assessments as part of the system. PHASE1: Establishing Standards, Assessments, and Public Reporting The first major phase in the development of California’s results-based accountability system es- tablished a foundation of standards, assessments, and public reporting of key data. The intention of the system was straightforward. Statewide content standards would set out expectations for what stu- dents should know and be able to do. Assessments aligned with these standards would allow educators to measure individual students’ learning and to assess the extent to which schools were developing students’ knowledge in key areas. Public reporting of state as— sessment results would then allow educators, parents, and the public to see progress made by individual schools and districts. Standards The move toward California’s current academic standards accelerated in 1994, when several factors converged to create an environment conducive to standards—based reform. In particular, California’s poor results on the 1994 National Assessment of Edu- cational Progress (NAEP) convinced legislators that change was necessary, while the 1994 Elementary and Secondary Education Act reauthorization required that Title I students be taught according to the same state—adopted standards as other students, at least in reading and mathematics. These factors pushed state leaders in 1995 to build on and revise the content of the pre—existing curriculum frameworks in order to create statewide academic standards in the four core subject areas—English/ language arts, mathematics, history/social science, and science. The final adoption of the current academic content standards in these subjects began in 1995 and was completed in the fall Of 1998. Assessments Attempts to institute a statewide assessment system in California date as far back as the 19605. Two critical precursors to the current system were the Cal— ifornia Assessment Program (CAP) and the California Learning Assessment System (CLAS). The CAP, first administered in 1972, provided a “matrix sampled” test that allowed for the assessment of school perfor— mance, but not of individual student performance. Nearly two decades later, in 1991, the Superintendent CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 of Public Instruction introduced the CLAS, designed to align with the curriculum frameworks in place at that time. While this test measured performance of both schools and individual students, it faced opposi- tion on many fronts and was eliminated in 1995. With continued pressure to implement statewide aligned assessments to measure school and student outcomes, the state established the Standardized Test— ing and Reporting (STAR) Program in 1996. Because the state standards were still being developed, the State Board of Education selected the nationally- normed Stanford—9 (SAT—9) as the test that all public school students in grades 2 to 11 would take in the spring of 1998. The plan, however, was to develop and roll in additional assessments aligned with the new academic content standards. The transition to this standards—based assessment system was completed in 2002, with the administration of the California Standards Tests (CSTs) in English/language arts and mathematics. Public Reporting Public reporting of data on school characteristics and performance can serve as a strong motivator for educators and schools to improve student outcomes. In addition, given the large number of public dol— lars allocated to education, the public has a right to receive clear and valid information about their schools. In order to make useful and valid informa» tion publicly available to educators and communities, California has implemented provisions requiring the dissemination of school—level performance data to educators and the public. The passing of Proposition 98 in 1988, which ensured (according to several cri— teria) a minimum amount of funding for elementary and secondary schools, also required that schools compile an annual School Accountability Report Card (SARC) to provide information on various measures, including student achievement, dropout rates, class size, school safety, expenditures, curricular programs and instructional materials, and other items. Federal reauthorization of ESEA in 1994 reaf- firmed the need for the SARC by requiring that assessment results be reported and broken out by demographic groups. The STAR program enabled the state to obtain similar information on student achievement across all schools in California. Informa- tion from the STAR could thus be incorporated into CALIFORNTA’S ACCOUNTABILITY SYSTEM the SARC. With such information available, the state could better identify struggling schools and districts that needed support or incentives to improve.2 PHASE 2: PSAA—Introducing Targets, Assistance, and Consequences The system of statewide standards, STAR assess- ments, and public reporting served as the basis for introducing the Public Schools Accountability Act (PSAA) in 1999. The PSAA legislation moved Califor— nia into the next phase of accountability by adding to this foundation: I a set of targets based on standards and assess- ments, l criteria to identify low-performing and high-per- forming schools, and I a system of assistance and consequences. This legislation moved California to a full results— based accountability system. Targets Based on Standards and Assessments One goal of PSAA and similar accountability sys— tems elsewhere is to focus the attention of educators and the public on student achievement, with the as- sumption that this attention will lead to increased ef— forts to improve that achievement. To that end, PSAA instituted specific school-level targets for student per— formance based on the Academic Performance Index (API). For the first time, schools could learn how they ranked in comparison to all other schools in the state, and in comparison to demographically similar schools. These results were published publicly to raise awareness among communities about their schools’ performance and to create incentives for schools to improve relative to others. The API is a numeric index assigned to each school, ranging from 200 to 1000. Initially based solely on the results of the norm—referenced SAT-9 portion of the STAR program, calculation of the API has now incorporated the California High School Exit Exam (CAHSEE) and the California Standards Tests in English language arts, mathematics, science, and history/social science, and has increased the weight assigned to these standards—based measures. Although the API ceiling is 1000, the Board of Edu- cation set an interim performance target of 800 for PACE TE 53 54 all schools to achieve. This goal has served as the basis for determining individual school API targets on a yearly basis.3 The characteristics of the API targets represent certain choices that California policymakers made, and certain assumptions about the behavior that those choices might engender. Several types of perfor- mance are woven into the API targets": I Performance relative to a set standard—that is, a score of 800 on the API. The fact that the API is an index is itself important, as it incorporates multiple subjects, different measures, and an increased weighting over time of assessments aligned with the California content standards. I Performance relative to past performance—that is, growth in API over time toward the absolute standard of 800. The amount of expected growth is based on the school’s API score for the prior yean I Performance relative to other schools. This is the norm—referenced aspect of the API. Schools are ranked against each other, and against schools with similar characteristics. These rankings are reported as deciles, which indicate the school’s rank in terms of their overall relative perfor~ mance and of their performance relative to simi— lar schools. I Performance disaggregated by “numerically significant subgroups.” For a school to reach its target, it must also show “comparable improve— ment” for all numerically significant ethnic and economically disadvantaged subgroups.S Another PSAA assumption is that the API growth targets will push schools not only to improve the average level of performance every year, but to do so for all their students, especially the lowest—performing students. The API formula rewards growth from the bottom upward more heavily than growth from the middle upward, creating an incentive for schools to provide the most help to pupils with the lowest scores. Identification of Schools for Assistance and Consequences PSAA introduced two programs— one for any school that did meet its targets (the High Perform— ing/Improving Schools Program, also known as the Governor’s Performance Award), and one for low-performing schools that did not meet their API growth targets (11/ USP) -—and established criteria for identifying eligible schools for each program. Under the Governor’s Performance Award pro- gram, schools qualified for a monetary award if they met their growth targets, showed comparable growth among all significant ethnic and economically disad- vantaged subgroups, and satisfied testing participa- tion rate requirements. However, because no funds were appropriated for awards after 2002, GPA identi— fication did not occur after this time. With respect to II/ USP, two aspects of the iden- tification criteria are important to note. First, in contrast to many state accountability systems, partici- pation in California’s program for low—performing schools was voluntary. By applying for II/USP, schools traded off the possibility of additional resources for the potential consequence of receiving sanctions down the road, should those resources not produce the desired improvement. Second, California chose to target a very broad range of schools as potentially un— derperforming. Schools that qualified for II/USP had scored in the bottom half of the state (Deciles 1-5) on the STAR tests and had failed to meet their growth targets in the qualifying year.6 Schools could also be identified if even one subgroup failed to achieve the comparable growth target. The assumption/ hope was that the potential for identification would encour- age improvement efforts across this broad range of schools. Assistance and Sanctions under II/USP Once schools were identified, PSAA incorpo— rated a system of assistance and consequences for those participating in II/ USP. The legislation placed a strong emphasis on the role of improvement planning and on the value of assigning “External Evaluators” to assist these schools during the planning process. PSAA also threatened potential negative consequences for II/USP schools that did not meet their API growth targets. Funding and Assistance. II/USP schools in the first year (planning year) received $50,000 to create and implement an Action Plan for school improve— ment. Subsequently, they received implementation funding from one of two sources: state funds ap— propriated for 11/ USP and funds from the federal CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Comprehensive School Reform (CSR) program.7 State—funded schools received $200 per pupil for the following two years to implement their Action Plan, while CSR—funded schools received the same level of funding for the following three years to implement a research—based school reform model. Schools participating in II/USP were required to hire an “External Evaluator” during the planning year to assist in developing the Action Plan for school improvement. External Evaluators were individuals or organizations with experience in school reform, ap— proved by the state to assist low—performing schools. The assumption underlying this requirement was that lower—performing schools often lack the capacity to identify problems and institute changes on their own. Therefore, they need external “eyes” and assistance to move forward. Sanctions. Since II/USP was intended to be a vol— untary program for schools, schools that participated in II/USP made an explicit trade—off of additional resources over three years for potential consequences at the end of this period should those resources not result in improved student performance. Schools that did not improve sufficiently over the course of the funding were subject to intervention by the Super— intendent of Public Instruction, who would have the authority to take such actions as reassigning the prin— cipal or reorganizing or closing the school. The threat of sanctions was expected to increase motivation (effort) to improve student achievement. As will be discussed later, the large number of schools that failed to meet growth targets through II/USP led to the de— velopment of a new program to monitor and provide additional support to these schools. PHASE 3: Fine Tuning of PSAA to Focus on the Lowest-Performing Schools—Implementation of HPSGP and School Assistance and Intervention Teams (SAIT) Initial evaluation of the PSAA revealed that pro- grams like II/USP and GPA had no significant or sus— tained impact on student achievement across partici- pating schools when compared with similar schools that did not participate in the programs. More specifi- cally, while some II/USP schools appeared to benefit and improve considerably, others did not, and some even lost ground. The net effect across all schools was negligible. One concern raised about 11/ USP was that CALIFORNIA’S ACCOUNTABILITY SYSTEM. it targeted a wide range of schools, some of whom were more demonstrably in need of resources and assistance than others, and that the level of funding was simply too low to produce results in the neediest schools. Responding to this and other concerns, the state modified the programs after the first few years in order to focus on a smaller number of the lowest performing schools and to better address those that failed to make substantial improvement during 11/ USP participation. Narrowing of Eligibility Criteria As stated above, the criteria for identification of II/ USP schools were broad and thus resulted in a large number of schools identified as underperforming. This policy design has the advantage of capturing the attention of a broader range of schools (schools in the middle performance levels as well as low performers), but also has the disadvantage of potentially draw— ing in schools that have less need of the additional resources and assistance. Schools that had previously made all their targets but missed on one subgroup goal were as eligible to participate as schools that had failed to make any targets on multiple occasions. In order to concentrate resources where they were most needed, the state—funded High Priority Schools Grant Program (HPSGP), established in 2001 as a part of PSAA, narrowed the identification of schools to those with API scores in the lowest decile (including II/USP Decile 1 schools). HPSGP provides funds for planning and three years of implementation to schools that apply and are selected for the program. Schools participating in HPSGP receive twice as much implementation funding per year ($400 per pupil) as those only participating in II/USP. Thus HPSGP focused a larger amount of funds towards a smaller number of high—need schools.8 Defining Consequences for Schools That Did Not Improve Schools that volunteered to participate in II/ USP made an explicit trade—off, choosing additional sup- port and resources in exchange for subjecting them— selves to potential sanctions should they not improve. Initially, schools in II/ USP were scheduled to be sub- ject to sanctions if they did not meet their growth tar- gets for both 11/ USP implementation years. However, those schools that showed some positive growth in PACE a 55 56 either implementation year—but did not meet their growth targets—received funding for an additional implementation year.” Schools that showed no growth in both imple- mentation years became “state—monitored” schools and entered into a contract with a School Assistance and Intervention Team (SAIT).10 SAITs are teams of educational consultants—often retired educators, and other individuals from private companies, county of— fices of education, and nonprofit organizations—who work with and monitor schools to improve student achievement.11 A school remains state-monitored until it makes positive API growth for two consecutive years. The consequences for schools that do not show growth through the SAIT process have been unclear. Both the HPSGP and SAIT programs focused ad— ditional funds and assistance to the lowest performing schools in the state: the Decile 1 schools and the II/ USP schools that continued to fail. At this point it is still too early to tell definitively if these two programs are realizing better results than 11/ USP did. See Chap— ter 5 of this volume for preliminary findings on the effectiveness of HPSGP. PHASE 4: State Accountability in Federal Context—”PSAA Meets NCLB" The federal context has always played a key role in California’s accountability program. Federal pro- grams and legislation including IASA in 1994 and Goals 2000 have served as catalysts for the standards movement, the implementation of aligned assess— ments, and the public reporting of disaggregated test scores. They have influenced the evolution from Cal— ifornia’s original curriculum frameworks to the cur— rent—day content standards. However, with the imple— mentation of NCLB, accompanied by a new layer of rigorous targets, identification criteria, assistance, and consequences for low—performing schools, the need to reconcile the federal and state systems has come to the forefront. Here we outline some of the key differences between the components of NCLB and California’s PSAA. Conflicting Targets In place of the API, NCLB requires Califor~ nia and all other states to set targets for assessing whether schools and districts are making “adequate yearly progress” (AYP) toward the goal of all students achieving proficiency on state tests in reading and mathematics by 2014. Following federal guidelines, California’s AYP criteria encompass several measures, including participation rates and “percent proficient" on state assessments (CST and CAHSEE), API perfor- mance for elementary and middle schools, and gradu- ation rates for high schools. All schools, districts, and numerically significant subpopulations are expected to meet specific targets within each of these categories. API and AYP targets differ in several important ways, with implications for school identification and the allocation of resources and consequences. First, while annual API targets are based on an expected increase in aggregate school performance as measured along an index scale, AYP targets set out an absolute percentage of students who must be at or above a specified level—proficiency on the CST——each year.13 Second, as mentioned earlier, a school’s API score increases more when the achievement levels of the lowest performing students increase, thus rewarding growth from the bottom upward more than from the middle upward. This structure can create an incentive for schools to raise the achievement of their lowest performing students. AYP targets, on the other hand, may encourage schools to focus on moving students from the “basic” to “proficient” categories, potentially neglecting the lowest performing students, in order to increase the percentage of students in or above this targeted band. Third, AYP also incorporates English Learners and special education populations into the list of subgroups and requires each subgroup to meet the same performance standards of all other students. The API initially did not. Finally, AYP includes gradu— ation rates and the API as the NCLB—required “other academic indicators” of school performance for high schools and elementary/ middle schools respectively. One consequence of these differences in formula is that it is possible for a school to meet one set of tar— gets without meeting the other, in any given year. For example, a school could increase test scores enough to meet their API growth targets schoolwide and among significant ethnic/racial and socioeconomic subgroups, but not meet the percent proficient target for their EL students and therefore fail to make AYP. Alternatively, a school could meet all percent profi- cient targets (as well as the additional targets such as participation rate) to make AYP, yet not increase their scores enough in one year to meet their API growth CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 targets. In 2004—05, nearly 1,561 schools met their API growth targets,13 but did not meet all 2005 AYP criteria. Alternatively, 1,200 schools met all 2005 AYP criteria, but did not meet their API growth targets. One additional important difference between the two systems is that NCLB also holds districts accountable for meeting AYP targets. Districts must have 95% participation rates, as well as specified percentages of students meeting proficiency standards. The PSAA did not incorporate district-level targets. Differing Consequences Under NCLB, schools are subject to sanctions if they fail to make AYP for two consecutive years. At this point the school enters “Program Improvement” (PI). Sanction and intervention options gradually increase in severity each year that a school does not make AYP. These interventions include notification to parents of the school’s PI status, giving them the option to change schools; provision of Supplemental Educational Services; possible replacement of school staff; and ultimately, school restructuring. A series of other instructional modifications and support en— hancements are required along the way. A school does not exit PI status until making AYP for two consecu— tive years. As of the 2005—06 school year, California has 1,772 schools identified for Program Improvement Under NCLB, local education agencies (e.g., districts) can also be identified for Program Improve— ment by failing to make districtwide AYP for two years in a row. PI districts are expected to create a plan for improvement and to implement that plan in the upcoming year. They face additional correc— tive actions in the third year.14 Districts do not exit PI status until they have made AYP for two consecutive years. As of the 2005—06 school year, 154 local educa- tion agencies in California had been identified for Program Improvement. EXPLAINING THE IMPACT OF PSAA: THEORY VERSUS PRACTICE California has made substantial progress towards establishing a comprehensive results—based account- ability system, beginning with the development of content standards and assessments and progressing to a system of targets, assistance, and consequences for CALIFORNIA’S ACCOU NTABIUTY SYSTEM low—performing schools. We now turn to the effect PSAA has had on student achievement and discuss its impact through the lens of the underlying theory of results-based accountability. Impact on Student Achievement Given the significant investment of public funds for PSAA, the state legislature required that the program’s impact be evaluated regularly (O’Day and Bitter, 2003; Bitter et al., 2005). One noticeable find- ing of the first two evaluations is that achievement for all low—performing schools (both 11/ USP and similar schools that qualified for 11/ USP but did not partici- pate) has increased sharply and significantly since the institution of the STAR testing program and the passage of PSAA.15 The gains have been the greatest at the elementary level, more moderate for middle schools, and much lower for high schools. These over— all increases in performance could be an indicator of increased attention to student achievement outcomes and low—performing schools, though it is not possible to link them directly with any aspects of the account- ability system per se. Despite this overall increase in state assessment scores, scores on the National Assessment of Educa- tional Progress (NAEP) from 1992 to 2003 reveal less promising trends.16 While these data also show that student achievement has increased over time, Califor- nia has continued to perform well below the national average in both fourth and eighth grade mathematics and reading.[7 In addition, California’s average NAEP scores are the lowest among the five most populous states in the US. (including Texas, New York, Florida, and Illinois). These trends are consistent among all racial and ethnic groups. Linking these overall trends in achievement to the implementation of PSAA is difficult given the complex environment in which PSAA was introduced. One way to evaluate PSAA’s effect more directly, however, is to examine the impact of its primary intervention programs, II/USP and HPSGP. At this point in time, analyses have not been completed to address the effect of HPSGP. However, analyses of the II/USP impact showed at most only modest and inconsistent differ— ences in student outcomes between II/USP and relevant comparison schools (O’Day and Bitter, 2003; Bitter et al., 2005). In some cohorts and grade levels, II/USP schools experienced slightly larger achievement gains PACE fl 58 in comparison to similar low-performing schools in the state during 11/ USP planning and/ or implementa- tion years. More often, however, there was no observed effect, and trends were inconsistent across cohorts and grade levels. Moreover, any modest gains observed during the period of program participation dissipated after program completion (Figure 1; Bitter et al., 2005). Thus, while there were some statistically signifi- cant but small positive effects on school—level achieve- ment gains during II/USP planning and implementa- tion, there were significant but small negative effects on the growth of II/USP schools relative to their comparison schools in the years after the program ended.18 For some groups of schools, this negative effect on achievement growth occurred even earlier, during the implementation years. In addition, when results were disaggregated by subgroups (English Learners, students in poverty, and special education students), they revealed no consistent positive or negative 11/ USP effects on achievement gaps among groups of students. Overall, [1/ USP showed no advan- tage over similar schools that did notparticipate in the program. One main reason for the program’s limited overall effect is that achievement trends among II/USP schools varied widely, such that some schools appeared to benefit substantially from program participation and funding while others gained little or even lost ground. We will explore additional potential reasons for this lack of effect in the discussion of issues that follows. The Theory of Results-Based Accountability The lack of observed benefits for programs like 11/ USP, coupled with the challenges that have arisen with the overlap of the state and federal systems, forces California now to reflect on and refine its accountability system. It is an opportune time to examine what is working within the system, what should be recon- sidered, and what to do to resolve the differences between the federal and state programs. To that end, we move to an analysis of the impact of the system in light of its underlying assumptions. While California’s current accountability system has evolved in many ways since its inception, the basic components of the program have remained quite consistent over time. These include standards, setting FIGURE 1: API Performance of II/USP Cohort 2 and Comparison Elementary Schools,a Controlling for School Characteristicsb 700 680 660 640 620 600 580 560 540 520 500 480 460 440 420 400 API Time of li/ USP award 1997-98 1998-99 1999-00 2000—01 Year 2001-02 End ofII/USPaward —+— Non” __ IIUSP -O- iIUSP 2002-03 2005—04 a. ”Cohort 2” consisted of 430 schools that began ll/USP participation in 2000—01 The pool of comparison (“non—II/USP”) schools were those that qualified for, but did not participate in, Il/USP that year. b. The analysis models controlled for a variety of school—level variables including student ethnicity, percentage of students eligible for free/reduced price lunch, percentage of English Learners, student mobility, average parent education, and percentage of teachers with full credential. Source: Bitter et al. (2005) CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 out what students are expected to know and be able to do at key points in their schooling career; targets for assessing progress toward the standards; data on student performance (through assessments aligned with standards) made available to all stakeholders; identification of schools (and districts) for improve- ment; and assistance and consequences to stimulate growth. These components reflect a now nearly universal model of results—based accountability that identifies the school as the primary unit of account— ability, holds schools accountable for producing specified results, and extends assistance and/ or re— sources to schools that lack the capacity to do so. As part of a broader standards-based reform strategy, this accountability approach is intended to foster instructional improvement and student attainment of common, challenging standards. While the specific provisions of results—based accountability policies vary from jurisdiction to ju- risdiction, they are based on a common set of assumptions (a generic “theory of action” Agyris and Schon, 1978) about the mechanisms and causal rela- tionships through which accountability is supposed to work (O’Day, 2002; Finnigan and O’Day, 2003). In the case of PSAA, the primary components of the system are expected to: "sitinnsitui I Focus attention on student outcomes and low—per— forming schools. The ability of the system to effec— tively focus attention depends on the use of clear and consistent targets and the availability of valid and accessible information on student perfor- mance and school processes contributing to that performance. I Motivate educators (and others) to expend the effort needed to improve. Targets themselves serve as motivators, but accountability systems rely largely on incentives such as promises of rewards or the threat of sanctions tied to results. I Ensure that schools and educators have the capacity to implement strategies for improvement through: Developing knowledge and skills of school person- nel to interpret relevant information on learning and to select appropriate strategies in response. Allocating resources where they are most needed. These assumptions are shown graphically in Figure 2. As we have found little to no overall achievement effect of California’s intervention programs for low- performing schools, it is important to consider how California’s overall accountability program fulfills the assumptions outlined above. Below we discuss find- ings related to each of these assumptions. 19 FIGURE 2: Simplified Accountability Theory of Action Model RESULTS-BASED ACCOUNTABILITY POLICY Components of State Accountability System I Clear targets/identification of schools I Information on achievement I Consequences for performance (interventions Garner Attention and rewards) Develop Capacity (0f educators] (educator and public) to school) outcome goals Enhance motivation to improve \ Improved instruction and School Functioning / Increased Student Achievement PACE W1 59 60 Focusing Attention on Student Outcomes and Low-Performing Schools PSAA, like other performance-based accountabil- ity systems, seeks to focus the attention of the public and the educational system on the improvement of student achievement and on low—performing schools. Clear standards and targets, coupled with useful in— formation on student performance, enable educators to make informed decisions about strategies for im- provement. In addition, the identification of schools most in need of improvement focuses the attention of states, districts, and educators on the specific needs of these schools. The first II/USP evaluation (O’Day and Bitter, 2003) found that PSAA had indeed focused the atten— tion of educators on student achievement outcomes, as measured by the API, and on the improvement of low—performing schools. School personnel were aware of their API scores, targets, and deciles. They knew that they must achieve both overall growth and suf— ficient subgroup growth to meet those targets each year. Data also suggested that the establishment of specific goals for each school may have contributed to the overall achievement growth in the state. In ad— dition, the API seemed to be garnering more support as it incorporated greater and greater emphasis on the California Standards Tests, which practitioners viewed as better aligned with their instructional goals, and thus a better indicator of their students’ progress towards state standards. Perhaps equally important, the evaluation found that PSAA also focused attention on the lowest per— forming schools in the state. Not only was this atten— tion evident at the state level, but it was also taken up by many districts, often leading to additional specific actions and programs within the district to support low-performing schools—in some cases whether or not those schools were participating in II/USP. The state focused attention even more on the lowest per— forming schools through the implementation of the HPSGP. However, by 2005, the time of the second II/USP evaluation (Bitter et al., 2005) the environment had changed. The implementation of NCLB, as discussed earlier, had created an additional layer of achieve- ment targets and expectations for educators in California. While both API and AYP measure school performance based on several measures of student achievement, the differences outlined earlier have made the targets for student achievement growth less clear and have somewhat diffused educators’ at- tention as they attend to multiple targets. Some practitioners participating in the second evaluation of II/USP (Bitter et al., 2005) did not see the additional layer of targets as a major distraction, saying that their plans to improve student achieve- ment overall should meet the needs of both the state and federal programs. Others, however, at both the school and district levels, reported that having two separate accountability systems in place has confused and/or overwhelmed school and district staff. Keeping track of these two sets of targets and thinking strate- gically about how to meet them were considered dif- ficult, particularly in light of the differences in student populations targeted through the two policies. In addition, more and more schools are being identified for program improvement each year under NCLB. AYP incorporates many specific targets, and schools only need to miss one of them to fail to make AYP. As a result, an increasing pool of “under- performing” schools is demanding the attention of state policymakers and district administrators. The expanded pool has the potential to divert attention from schools that are truly the lowest performing and that need the most assistance. In addition, while there is considerable overlap, some schools are identified as low-performing under one system, but not in an- other. These differences send mixed signals to educa— tors about the progress they have made in improving student outcomes. In addition, it is becoming more and more likely that higher performing schools will be identified as “low—performing” according to the federal standards. These changes call into question the ability of the combined state and federal accountability systems to continue sending clear signals that will motivate edu— cators who are working towards school improvement, and to maintain a focus on the lowest performing schools. Enhancing Motivation of Educators While increasing and focusing the attention of educators on student achievement is necessary to implement systemic improvement, educators must also be motivated to establish goals for improvement and work hard to improve instruction and student CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 outcomes. A key goal of the PSAA is to motivate edu— cators to implement strategies for school improve- ment—most explicitly through the establishment of clear targets and the provision of extrinsic rewards and sanctions for meeting or not meeting those targets. Realistic and Fair Targets Targets can enhance the motivation of school practitioners by providing reasonable steps to reach their ultimate improvement goals. However, the research literature on motivation and employee per— formance suggests that targets will only be effective to the extent that individuals believe the targets are attainable (Locke and Latham, 1993). Alternatively, unreasonable targets can serve as a disincentive if staff believe that no matter what they do, they will never reach the goals set out for them. The implication for school accountability is that to motivate educator performance, targets must be perceived as realistic and fair. Data from the II/USP evaluations indicate that practitioners believed the growth targets associated with the API were generally reasonable, while federal AYP targets were often not. In particular, practitioners spoke of the difficulty they faced in meeting NCLB’s absolute targets for certain groups of students, espe— cially California’s large English learner and special education populations (Bitter et al., 2005). Practitio— ners also noted that the many ways to fail AYP made it difficult to know whether increasing efforts on their part would actually result in success. Incentives—Sanctions and Rewards. Rewards and sanctions for performance were also incorporated into PSAA as explicit incentives to motivate improve— ment. PSAA’s theory of action anticipated that both the threat of sanctions (should a school fail to im— prove) and the promise of financial rewards (should a school meet its growth targets) would serve to in— crease schools’ attention and motivation to improve student achievement. However, for promised conse— quences to motivate educator performance, they must be salient to school staff—i.e., staff must be aware of the incentives; they must value the outcomes; and they must believe the incentives will be implemented under the conditions outlined by the state. Sanctions. Prior to the implementation of SAIT, the salience of the sanctions threatened by II/USP was weak (O’Day and Bitter, 2003). While some CALIFORNlA'S ACCOUNTABlLITY SYSTEM practitioners found the threat of sanctions to be dis- heartening to their improvement efforts, others were skeptical that the severe sanctions (e.g., state take- over) would actually be implemented. This skepti- cism was born out by subsequent events: when many schools did not meet their growth targets for both years of II/USP implementation, most were simply provided another year of additional funding while some (those that failed to meet their growth targets during both years of II/USP) were assigned a School Assistance and Intervention Team (SAIT). Now the question arises: what do school personnel be— lieve will happen to schools that continue to fail within the SAIT program? The only data available to address this ques— While increasing and focusing the attention of educators on student achievement is necessary tion comes from the second II/USP evaluation (Bitter et al, 2005) and suggests some cause for concern. Noting that the to implement systemic improvement, educators , must also be motivated... consequences for continued failure had not yet been spelled out, some principals reported that the resulting uncertainly made staff feel that the outcome was out of their control, thus creating a potential disincentive for them to put in extra effort to improve. By contrast, others simply dismissed the most severe sanctions as unlikely to ensue, regardless of performance. With respect to the federal—state program interac- tions, sanctions associated with PSAA have continued to lose salience with the implementation of NCLB. Teachers and administrators generally considered the sanctions associated with not making AYP to be more severe than those associated with missing API growth targets, and therefore reported placing greater focus on avoiding sanctions associated with the federal system. Practitioners were aware of the potential to undergo corrective action or restructuring as a PI school, and were aware of the sanctions affecting many of them currently, including letters sent to parents allowing for school choice. It is important to note, however, that this situation may change as the number of schools identified for improvement rises and it becomes in- creasingly difficult to mete out the specified NCLB sanctions, especially the more severe ones. Should this PACE a 61 62 pattern develop, the salience of the NCLB sanctions may also dissipate over time, eroding any motivational benefit they might otherwise produce. Rewards. As described earlier, the PSAA also orig- inally provided a positive incentive for schools to im- prove, through the GPA program. In the case of this program, implemented in 2001 and 2002, the aware- ness and salience of awards—therefore their motivat- ing power—appeared to be minimal. Though recipi- ent schools were pleased to have received the rewards, respondents did not see them as a primary factor mo- tivating continued growth. This weak incentive power may have stemmed from a distrust that schools would actually receive the rewards even if they reached their targets. The majority of school staff in the first II/USP evaluation study (O‘Day and Bitter, 2003) did not expect rewards to ensue if their outcomes improved. The late disbursal of funds during the implementa— tion of GPA, and the ultimate termination of the awards, is likely to perpetuate this distrust for years to come. At this point, no system of positive incentives exists in California’s accountability system. Developing Capacity for School Improvement Focusing the attention of educators on student outcomes and then providing incentives for them to change these outcomes are two key components needed for the systemic improvement of schools. However, educators must also have the requisite knowledge and skills, and schools must have the re- sources, to develop and implement strategies for ef— fective change. Thus, developing the capacity of edu— cators and systems (including schools and districts) is vital to successful school improvement efforts. For one thing, people are more motivated to work harder and make changes if they have the capacity to do so. Perhaps more importantly, educators and systems require capacity to enable change to occur. Key forms of capacity include appropriate knowledge and skills of educators, adequate resources for schools and districts, and system and school infrastructure to sup— port and sustain improvement. Many outcome—based accountability policies now involve some form of professional development or other assistance to build the capacity of teachers and schools to implement effective strategies, improve instruction, and understand how to interpret and use data to inform instruction. PSAA included two components to develop educator and school capacity: external assistance and additional funds to implement school improvement strategies.20 Shortcomings in both design and implementation of these components may have contributed to the policy’s limited effects. Improving Capacity of Low-Performing Schools. External Evaluators. For II/USP, the External Eval— uator hired during the planning year was expected to assist the school in developing a plan for improve— ment. Specifically, the PSAA legislation required the External Evaluators to provide technical assistance to the school site. Evaluators were to work with the school community team to identify weaknesses that contribute to below average performance, make rec- ommendations for improvement, and begin to devel- op an Action Plan to improve academic performance at the school. Schools were then expected to follow through on this Action Plan during the subsequent implementation years. This policy placed considerable emphasis and faith on the capacity—building efforts of the school improvement planning process. While the planning year was generally implemented as the legislation specifies, there was wide variation in both the quality and depth of the planning and in the quality and ca— pacity of the External Evaluators and their organiza— tions. Even where External Evaluators were strong and the planning process was considered successful, their influence on subsequent practice was often minimal. One potential reason for the lack of relationship between planning and changes in either practice or outcomes is that the planning process was divorced from implementation in many respects. Specifically, the legislation did not include expectations for moni- toring of implementation or continued support from the External Evaluator. Nor did it specify particular ways for schools to target instruction in their im— provement strategies. For example, capacity—building strategies and instructional coherence were factors that were seen to contribute to schools’ growth during II/USP. However, schools were differentially able to build capacity and instructional coherence. We expect HPSGP may experience a similar disconnect as well, since this policy did not further specify the link be- tween the planning year and implementation years. In contrast, the SAIT process lays out specific requirements for teacher and principal training:1 and CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 adds guidelines to ensure that teachers use data to inform instruction. However, SAIT still relies on the knowledge and skills of the external team hired by the district and school. These teams face a huge chal- lenge in attempting to turn around underperforming schools. Their efforts are, at a minimum, based on an initial audit and three monitoring visits throughout the year. Given the magnitude of the task facing school personnel, this level of intervention and assistance is relatively small. While many SAITs provided techni- cal assistance above and beyond that specified in the guidelines, others were only able to provide what was required. In addition, there is still no system in place to assess individual SAIT’s effectiveness in improv— ing these low—performing schools. Thus, the ability of these teams to provide meaningful capacity-building assistance and support still remains to be seen. Funds for School Improvement. The development of improvement strategies can only go so far without the requisite resources to implement the strategies. Given the inequities in the resources children bring into schools, most educational systems attempt to incorporate some degree of resource targeting to high-need, low—performing schools. The PSAA has attempted to do this in several ways by allocating ad— ditional funds to low—performing schools through the II/USP and HPSG programs, but again, problems in design and implementation may have undermined impact. The first II/ USP evaluation (O’Day and Bitter, 2003) found that schools generally spent their funds on goods and services directly related to instruction, including support providers, professional develop- ment and release time, instructional materials, and instructional personnel. However, schools often received the funds so late in the school year that they were unable to use them in the ways intended. Planning in II/USP schools was hampered by late ar— rival of funds to pay the External Evaluators, coupled with the tight deadlines for carrying out the required activities before the plan had to be submitted to the state. Similar delays in subsequent years meant that schools were often unable to implement all of the activities laid out in their Action Plans, especially if those activities involved professional development or purchases to occur before the start of the fall semester. In some cases, districts advanced money to schools in anticipation of the forthcoming state funds, but when CALIFORNlA’S ACCOUNTABlLlTY SYSTEM the number of affected schools in the district was high, this accommodation was not always possible. A further concern that emerged in the second II/USP evaluation was the ability of schools to main- tain their improvement strategies when funding ended. Several schools discussed ways in which they would either have to eliminate or reduce programs implemented during 11/ USP or staff hired during II/USP upon exiting the program. In addition, the achievement analyses indicated a drop—off in achieve— ment gains among 11/ USP schools relative to similar non—II/USP schools after the funding ended. While this drop~off could be due to a variety of factors (e.g., the increased pressure placed on non-II/USP schools from the federal accountability provisions), one pos— sible explanation is that the level of resources without II/USP funds is insufficient to maintain the growth they were making. The recent budget crisis in Califor- nia has only exacerbated these resource constraints. While HPSGP provides even more concentrated funding to the schools most in need, this program will also come to an end for participating schools. With limited ability to carry over funds beyond the four years of participation, schools will face an even greater differential between funding levels during and after program participation. Improving the Capacity of Systems— the Missing Link The drop in relative growth of II/USP schools in relation to comparison schools after program partici- pation, coupled with reports from schools that they were unable to maintain strategies after the additional funding ended, raises questions about the long—term infrastructure supporting these schools. Although PSAA included components to address school—level capacity, it did not include provisions to address system—level capacity, such as the adequacy of the base resources available to all California schools or the ca— pacity of districts to support their schools during and beyond participation in programs like II/USP. Adequate Resource Base. The improvement of low-performing schools has long been a difficult and intractable process. In California, a contributing factor may be the overall level of resources available to all schools. Accounting for variations in the cost of education, California schools are among the most poorly funded in the nation. In 1999—2000, spending PACE % 63 66 Consistent and Fair Targets. In order to send clear and consistent signals and maintain the focus on the lowest-performing schools, we recommend that state policymakers focus on harmonizing the accountability systems by better aligning the targets associated with the two systems. California should have a coher- ent system of standards, targets, and goals that build on the most effective aspects of the two policies. I Recommendation 1a:To send consistent signals to schools, the state should continue to press the federal government to negotiate a unified set of targets that incorporates the best of both the PSAA and NCLB. The unified set of targets should be realistic and fair. The 2005 II/USP evaluation (Bitter et al, 2005) suggested that a growth model for ac— countability is important. Practitioners in II/USP schools generally reported that the PSAA’s focus on aggregate growth in school performance was a more realistic goal for improving low-performing schools than AYP’s focus on getting more students over a particular proficiency bar. The API growth calculations place more weight on the growth of the lowest—performing students, many of whom are concentrated in low-performing schools. In addition, the API targets recognize the growth and progress of low-performing schools that may still be far from meeting the proficiency standards associated with AYP. On the other hand, the AYP includes alterna- tive measures, such as graduation rates, that are important indicators of school and student progress. In addition, AYP sets the same high expectations for all subgroups, such as English Learners. In establishing one set of clear targets for schools within a unified accountability system, the state may consider including some of these same standards and indicators in API. However, the targets must be set appropriately to ensure that they are realistic and fair. We should note that neither set of targets incorporates measures of individual student growth. The lack of a longitudinally-linked data system in California prohibits the track— ing of this growth over time, and has also limited the state’s ability to take advantage of the federal government’s pilot program for growth models. We address the need for such a system below under Goal 3. Common Criteria for Identification of Schools and Districts. In order to further align the state and federal systems, the state should consider methods to better align the criteria used to identify schools for intervention. I Recommendation 1b: The identification of schools for interventions through PSAA and NCLB should be aligned. At this point, varying numbers of schools have been identified as “low-performing” within the multiple systems of support and accountability. For example, 1290 schools were identified for II/USP, 290 of which were also identified for participation in HPSGP. An additional 368 were identified for HPSGP alone, or HPSGP and the federal CSR pro— gram. Overlapping with these schools, 1772 are in Program Improvement (identified through NCLB). Finally, 163 are now being state-monitored in the SAIT program.23 The state could further focus on having a more consistent pool of schools identified for sanc- tions under both federal and state programs. In determining the criteria for identifica— tion, the capacity of the state to intervene in schools should be considered. Aligned System of Support and Consequenc- es. Both the federal and state accountability programs provide systems of support and interventions for schools that do not improve. To make the consequences of poor perfor- mance clear to all educators, the sanctions for schools and districts within the federal and state programs could be better aligned. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 I Recommendation 1c: The nature of support and consequences for failing schools should be aligned across PSAA and NCLB. The state has made progress towards this alignment through the development of the nine Essential Program Components (EPCs) that are currently being used to guide im- provement strategies in SAIT schools and PI schools. Additional measures to maintain this alignment will send clear signals to schools and districts about assistance and conse- quences they will face if improvement is not forthcoming. GOAL 2: Allocate sufficient additional resources and assistance for low- performing districts and schools, and take appropriate actions when improvement is not forthcoming. The state will continue to have low—perform— ing schools and schools with high—need student populations. To assist these schools, we recommend that the state implement a triaged system of support and intervention that focuses intense efforts and resources on the lowest-performing schools. Such a system should include policies for the identification of schools, supports, and consequences that are realistic and fair and that address gaps in achievement among groups of students. The programs must provide the necessary re— sources for change, and effective assistance to develop coherent capacity building strategies for improvement. The II/USP and HPSGP programs have taken initial steps towards such a system, but have not yet proven to have the desired effects. The recommenda- tions below address next steps for these in— terventions. Targeting of Resources. First, the state must consider how best to allocate additional funds most efficiently and effectively to enable school improvement in its low-performing schools. The current budget crisis has limited the funding available for these allocations, and the pool of schools considered “low-per- forming” continues to grow. The state must therefore carefully consider which schools should be targeted for additional resources, and allocate sufficient funds to those schools to enable improvement. I Recommendation 2a: The state should target additional resources to the schools most in need, and should appropriate sufficient additional funds for these schools to develop and maintain effective improvement strategies. California has made considerable progress towards this goal over the past few years. While II/USP targeted a wide range of schools, HPSGP concentrated its funds in approximately half the number of schools, all of them among the state’s lowest performers. In addition, HPSGP doubled the program’s supplemental expenditure from $200 per pupil (under II/USP) to $400. This concen— trated attention and support runs counter to the current requirements for NCLB. As more and more schools are identified for Program Improvement under NCLB, the state must consider ways to maintain the focus on the lowest—performing schools developed through HPSGP. In addition, the state should examine closely the time frame needed to make effective change in schools, and maintain funding across this time frame. A sudden drop in funding has the potential to undermine effec— tive strategies initiated with assistance funds. Providing schools with greater flexibility to carry over funds and additional opportunities to apply for continued funding if they have made progress could ensure that schools use their money effectively over time and ease the transition out of the supplemental assistance program. PACE $1 67 CRUCIAL ISSUES Finally, the state and districts should pay particular attention to the development and deployment of instructionally strong lead- ers and teachers in low—performing schools. Incentives for experienced and high qual- ity teachers and principals to work in these schools could help build and sustain essential capacity for improvement over the long haul. Provision of Skilled External Support. PSAA provided a system of support for low—per- forming schools through the provision of an External Evaluator to assist schools in their planning efforts. However, the knowledge and skills of the External Evaluators, the intensity of their support, and the strategies developed during the planning year varied considerably among schools. I Recommendation 2b:The state should take steps to ensure that external assistance providers have the capacity to effectively assist low-performing schools. Ongoing monitoring of external assistance providers, including SAIT providers, will help to ensure that the lowest-performing schools receive strong support to enable them to im— prove. The state should examine how best to assess which providers have been successful and under what conditions, and should select external assistance providers for continued work based on these assessments. The provid- ers should be held accountable for providing ongoing support, guidance, and monitoring to schools during and beyond the planning year. The state should also develop methods to build the capacity of assistance providers in the state to provide high quality and effective support. In addition to monitoring, external assistance providers need to have the skills to provide guidance and assistance in develop- ing internal capacity at the schools and a co— herent instructional program. Improvement efforts should incorporate capacity-building strategies such as instructional collaboration and professional development, particularly around the use of data to inform instruction. To this end, assistance from external agents should be focused in this direction. Support around these strategies should be intensive and should address the needs of the unique student populations at each school, in par— ticular English Learner and special education populations. Consequences for Continued Low Perfor— mance. Inevitably, some schools, even with additional and appropriate resources and support, will fail to improve. The state must consider how best to address persistent lack of growth. I Recommendation 2c: The state should develop a plan to address schools that continue to fail. The consequences put in place should be piloted for effectiveness, and should then be implemented consistently over time. California’s system of consequences for persistent failure has lacked consistency. Because the criteria and terms of sanctions have changed over time, educators have been unclear about what consequences they will face for success or failure. Now, with the NCLB consequences rising to the forefront, schools are facing possible sanctions from multiple sources. This situation is likely to increase the confusion around sanctions and thus to decrease their motivating qual— ity. To avoid this outcome, the state needs to consider how to implement a unified system of consequences for schools that continue to fail, one that is clear and consistent over time. The consequences need to be realis— tic in scope—i.e., possible to implement given available funds and resources for sup- port—~and fair. The state should consider strategies that other jurisdictions have tried or discussed, for example contracting out school governance or reopening a school in a different configuration or with a different IN CALIFORNIA EDUCATION 2006 governance structure (e.g., a charter school). Specified sanctions should be based on evi— dence demonstrating their effectiveness in other locales or states. GOAL 3: Build a long-term infrastructure to support the continuous improvement of schools throughout the state. The recommendations above address ways to align and enhance the state’s accountability system, including additional resources and support to low—performing schools. However, no system will result in meaningful increases in student learning and achievement without adequate and equitable resources and infra- structure to back it up. The state needs to develop a structure that will continue to pro— vide adequate resources and supports to all schools in the state. This structure must ac— count for the varied needs and circumstances of schools across California. Adequate Base Level of Resources. Califor— nia’s base level of resources needs to be exam— ined, particularly in high-poverty low-per— forming schools. If the state’s most challenged schools are starting at a disadvantaged level of resources, it is unrealistic to expect short— term monetary infusions like 11/ USP to have a sustained impact. I Recommendation 3a:The state should define and allocate the base resources needed to produce and sustain academic success across all schools, taking into account the unique needs of each school and district’s student population. Given that schools reported being unable to sustain key strategies implemented through II/USP, sufficient resources to sustain improvement efforts must be an important component of any program that provides focused resources for change. The state should examine carefully the resources necessary to address the needs of the state’s schools, and ensure that these resources are allocated effectively across schools. The level of resources should take into account the individual needs of students in each school, providing appropriate funding to support effective strategies for English Learners and special education students. To this end, several foundations have funded a set of school finance studies in California (see Chapter 3) that will systematically examine the funding needed for schools and districts, based on the state’s goals for student achievement and on student needs. These studies should provide valuable information to inform the resource standards for the state. Since the district is a key intermediary be- tween state-level policy and school-level im— plementation, the state must also ensure that districts have adequate resources to provide the necessary assistance and support to their schools and that they allocate them to low- performing schools as needed. District Capacity and Accountability. Given the substantial impact that districts can have on building the capacity of their low— performing schools, districts are an impor— tant vehicle, not only for ensuring effective implementation of programs such as II/USP, but also for facilitating growth among all low—performing schools in their jurisdiction. Recognizing this key role, state policymakers should ensure that districts have the capacity to assist and support their schools undertak— ing improvement efforts. Districts should then be held accountable for creating and maintaining conditions that ensure success within programs like II/USP and for provid— ing long—term and sustained support and monitoring to maintain progress made. I Recommendation 3b: The state should build the capacity of low-performing districts and of districts with high numbers of low-performing schools. PACE an 69 CRUCIAL ISSUES Districts should then be held accountable for the performance of their students and schools. Districts should also be encouraged to play a key role in this effort by ensuring that funds are allocated in a way that supports their lowest-performing schools. One district in the 2005 II/USP Evaluation, for example, al— locates dollars to schools on a per pupil basis, rather than in full—time equivalents (FTEs). Under such a plan, schools with less experi- enced and therefore less expensive staff have additional resources available to spend on in— structional resources or additional staff. Such student—based funding systems within dis- tricts generally feature weights, ensuring ad- ditional district resources for schools enroll- ing high percentages of students with special needs, such as students in poverty, English Learners, and special education students. This is just one example of ways in which districts can allocate money to better align resources with the needs of their schools, based on the characteristics of the students they enroll. Recognition of Success. Focusing only on the identification of failing schools and on punitive consequences for these schools has the potential to discourage educators. While such negative incentives may be unavoidable in a comprehensive accountability system, the fact that there is currently no program in place to recognize schools that have greatly improved may hinder the motivation fostered through PSAA. In fact, schools that improve in programs like II/USP lose supplemental resources when they succeed in meeting their growth targets. Such “punishment” for doing well can serve as a disincentive. I Recommendation 3c: The state should implement positive incentives that recognize and reward the improvement of low-performing schools and motivate educators to continue improvement efforts. To ensure proper acknowledgment of suc— cess, high poverty schools that are beating the odds should receive public acknowledge- ment. This acknowledgment can serve both as a continuing motivator for growth, as well as a means to identify models for other high poverty schools that are not succeed- ing across the state. Successful, high poverty schools can provide concrete examples of what can be done to successfully improve student achievement. Given the state’s fiscal constraints and the lack of evidence that monetary rewards provide a meaningful incentive, however, the state should think creatively about other forms of incentives for educators. For example, the state could consider incentives for highly qualified and experienced teachers to teach in high-need schools, incentives such as reduced regulatory requirements for districts to sup— port schools’ improvement and additional technical assistance to schools undergoing improvement initiatives. Individual educators can be motivated by a variety of incentives including good working conditions, oppor— tunities for growth through professional de— velopment, and opportunities for meaningful collaboration with colleagues. Comprehensive Longitudinal Data System. The state has created a number of systems to provide useful and accurate information on student and school level academic perfor— mance for educators and the public. Educa— tors can now examine aggregate progress of schools over time and can use yearly disag— gregated assessment data to plan instruction and strategies for school improvement. In addition, parents can access information on school performance, demographics, and other indicators. However, several vital forms of data are still lacking, including a longi- tudinal dataset that provides information on individual students’ progress over time. Also lacking are frequent data on student IN CALIFORNIA EDUCATION 2006 achievement that teachers and principals can use to inform instruction and school—level strategies for improvement. I Recommendation 3d: The state should develop and fund the implementation of a comprehensive, longitudinal, accessible data system that will make possible the analysis of school, student, and district level progress over time in more accurate and useful ways. In 2002, Senate Bill 1453 authorized the establishment of such a data system. Since then, however, internal disagreements among state—level departments have hindered prog- ress on the system’s development.24 More recently, the allocation set aside by the State Legislature to support districts in collecting and entering the initial requisite data for this system was suddenly cut from the 2006 state budget. For this reason, we strongly urge the various branches of the state govern— ment—the Legislature, the Governor’s office, Finance, and the CDE—to work to resolve whatever issues continue to hold up this effort. We also encourage the state to provide the funding necessary for districts to imple— ment this system, at least during the initial start—up period when new data collections and systems must be established at the local level. Such a database and information system is a statewide concern; its quality should not be dependent on the uneven capacity and will at the local level. From the state perspective, a comprehensive student—level dataset would enable more ef— fective evaluation of educational programs and policies over time. School progress could be assessed based on the growth of indi- vidual students across years, rather than on improvement from one cohort of students to another. In addition, a system that also tracks teacher assignments would enable further evaluation of programs that could impact teacher mobility. These data would allow for finer-grained analyses, which would enhance research on the effectiveness of programs and policies. Finally, the presence of such a system would enable California to conduct further negotiations with the federal government to establish a growth model for AYP in California. In addition, of course, such a system would have direct applicability for instructional improvement at the local level. It would allow educators to track individual students’ progress over time, a particularly important goal given the high level of student mobility in many California districts. A longitudinal system would enable educators to examine past performance of students who attended other schools or districts in California, and then plan an instructional program appropri— ate to the students’ needs. To ensure that such a system is used effective— ly and to its potential, the state must consider ways to make the data accessible to educators and researchers while maintaining student privacy. Educators should have access to in- dividual students’ records of performance and teacher assignments for students in their jurisdiction in order to plan instructional programs. In addition, educators should be trained how to access and use these data ef— fectively Use of Benchmark Data. Beyond a central longitudinal system of data, the state should also consider ways to foster the use of data to directly inform instruction. In both a recent EdSource report (Williams et al., 2005) and the second 11/ USP evaluation, systematic analysis and use of data to inform instruction was identified as a key factor for growth in student outcomes and higher achievement in high poverty schools. I Recommendation 3e: The state should support training for principals and teachers on the use of benchmark and PACE $2 71 72 formative assessment data as a feedback mechanism for informing instruction and identifying students for extra support. Respondents in the second II/USP evalua- tion emphasized the importance of frequent (e.g., monthly or every 6 weeks) benchmark assessments tied to the school’s curriculum and to the state standards. Related ongoing communication and collaboration around data among grade-level and cross-grade-level colleagues were vital to this process. Thus, we recommend the state incorporate such a focus into its professional development and leadership training programs to better build the capacity of teachers and principals to regularly and effectively use data. Successful systems are already in place in some districts and schools that are realizing considerable gains in student achievement. For this reason, the state role may include further identifying such systems and en- couraging their continued development and dissemination. Those systems already in place were typically based on assessments as- sociated with districtwide curricula tied to the state standards, and were supported by software packages that enabled the analysis and dissemination of assessment results in a user—friendly format. CONCLUSION The state has made significant strides towards creat- ing a coherent, results—based accountability system in California. An aligned set of standards, assessments, curricula, and targets are in place, and the PSAA al— locates resources and sets consequences based on this aligned system. Nonetheless, the state must continue to examine the ability of this system to garner the at— tention of educators and the public towards student achievement, to motivate educators to implement new strategies for instructional improvement, and to build the capacity of educators and schools. The current context in California, particularly the implementation of NCLB and fluctuating fiscal constraints, has placed additional stresses on the cur- rent system. NCLB has diffused the clear signals in place when PSAA was first implemented, and past fiscal crises have strained the base resources available to schools undertaking improvement efforts. In the coming years, the state needs to think creatively about how best to accommodate these stresses in order to reestablish a coherent and effective system. It needs to consider ways to align the state and federal systems so as to establish one unified and coherent accountabil- ity program of targets, assistance, and consequences. In doing so, the state must determine how best to build on the most effective aspects of the federal and state systems; to target resources effectively and ef- ficiently; and to create incentives that are realistic and fair. Recognizing and supporting the role of districts in these efforts will be key to their success. ENDNOTES The Improving America’s Schools Act (IASA) was the name given to the 1994 reauthorization of the Elementary and Secondary Education Act (ESEA) of 1965. When Congress reauthorized ESEA again in 2001 they called the revised legislation by a new name, the No Child Left Behind Act (NCLB). California’s standards and results—based accountability have been influenced by both reauthorizations of the ESEA. 2 Later, NCLB further specified the content of the SARC, requiring information such as AYP results and the extent to which “highly qualified” teachers are teaching core subject areas. For a school with an API score below 800, for example, the annual performance target is to grow by 5% of the difference between its actual API score and 800. For a school with an API score of 800 or above, the target is to maintain a score of 800 or above, hopefully increasing over time as well. lt is important to note that these policy choices, individu— ally and collectively, result in markedly different criteria from those that NCLB would require of all state account— ability programs two years later. We address the conflicts between PSAA and NCLB programs in later sections of this chapter. An alternative accountability system has been established to hold accountable schools with fewer than 100 pupils, special education schools, and alternative schools. When 11/ USP was first implemented in the summer of 1999 (prior to the availability of API growth informa— tion), the CDE invited schools scoring in the bottom CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 half of the state’s schools on the SAT—9 for two consecu- tive years (1998 and 1999) to submit an application for II/USP. Schools entered II/USP in three cohorts. The first cohort of 430 schools entered II/USP in 1999; Cohort 2 included an additional 430 schools in the fall of 2000; and 430 were included as Cohort 3 in the fall of 2001. II/USP schools were chosen to represent a range of grade levels, SAT—9 deciles, and geography. Formerly known as the Comprehensive School Reform Demonstration (CSRD) program. The CDE selected 351 schools that had never participated in II/USP or CSR for this new program. In addition, 307 schools that had been participating in either II/USP or CSR received additional funding through HPSGP. Cohort 1 jointly-funded schools received one year of implemen— tation funds under HPSGP, Cohort 2 schools received two years of implementation funds under HPSGP, and Cohort 3 jointly—funded schools received all three years of implementation funds under HPSGP. A third year of implementation for state—funded schools _. 0 Schools could also enter state monitoring if they made some growth during the two implementation years (but did not meet growth targets), but experienced negative or no growth in any year thereafter. .... .— The SAIT first assesses whether a school has the “es- sential program components” (EPCs) deemed necessary for improved student achievement. The EPCs vary by grade level, but in general include components such as the implementation of State Board of Education—adopted curricula, AB 75 training for principals, and the imple— mentation of an assessment system to monitor student progress. In schools where these components are missing or not fully implemented, the goal is to fully implement them. 12 Neither system accounts for the achievement growth of individual students over time. 13 Both schoolwide and “comparable improvement” targets ‘4 Corrective actions could include the replacement of district staff or appointing a state trustee in place of the superintendent, among others. 15 For example, Cohort l II/USP and comparison elementa— ry schools gained approximately 190 API points between 1998 and 2004, while Cohort 1 II/USP and comparison middle schools gained an average of approximately 107 points since 1998. Cohort 1 II/USP and comparison high schools increased by an average of approximately 79 API points in this same time period. 16 NAEP, a norm-referenced and criterion-referenced test of student achievement, has been administered nationally approximately every two years since 1990. '7 When states were ranked by average performance on NAEP tests between 1990—2003 (normalized on the na- tional mean and standard deviation), California ranked lower than all other states except for Louisiana and Mississippi (Carroll, et al., 2005). CALIFORNIA’S ACCOUNTABILITY SYSTEM 18 After the first two years of implementation. However, many schools did receive an additional year of funding. 19 We draw primarily on findings from the 2003 and 2005 state-sponsored independent evaluations of the II/USP. 20 For HPSGP, the legislation requires a school site Action Plan, but does not require a planning year nor the hiring of a state—approved External Evaluator. However, nearly all schools did participate in a planning year, and schools were expected to hire an external entity for support in de- veloping the Action Plan. Schools could work with their district or county office for this support in developing the Action Plan, and the legislation for HPSGP required that school staff and administrators participate in specific trainings, including AB466 professional development associated with the state—adopted curricula. 21 Including training for teachers and principals on State Board of Education—adopted instructional materials and intervention programs (AB 466 and AB 75) 22 Texas, New York, Florida, and Illinois 23 230 out of the 368 HPSGP schools that were not in II/ USP are also in Program Improvement. 24 SB 1453 requires the establishment of the California Longitudinal Pupil Achievement Data System (CALPADS) that includes statewide assessment data, enrollment data, and other demographic data required to meet federal NCLB reporting requirements. In addi- tion, unique student identifiers must be assigned to all K—12 students enrolled in California public schools. (See: http://info.sen.ca.gov/pub/O1—02/bill/sen/sb_1451—1500/ sb_1453_bill_20020927_chaptered.html.) REFERENCES Argyris, C., & Schon, DA. (1978). Organizational learn- ing: A theory of action perspective. Reading, MA: Addison Wesley. Bitter, C., Perez, M., Parrish, T., Gonzalez, R., Socias, M., Salzfass, L., Chaney, K., Gubbins, P., Anand, P., Dawson, K., Yu, V., Delancey, D., St Esra, P. (2005). Evaluation Study of the Immediate Intervention/Underperforming Schools Program of the Public Schools Accountability Act of 1999. Palo, Alto, Ca: American Institutes for Research. Carroll, S.I., Krop, C., Arkes, 1., Morrison, P., Flanagan, A. (2005). California’s K—12 public schools: How are they do— ing? Santa Monica, CA: RAND Corporation. Esch, C.E., Chang—Ross, C.M., Guha, R., Humphrey, D.C., Shields, P.M., Tiffany-Morales, ].D., Wechsler, M.E., 8r Woodworth, KR. (2005). The status ofthe teaching profes- sion in 2005. Santa Cruz, CA: The Center for the Future of Teaching and Learning. Finnigan, K.S. & O’Day, ].A. (2003) External Support to Schools on Probation: Getting a Leg Up? (In Press) Research report jointly published by the Consortium for Policy Research in Education and the Consortium on PACE I 73 74 Chicago School Research, University of Pennsylvanian, Philadelphia, PA. O’Day, IA. (2002). Complexity, accountability, and school improvement, Harvard Educational Review, 72, 3. 293- 329. O’Day, LA. and Bitter, C. (2003). Evaluation of the Immediate Intervention/Underperforming Schools Program and the High Achieving/Improving Schools Program of the California Public School Accountability Act of 1999. Palo, Alto, CA: American Institutes for Research. Williams, T., Kirst, M., Haertel, E. et al. (2005). Similar stu- dents, dijj‘erent results: Why do some schools do better? A large—scale survey of California elementary schools serving low-income students. Mountain View, CA: EdSource. CRUCIAL ISSUES IN CALIFORNlA EDUCATION 2006 N THE DAYS \VHEN THERE STILL WAS SUCH A Tl'llNG AS: _. .e%=> A C ALI E 0 RN T A ASS {BSA/1175.19 T P R0 G RAA '1 (CAP i), N EW SEAT) ERS JAR(§.)UI\TD THE STATE ROUTlN ED“ PUBHSHED SCHOOL TEST SCORES. FOR THOSE \"i’HO KEPT TRACK AND {Abilil lT \‘x‘AS A DlSHEARTENlNG R1'Ti_,b’xl_., MOSTLY TOR JTS PRET)lC’J‘r‘ti‘leJT‘t. TT \‘\—‘8 liAlRlX EASY TO Jed—{El '1 JCT A SCH O 0138 SCORE EROM YEA R TC) YES/Alt. lxNOlwlN C} NOTHlNG MORE T 1 if", J Tl l Ji PR 15“»‘lC‘1 L78 YEARS SCORE. And while there may have been some marginal changes here and there, yearly rankings were quite consistent. Schools at the bottom tended to stay there. What made bad news worse was the inexorable connection between the socioeconomic status (SES) of a school’s student population and test scores. California has long since stopped administering the CAP, but low achievement persists among large numbers of students— particularly those who are poor, do not speak English, and whose parents lack formal education. What has changed is how policymakers have structured the problem of persistently low levels of student achievement. The impetus for that change has been the enactment of state accountability laws (in combination with No Child Left Behind) and the de- velopment of curriculum and performance standards. As a result, it is now much more difficult, if not im- possible, for policymakers, teachers, administrators, school boards, and the public to simply accept persis- tently low student performance as an unpleasant fact. Enactment of the Public School Accountability Act (PSAA) in 1999 created a massive and complex regu— latory structure in the state that holds schools respon- sible for student achievement. A critical feature of the accountability system is a variety of state interven— tions—a combination of technical support and sanc— tions—that are meant to force schools to address the problem of low student performance in their schools. The year PSAA was enacted, there were roughly 1.2 million students in Decile 1 through 3 schools. The average Academic Performance Index (API) of those schools was 473, with a low of 297 and a high of 561.1 Of those students, 79% were eligible for free or reduced meals, 45% were English learners, 67% were Hispanic, 12% were African—American, and 32% of students’ parents had only a high school educa— tion, while 38% of students’ parents did not have a high school education.2 The characteristics of Decile 1 schools in 1999 were similar though more pronounced. 75 76 The average API for the 685 Decile 1 schools was 417. Of those students, 75% were Hispanic and 13% Afri- can—American, 56% were English learners, and 86% were eligible for free meals. About one-half of the parents of children in Decile 1 schools did not have a high school education. In rather stark contrast, in 1999 there were slightly over 1 million students in Decile 8 through 10 schools. Average scores in these schools was 787. Among students in those deciles, 14% were Hispanic, 4% African-American, 8% English language learners, and 18% eligible for free meals. Among the parents in this group, 35% were college graduates and 21% had graduate degrees. School accountability and the sanctions associ- ated with them raise the question of what to do about low—performing schools to help them improve. The Immediate Intervention/Underperforming Schools Program (II/USP) that Bitter and O’Day discuss in Chapter 4 was the policy centerpiece for state inter- vention. Policymakers believed that a combination of discretionary funds, school—wide planning, and exter— nal technical assistance would coalesce into solid gains in teaching and learning in these schools. As Bitter and O’Day show, these programs were voluntary and only enrolled a fraction of eligible schools. In addition to II/ USP, the Legislature created the High Priority Schools Grant Program (HPSGP) to target Decile 1 schools. This chapter examines state efforts to improve instruction in the lowest performing schools—those receiving HPSGP funds. The chapter first assesses past strategies to address the problems of persistent low achievement in schools, a problem that is most acute among schools that serve large numbers of non—Eng— lish—speaking, students of color from disadvantaged socioeconomic backgrounds. It then discusses pre— liminary findings from an ongoing study of schools in the HPSG program.3 The paper identifies some of the strengths and weaknesses in the program and makes recommendations for its improvement. COMPENSATORY PROGRAMS AND SCHOOL IMPROVEMENT California’s Public School Accountability Act (PSAA) signals an important shift in how policymakers and practitioners think about the problem of low student performance. In theory, the change is away from a compensatory, regulatory model based on categorical program support to a capacity building and account— ability model. It reflects a major change in how the problem of low student achievement is defined and how solutions to it are structured. Most significantly, the focus has shifted from low-achieving children to low-performing schools. This sea change in education policy redefines educational roles, responsibilities, and professional relations within the public education system. The history of compensatory education is syn— onymous with the history of Title I of the Elementary and Secondary Education Act. A primary purpose of the law was to provide financial assistance to school districts that suffered from the adverse “impact that concentration of low—income families have on [their] ability. . .to support adequate educational programs.”4 Its other purpose was to provide direct support to children by funding programs to meet their “special needs.”5 In an effort to secure local compliance—to guarantee that federal funds were flowing only to eligible students—the US. Office of Education cast an ever widening regulatory net. While these efforts are well documented, it is important to note that regulations implementing Title I focused primar- ily on changing the legal and political organizations of schools.6 These regulations—which included the elaboration of substantive and procedural rights, the requirement for clear audit trails for local expendi- ture of federal dollars, federal and state sanctions for misuse of funds, the growth of a vast state and local bureaucracy to monitor local compliance, and the empowerment of local community groups as a countervailing force to local school authori— ties—eclipsed the pedagogical dimensions of federal compensatory aid. The policy framework of Title I shaped behavior in schools in several unintended ways that, in the long term, inhibited organizational effectiveness. The preoccupation of policy with regulatory compliance denigrated instructional practice by undercutting professional judgment and authority and fragmenting both schools and students. Instead of focusing on the whole child, policy dissected children into disparate program targets. Though it seems naive in retrospect, federal policymakers believed that stretching a regula— tory net over schools could overcome the incapacity, CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 ineptitude, or indifference of local schools serving poor, low—achieving students. While such strategies did force some schools to improve, it undermined those educators who were making good-faith efforts to serve those children (more in Chapter 8). State policies directed at schools serving disad- vantaged students mirrored federal policy. California’s funding counterpart to Title I was Economic Impact Aid, which targeted low—income, minority students. At the end of the 19705, the largest administrative unit within the California Department of Education was the Field Services Unit, which was responsible for monitoring and reviewing local compliance with federal and state compensatory programs. The state regulatory framework for education was rooted in distrust of the motives and capacity of local school officials. At the state level, officials came to share Washington’s belief in stressing compliance as distin- guished from assistance.7 The major difference between the regulatory model and the accountability model is that under the previous model schools could be sanctioned for fail- ing to follow rules, but they could not be sanctioned for not teaching students. Implicit in both federal and state policies was the belief that schools could develop effective programs for disadvantaged children without paying attention to the overall quality of the school. Simply put, they believed that good programs could trump bad schools. The shift toward accountability and student outcomes began with the Hawkins—Stafford Amend— ments to Chapter 1 (which replaced Title I during the Reagan years) enacted in 1988. Among the many changes initiated by the legislation, the most impor— tant were those concerning program coordination, school-wide projects, school performance account— ability, and parental involvement. The amendments marked a significant shift in Chapter 1 policy by em— phasizing program effectiveness and accountability. Chapter 1 schools were required to develop student outcome goals, and schools failing to meet those goals were required to develop school improvement plans. Congress also urged districts to adopt local standards and measures of student progress that were based on proficiency. As Bitter and O’Day point out in the previous chapter, California moved in a similar direction. In part, this was due to federal requirements contained EVALUATING STATE INTERVENTION: THE HIGH PRIORITY SCHOOLS GRANT PROGRAM in the reauthorization of Title I. The law required states to develop performance standards and assess— ments as a condition of receiving federal funds. As the authors show, California was first among states to develop curriculum frameworks, academic content standards, and assessments. Enactment of the Public School Accountability Act (PSAA) in 1999 completed the shift to an outcomes—based accountability system in which schools, theoretically at least, were respon- sible for the academic progress of all students, and instructional improvement superseded regulatory compliance. THE HIGH PRIORITY SCHOOLS GRANT PROGRAM A central feature of California’s school accountabil- ity system are programs to engage low—performing schools in improvement efforts. One of these, which is discussed in the previous chapter, is the Immediate Intervention/Underperforming Schools Program (11/ USP). The other, the High Priority Schools Program (HPSGP) is similar, but targets Decile 1, rather than Decile 1 through 5, schools. While the two programs are structurally similar, the HPSGP places some ad- ditional requirements on schools. The HPSG Program was created by Assembly Bill 961 (Chapter 747, Statutes 0f2001) to provide additional funds to the lowest performing schools in the state. To be eligible for funding, schools must rank in the bottom decile of the state’s API. Participating schools receive $400 per pupil for a period of three years. Districts are required to match state funding with $200 per pupil annually. Over the life of the program, this amounts to $1,800 per pupil or $1.4 million (including the local match) for a school of 900 students—the average school size in the HPSG program. A few of the largest schools received over $5 million. Over the three-year funding cycle——2202-03 through 2004-05—HPSGP allocations to districts were slightly over $754.9 million. In return, schools had to meet state benchmarks for improved student academic performance. Schools failing to improve face various sanctions and interventions, including state takeover and dissolution. HPSG schools must also participate in a state— specified professional development program for PACE a 77 78 teachers (AB 466), yet another for principals (AB 75), and must purchase state—adopted textbooks in read- ing-language arts and mathematics. Schools must also hire external consultants or “evaluators” and create “action-plan teams” to assist in developing a school improvement plan—the components of which are specified in state law. Finally, schools must comply with a long list of state requirements specifying parent and community engagement in the improvement process. The structure of the HPSGP is such that schools receive funding for three years with the possibility of an additional year if they are making “adequate progress.” The implied rationale is that three years of funding will result in significant capacity building and sustainable organizational improvement. Suc— cess is conditioned by several factors. First, it requires schools to take a long View; yet, schools are under considerable pressure to show short—term results in order to avoid sanctions. Second, it assumes that schools have the flexibility and autonomy to allo— cate HPSG funds in a manner that is consistent with reform priorities and objectives. As a condition of receiving HPSGP funds, schools must also agree to engage in various activi- ties. Teachers and administrators are required to participate in professional development programs.8 Schools must purchase approved, standards—aligned textbooks in reading and mathematics if they do not already have them. They must also engage external consultants to assist in the development of a school Action Plan. Schools may apply for a $50,000 plan— ning grant to develop their actions plans. The Action Plan must be based on an initial needs assessment, it must be research based and data driven, and must encompass a strategic plan for helping low—perform— ing students. The legislation lists a number of options that may be included in the strategic plan. They include common planning time for teachers, support staff, and administrators; mentoring for site administra— tors and peer assistance for teachers, particularly new teachers; professional development activities, particu— larly in mathematics and reading and literacy; and incentives to attract credentialed teachers and quality administrators.9 External evaluators are required to engage parents throughout the planning process, and each school’s site council is required to sign off on the school improvement plan. While legislation creating the HPSGP provides a lengthy list of school improvement actions that schools may take, the legislation is more permissive than prescriptive. Schools are required to address pupil literacy and improvement; quality of the staff; parental involvement; and adequacy of facilities, cur— riculum, instructional materials, and support services. Legislation provides various examples of how schools may address those issues, but leaves it to districts to adjust the details to the specific needs of each school. More than anything, the legislation embodies a set of expectations for schools about how they might ad— dress instructional improvement. The bill’s language places considerable emphasis on comprehensiveness, collaboration, planning, assessment, focus on read— ing across the curriculum, community engagement, mentoring, professional development, and beginning teacher training. The measure delineates the essential components of a school improvement plan, but leaves schools considerable room to develop a plan that meets local conditions and needs. In order to assess their progress in meeting aca— demic growth in core curriculum areas and to moni— tor the efficacy of their school improvement plan, schools are strongly encouraged to revisit their action plans and to modify them as necessary. After three years of participating in the program, a school that has not met its growth targets or has failed to show “significant growth,” as determined by the State Board of Education, is required to enter into a contract with a school assistance and intervention team (SAIT). Members of the SAIT are individuals who “possess a high degree of knowledge and skills in the areas of school leadership, curriculum and instruction aligned to state academic performance content and performance standards, classroom disci— pline, academic assessment, parent—school relations, and have proven expertise specific to the challenges inherent in low—performing schools.” (EC 52055.650. (1)(A) Finally, schools that fail to meet their growth targets are subject to various sanctions. These in— clude reassigning students to other schools, reassign— ing teachers, renegotiating the collective bargaining agreement, reorganizing the school, and closing down the school. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 THE HPSGP AND SCHOOL IMPROVEMENT The shift from a school accountability system driven by inputs, regulation, and compliance to a system based on outcomes necessitates a major shift in the process of schooling and a new conception of the organization of schooling. The traditional view of the school sees it as a collection of classrooms. Teachers had considerable autonomy in what to teach and how to teach. More importantly, teachers were respon— sible only for their classrooms—for what went on behind the classroom door. Organizational theories of schools described them as “loosely-coupled organiza— tions,” whose commonalities were anchored in “myth” and “ritual” that had little to do with the underlying technology of teaching.10 Organizational coherence was thought to be imposed by textbooks, teacher preparation and professional development, and some level of super— visory oversight. Consistent with theories of “loose coupling” were school decision—making theories de— scribed as the “garbage—can” model. Instead of a co— herently articulated model of decision making based on organizational goals and strategies to attain them, decision making in schools was best described as in— dividuals reaching for readily available solutions to satisfy immediate needs. Both the traditional school organizational model and decision-making model are the antithesis of coherent, long-term, organizational planning. Consequently, a central hypothesis of the study was that HP schools that improved were ones able to transform themselves from a collection of class— rooms into coherent, purposeful organizations. We assumed that this would occur as schools increas- ingly were held directly accountable for the perfor— mance of their students. Unlike in the past, it no longer mattered if the first grade teacher was doing a wonderful job with her students if the other teachers in the school were not. It soon became evident that the need for orga— nizational coherence and collaboration was all the more important in low-performing schools. As Table 1 shows, Decile 1 schools serve large numbers of poor children, many of whom are English learners, and come from families lacking formal education. Unlike high-performing schools that serve high SES students, EVALUATING STATE INTERVENTION: THE HIGH PRIORITY SCHOOLS GRANT PROGRAM organizational factors are likely to be much more important in low-performing schools. As a group, students in low—performing schools lack the social and intellectual capital of high SES students. As orga- nizations, high performing—schools (largely because of their students) can continue to do what they were doing. Low—performing schools, on the other hand need to learn how to do things very differently. Doing more of the same is unlikely to raise levels of student achievement. 1 I The strategic underpinning of the HPSGP is that an infusion of money, external technical support, a comprehensive school plan, and the threat of sanc— tions for failure to perform will catalyze the kind of organizational transformation that turns low-per— forming schools into high-performing (or, at least higher-performing) ones. The policy assumes, more- over, that three years is sufficient time to build the necessary capacity in schools to affect those changes. STUDY METHODOLOGY The magnitude of the state’s investment in low-per— forming schools—nearly $3 billion over a three-year period—begs the question of what difference that investment has made. This study does not attempt to answer that question on a systemic level.12 That is, the study does not address the question of whether schools receiving HPSG funds did better as a group than those schools that did not. Instead, it seeks to answer the question of whether some schools partici- pating in the state intervention program were more successful than others in meeting student achieve— ment goals. Did schools that met their API growth targets each year and by all subgroups share common characteristics? Conversely, what did schools that did not meet growth targets have in common? As noted earlier, the legislation that created the HPSGP gives schools considerable flexibility in designing improve- ment strategies. While such flexibility is desirable—as it allows schools to tailor school improvement strate— gies to their particular needs and circumstances—it also creates the possibility that program funds will be mismanaged and wasted. The schools we studied ex— emplify both outcomes. The study’s research questions flow from two key features of the HPSGP. One was the amount of money PACE fl 80 it provided schools. Participating schools received $400 per pupil for each of three years. For the average Decile 1 school of 900 students, that meant over $1 million. Some schools received as much as $6 million over the three-year period. The fact that funding was limited to three years became problematic for most schools. Over the three years that schools were in the program, they had to show continuous progress toward meeting API growth targets. If schools failed they were subject to various sanctions, beginning with the assignment of a school assistance and improvement team to oversee and direct improvement efforts. Additional sanctions could lead to reassignment of the teaching and admin— istrative staff or, in the worst case, school closure. How these features play out in the program’s implementa— tion is discussed in greater detail below. Data for this study comes from school-site visits to 15 schools. Of the 15 schools we studied, 10 were high schools and the remainder elementary schools; 11 were urban, while 4 were rural. The site visits com- prised structured interviews with principals, teach- ers, HPSGP and special program coordinators, and school-site council members. Interviews took place between February and May of 2006. In order to assess the impact of participation in the HPSGP, we wanted our sample to include both what we regarded as improving schools and non- improving schools. Initially we classified improving schools as those that had meeting API targets for all subgroups for each year of program participation. Due to the difficulty of getting access to districts, we relaxed this condition to include two schools TABLE 1: Comparison of Selected Characteristics of Case Study Schools to all California Schools School Indicators Mean Minimum Maximum Enrollment 1400 ~260 ~SOOO % English Learners 4O 1 1 76 All Decile 1 schools 46 All CA schools 25 % Free Meals* 70 45 97 All Decile 1 schools 80 All CA schools 50 % Minority** 92 69 100 All Decile 1 schools 90 All CA schools 64 % Full Teacher Crd. 90 73 100 All Decile 1 schools 88 All CA schools 94 API Base Scores 04—05 619 531 726 API Gain (01-02 to 04-05) 138 68 240 % change 30 14 62 Avg. Parent Education*** 1.97 1 3 2.7 All Decile 1 schools 2.0 All CA schools 2.56 Source: California Department of Education * ”Free Meals” represents students who are eligible for the free and reduced lunch program; it is a proxy for poverty. ** The”Minority”category comprises Hispanic and African American. *** Average parent education is represented by values from 1 to 5 where 1 represents "Not High School Graduate”and 5 represents ”Graduate School.” CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 that had not met program targets in the first year, but made them in subsequent years. The so-called non-improving schools were those that had missed API growth targets in two or more years. The CDE assigns “red,” “yellow,” and “green” lights to schools based on their success in meeting growth targets. Our “improving” schools all had green lights in the last three years, while non—improving schools had red and yellow lights. Table 1 profiles the schools in our study sample. The data in Table 1 reveals that the study sample of schools mirrors the general population of Decile 1 schools fairly closely. It also shows that Decile 1 schools generally—and this is certainly true for the schools in our sample—have a higher percentage of poor, non—English—speaking students than the aver— age school in the state. Parents of students in Decile 1 schools and in our sample have lower levels of formal education than the state average. Roughly half of the parents in our sample have not completed high school. The average parent education level for all stu- dents, on the other hand, is to have completed some college. Parent education is particularly important in relation to student achievement due to the high correlation between the two as measured by the API. Nearly 50% of the differences in average school API scores is explained by differences in the average edu— cation levels of parents. STUDY FINDINGS: THE CONDITIONS FOR SCHOOL IMPROVEMENT While we classified schools as improving or non—im— proving in selecting schools for study, in reality the classification was not clear-cut. We found signifi— cant variation among “improving” schools as well as among “non-improving” schools. One improv- ing school had a gain of 146 points over a four—year period, while another had nearly 300. Non—improv- ing schools may have missed their targets by only a couple of points or may not have tested enough students. Some schools that fit the non—improve- ment profile had many of the same organizational characteristics as the improving schools. On the other hand, some schools that were, by most organizational standards, fairly chaotic and unorganized, were in fact improving. EVALUATING STATE INTERVENTION: ’ THE HIGH PRIORITY SCHOOLS . GRANT PROGRAM Initially we thought that it might be possible to clearly differentiate the characteristics of improv- ing schools from non—improving schools. The real— ity turned out to be less clear-cut than we originally envisioned. Some of our non-improving schools were engaged in many of the same activities as our improving schools, yet API gains were quite different. Conversely, schools that might have been predicted to be non—improving made considerable gains in their API scores. However, as we looked more carefully at schools in their efforts to improve, we were able to identify some key factors that affected implementa— tion of the HPGSP. In the following section we discuss the factors that either facilitated or impeded implementation of the HPSGP. Figure 1 contrasts factors facilitating school improvement as opposed to factors impeding school improvement. As already noted, these do not map perfectly into improving and non-improving schools, but they do represent extreme ends of a con- tinuum for school improvement. This section focuses on what we consider to be the most important differ— entiating features of improving and non—improving schools. Organizational Stability and Continuity Without doubt, among the most significant fac— tors facilitating school improvement are organization- al stability and continuity. These two factors can take various forms. In our study we found a high degree of trust among teachers, administrators, and support staff. In improving schools generally, there was a high level of professional regard among faculty, support staff, and administrators. In interviews, teachers, pro- gram specialists, and principals talked about the im— portance of working together as a team. In one high school in particular—coincidentally the one with the greatest increase in API scores—teachers talked about how they not only enjoyed working together, but also enjoyed socializing together. In addition to being colleagues, they considered many of their colleagues their friends. It should be noted that this was a fairly young faculty, with teachers possessing five or fewer years of teaching experience. Many had gone through the same master’s program in teaching at a nearby university, so they had known one another there. Even though the school did not have a principal (two individuals served as acting principals), those whom PACE w 81 82 FIGURE 1: Factors Facilitating Improvement vs. Factors Impeding Improvement Factors Facilitating Improvement Organizational stability and continuity - High degree of social capital and trust - Stable teaching staff - Stable and competent leadership - Focus on developing leadership among teachers - Focus on school as the organizational unit rather than a collection of classrooms Leadership and vision Action plan that is working/living document that reflects strategic planning - Organizational coherence . Commitment to an improvement strategy . Ongoing assessment and evaluation Collaboration and professional development External support Coherent program funding is tied to strategic plan we interviewed expressed a great deal of respect for their competence, dedication, and leadership. They trusted their professional judgments, regarded them as knowledgeable about school improvement, and looked to them for professional support. In contrast, the loss of a sense of community and lack of program continuity was quite apparent in a non—improving high school that had become a SAIT school (a school subject to intervention by a state ap- pointed School Assistance and Intervention Team). The school’s attendance area had changed within the past five years because a nearby high school had been converted to a junior high, and many students living in the area were now being bused in to the SAIT school. According to school staff, students did not feel connected to the school. Annually, there was high turnover among students (not counting incom— ing freshmen and graduating seniors). High numbers of disciplinary referrals, frequent altercations among students, and a lack of respect for others became rou— tine features of the school. When asked what seemed to be the major prob— lems confronting the school, a parent who chaired the school site council noted the high turnover among staff and students. She saw little incentive for teach— ers to stay in the school. As she saw it, there was little continuity from one year to the next. She noted rather ruefully (and at times tearfully) that: CRUCIAL ISSUES Factors impeding Improvement l Organizational instability and constant change I Organizational fragmentation and individual isolation - Classroom rather than school centric focus - High turnover among teacher and administrators I Compromised leadership - Lack of district support - High staff turnover - Lack of leadership skills I Action plan developed forfunding purposes; ignored once funding approved - No coherent or consistent improvement strategy - No commitment to change I Little or no technical assistance or support I Program budgeting is opportunistic and ad hoc We moved from years of having money— money that was well spent and helped kids—to SAIT, and now it’s as though it [the improvements] never happened. Teachers were excited about the professional develop- ment that they got. And now the state is here with its scripted learning. It’s really demoral— izing for teachers. A consistent theme among improving schools was the importance of teachers taking responsibil— ity and leadership for school improvement. This response took several forms. One was a sense among teachers that school improvement is a collective re— sponsibility and a cooperative effort. Another was the importance of common planning time, or in the case of high schools, department—specific and teacher— planned professional development activities. Teachers, administrators, and staff working together to achieve a common goal was a consistent theme that ran through the interviews in improving schools. Leadership The importance of leadership is closely con— nected to the importance of organizational stability and continuity and social capital in schools. In both improving and non—improving schools, leadership played a central role. Leadership played out in various IN CALIFORNIA EDUCATION 2006 ways: stability and longevity, expertise, collegiality, and authority. In improving schools, principals had been at the school for a number of years, and gener— ally their tenure at the school preceded the school’s participation in the HPSGP. Some had taught at the school before becoming principal, while others had held various administrative positions. The relative longevity of principals in improving schools contrasts dramatically to the rapid turnover of principals and other administrative staff in non— improving schools. The most egregious case of lead- ership instability was in a school that, over a 30-year period, had had only one principal who was in the position for more than two years. One principal lasted for three years and was demoted in the middle of her fourth year by the school board. The constant turn— over at the school, moreover, mirrored the turnover of superintendents at the district. The school board hired and fired principals at will. One school had seen six principals come and go in eight years. The fact that it was a huge school with year-round attendance tracks made the need for stability and continuity even more acute. Teachers and administrators in schools and districts with high administrative turnover and dys— functional district leadership were generally operat— ing in “survival mode.” As a result, there was little to no focus on school improvement. In one school, most teachers carried on as best they could but were demoralized, and therefore sought other jobs. In one school, teachers talked about how fragmented the school was. Improvement programs would get de— signed, but not implemented. Money for programs was non—existent; and teachers had no idea what funds might be available for school improvement. According to teachers at one school, most school im— provement programs like AVID (Advancement Via Individual Determination) existed on paper, but not in reality. As our study revealed, leadership is about more than just continuity and stability. The principal’s ability to help the school shape a vision for reform, guide development of a strategic plan, and elicit co- operation and support from the school community is another significant factor. The extent to which the school remained faithful to the goals guiding the action plan was largely attributable to the principal’s leadership. While the principal’s role in managing EVALUATING STATE INTERVENTION: THE HIGH PRIORITY SCHOOLS GRANT PROGRAM improvement within the school is important, so is the principal’s role in connecting the school to the community. In a rural elementary school that serves largely Latino children, the principal stressed the im— portance of providing leadership to the community. As a Latina, she emphasized the importance of being “a role model for girls so that they can see the that they can have professional careers.” In a school where 63% of students are English learners and 100% are eligible for free lunches, the principal believed that an important aspect of her job was to make the school a “community place.” She told interviewers that she knew the names of 99% of the students in the school, and also made it a point to know students’ families and to have dinner with them. Leadership was the glue that held school im— provement efforts together. It was the principal who helped shape a vision for school improvement, kept the school on track and focused, mobilized the nec— essary resources, and generally helped to shape the school’s culture. In our conversations with teachers in improving schools, they consistently cited the impor— tance of leadership. They praised their principal for his or her dedication, hard work, and commitment to improvement. Among schools with the greatest im— provement, teachers readily acknowledged the criti— cal role played by the principal. “It would not have happened without her,” or his or her “leadership and dedication were what has made the school success— ful” were common responses to questions about the principal’s role in improving schools. The Action Plan All schools must develop action plans that detail their reform strategies over the course of the HPSGP. The plan must be approved by the school site council. Regulations related to the HPSGP require schools to contract with an “External Evaluator,” who helps the school develop its plan. Schools are required to report annually to the CDE on their progress in meet- ing improvement goals that they established in their plans. In addition, schools have to develop a budget that showed how program funds relate to specific improvement strategies. Among HPSGP action plans there were also significant differences between im— proving and non—improving schools. The main differ- ences could be found in the plan’s coherence, ability to measure progress, and its implementation. PACE u 83 84 In improving schools, the action plan tended to be a “living” document, one that mapped a strategy for school improvement efforts. Action plans and the program budgets that supported those plans were reviewed on a regular basis. The action plans in these schools also had measurable school improvement benchmarks—based on state content and perfor- mance standards—that could be used to measure a school’s progress toward meeting its goals. If needed, strategies were changed and resources were reallocat- ed in order to meet improvement targets. The overall vision did not change, however. What changed were specific activities. In some schools for instance, in which improving student literacy was central to im— proving achievement, the schools might change spe- cific professional development activities or focus on different kinds of supplemental support if goals were not met. The focus on literacy remained constant. In non—improving schools, improvement goals tended to be fragmented and expressed as disparate programs or activities. One school, for instance, spent most of its HPSGP funds in the first year on hand—held computers that students could use to help them with their homework. For various reasons (ac— cording to some teachers, students simply got bored with them), the computers were not used and the program was abandoned after the initial year. In non— improving schools, not only were action plans lacking a coherent, articulated plan for improvement, but they also tended to be ignored after the schools had applied for their first HPSGP grants. For example, a principal in one school not only knew nothing about the school plan, but knew nothing about the HP pro- gram. The principal only learned of it two days before our site visit. At that time this principal had been in the position for six months, having taken over when the previous principal was demoted to vice principal. In other schools, the action plan had nothing to do with the school’s improvement efforts or its budget for HPSGP. It was submitted with the applica— tion and then left to gather dust. The major differ— ence between improving and non—improving schools regarding their action plans was that in improving schools the action plan was exactly what it was meant to be—a strategic plan for charting a course for school improvement. In one school in particular, fac- ulty were surveyed each year to evaluate the school’s success in meeting the goals in the action plan. If there were deficiencies in the action plan, teachers were asked how those might be remedied. In non- improving schools, the action plan existed to comply with the state’s requirement that HP schools have one. Collaboration and Professional Development One of the most striking features of improving schools is their attention to collaboration among teachers. In improving schools, it was a singular focus among teachers at each grade level or department level. To facilitate collaboration, school schedules were changed to leave a portion of one day each week for various activities such as program planning or peer coaching. In other instances, teachers were paid to participate in various school-organized workshops held either on Saturdays or during summer break—— often both. In improving schools, planning, strategiz— ing, and evaluating activities were fixed features of a school’s regular schedule. They were ongoing and focused on the school’s improvement goals. Simi— larly, professional development in improving schools was not generic—that is, provided by the district or county or some other provider—but school specific. Professional development activities were an integral component of a school’s improvement goals. Professional development activities in improving schools took various forms as they were tailored to school needs. Most common were peer and literacy coaching. It is worth noting that all improving schools placed considerable emphasis on literacy and writ— ing. Most often schools would hire literacy coaches to work on a regular, full—time basis with teachers across all subject areas with the goal of improving their stu— dents’ reading and writing skills. One high school had instituted a program of reading and writing across the curriculum so that students in all courses had regular reading and writing assignments. In physical educa— tion dance classes, students read works of literature and had to discuss themes, movement, tone, and the like and discuss how these elements might relate to choreography. In biology students had to use the Cor— nell method of note taking. In one school, peer coach— ing took the form of teams of four teachers engaged in year—long activities. They would meet regularly to discuss teaching and learning strategies, would vid— eotape one another’s classes to observe instructional strategies, and debrief on what they had learned from observing one another’s teaching. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 In non—improving schools, the most pronounced difference to professional development was the generic nature of its activities. Generally they tended to be district sponsored, focusing on broad issues related to state content standards and assessments. While these issues may be important to teachers, they may be somewhat distant from the classroom if not incor— porated through specific activities. In non-improving schools, it was often teachers, on and individual basis, who decided what professional development activities to attend. One non—improving school, for instance, a1- Iocated part of its HPSGP budget for teachers to attend conferences on gifted and talented education programs. External Support One of the chief factors facilitating school im— provement among schools in our study was an ongo— ing relationship between the school and an external agency. The strongest and most enduring relation— ships were between the school and a university. One school had a seven—year collaborative relationship with a school of education in one of the California State University campuses. The school’s participation in HPSGP was initiated by their university mentor, and university personnel assisted in the proposal’s development and implementation. The university conducted needs assessments, and helped the school identify the resources necessary to meet their im— provement goals, develop program priorities, and assist in writing an action plan. The collaboration precedes and goes beyond the HPSG program. The university partners with the school in its teacher training program. Students in the teacher credential- ing program are placed in the school for their practice teaching experience. In turn, new faculty are hired from this pool. Another school had a close relationship with one of the University of California campuses. At the time of our study, the school had been open for about six years. Most of the teachers were young, and most had recently completed the master’s program in teaching at the university. The university also used the school for its teacher training program. Teachers whom we inter— viewed all regarded the ongoing relationship with the university as a key feature of school improvement. The relationship also provided teachers with a professional anchor—a way to stay in touch with educational issues and problems beyond the immediate school setting. EVALUATING STATE INTERVENTION: THE HIGH PRIORITY SCHOOLS GRANT PROGRAM In some instances the external evaluator provided mentorship and technical assistance to the school. But the role of the external evaluator as a source of support was quite uneven among the schools in our study. Some external evaluators helped schools to conduct needs assessments, assisted them with data analysis, guided development of school action plans, and continued to work with the school over the course of the HP program. Others provided what seemed like an off—the-shelf action plan (in one in— stance the external evaluator had not even bothered to change the name of the school). For the most part there was little or no ongoing engagement or rapport with the school. The specific source of technical assistance to the schools (whether it comes from a university, a nonprofit organization such as WestEd, or a private consultant) does not appear to be significant. On the other hand, factors that matter include the techni— cal assistant’s level of engagement and relationship with the school and its faculty. As discussed earlier, low—performing schools have tremendous challenges to overcome in order to improve. If schools already had the knowledge and skills to improve teaching and learning, it seems logical that they would do so. The fact is that they do not have the organizational capac— ity to turn schools around. They need assistance from an outside source willing to take time to understand a school’s problems and to put in the time and energy to develop strategies to overcome those problems. Program Funding While just about everyone whom we interviewed agreed that program funding “had made a difference,” just what that difference was varied widely. In some schools, the money was used to “backfill existing needs.” In these schools, funding was regarded as a windfall to the school to pay for a long list of things that the school had not been able to afford out of its regular budget. In some instances this meant funding new administrative positions; hiring teaching coaches or other supplemental personnel; purchasing com- puters, software, and instructional materials; support- ing a variety of professional development activities, including paying teachers’ costs to attend conferences; buying time for teacher collaboration; contracting for technical assistance; and purchasing assessment instruments and supplies. PACE m 85 86 Consistent with our other findings, the use of additional resources among schools can be dif- ferentiated by whether funds were integrated into a coherent program of school improvement or whether they constituted a school’s “wi h list.” Schools that had already committed to a school improvement strategy generally regarded program funding as an opportunity to continue what they were doing, but with additional resources. For instance, schools whose improvement strategy focused on improving literacy hired literacy coaches to work with each grade level in elementary schools or each department in high schools. Professional development activities focused on how to integrate reading and writing into all teaching and learning activities—including physical education at one high school. Regular assessments measured how well the school was doing in meeting its student achievement goals. In contrast, the “wish— list” schools had no coherent strategy for spending HPSGP funds. Money was simply regarded as an op— portunity to take care of a variety of unmet needs. Some schools had little or no idea how much money they were receiving from the HPSGP. In some cases, they were given a budget by the district and told to spend the money until it ran out. How each school used its HPSGP funds revealed a great deal about the school’s culture. Schools that were collections of classrooms and teachers with minimal interaction, planning, or collaboration~—in short, schools that were organizationally fragment- ed—used HP monies in a fragmented, opportunistic way. On the other hand, schools with a vision and a coherent plan used funds in a purposeful manner. An— other way to explain these differences is along a con— tinuum with program spending that is “need—driven” at one end and “goal-driven” at the other. Exemplify— ing the former, one non—improving school allocated its HPSGP monies by categories: 25% to technology, 35% to professional development and supplemental instruction, 15% for materials, 5% for improving the school environment, and 20% for administrative services. According to those involved in developing the HPSGP budget, the allocation ratios reflected the need to “give a little to everyone.” The principal was pessimistic about the benefits of additional resources, as he perceived it was more money for “just doing more of the same.” Exemplifying goal—driven schools were those with highly focused strategies for changing teaching and learning. In these schools, HPSGP fund- ing was allocated to support improvement goals. An overriding issue related to the HPSGP con- cerns the duration of funding and the ability of schools to carry over unexpended funds. Everyone whom we interviewed was concerned about the termination of funding at the end of three years. Several interviewees noted that making decisions about the use of resources for school improvement is not something teachers, parents, and administra- tors have much experience doing. Implementing the HP program requires schools to develop new decision—making skills. Traditionally schools have not had $1 million or more to allocate at their discre— tion for school improvement. Principals are given their annual budgets by the district. In high schools, department heads may have a budget for books or supplies, but those budgets are generally fixed. Using HP monies effectively for school improvement places huge and entirely new demands on schools. School site councils, administrators, and teachers must be able to conduct needs assessments, develop multi— year improvement goals and strategies, evaluate progress in meeting those goals, and revise strategies as necessary. As noted earlier, to implement HPSGP successfully, schools have to learn new skills and experiment with different strategies until they find those that work. This all takes time. No one person among those whom we interviewed thought that three years was enough time to develop those skills, much less be successful in applying them. Another problematic dimension to the HPSG program, one that was raised by a number of in— terviewees, concerns the lack of oversight and ac— countability for program expenditures. As long as schools are meeting their API growth targets, the state assumes that all is well. But even when they do not meet their growth targets and a SAIT is assigned to the school, there is no review of how HP monies were used. In one school, computers that were pur— chased with HP funds disappeared; while in others, no one really knew how much money the school had, how it had been spent, or what the money was used for in the current year. Lack of budget records is particularly evident in schools that had adminis- trative turnover. Some schools simply did not have the budgets for the HPSG program for prior years. Others show expenditures that had been charged CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 against HPSGP funds, but no budget to show how those monies had been allocated or how fit into an overall program of school improvement. POLICY IMPLICATIONS Based on the schools in our study, participa- tion in the HPSGP has produced somewhat mixed results. While some schools were able to benefit from the program and regarded it as an opportunity to transform the school into an effective organization competent to serve the educational needs of its students, others regarded the program as a financial windfall and a source of discretionary fund- ing. The difference between the two was in the commitment that teachers and ad— ministrators in the school made to school improvement. The purpose of this chapter has been to develop some understanding of the implementation of the HPSGP and to identify various factors that facilitate school improvement. One of the features that sorts schools in our study is their motive for participating in the HPSGP. As already noted, some applied for the grant because they saw it as an opportu- nity for discretionary resources. Some schools did a quick calculation and realized that HPGSP funds could bring in as much as $4 or $5 million over three years. Even for the aver- age-sized school, the HPSGP generated about $1.4 million over the course of the program. Some of those interviewed admitted that the HPSGP funds were the only source of “new” money available to schools, and they applied precisely for that reason. For these schools, program participation was opportunistic. They tended to be the schools that exercised little accountability over HPSGP funds, shelved the action plan after it was written, and used funds as spending needs arose. At the other end of the continuum were those schools that I EVALUATING STATE lNTERVENIlON: THE HIGH PRIORITY SCHOOLS 'AGRANTPROGRAM. .~ Viewed HPSGP funds as an integral part of their vision for school improvement. In the short term, if state policymakers con- tinue funding future cohorts of Decile 1 schools through the HPSGP, there are several modifications that they may want to consider. Over the long run, however, improving low— performing schools may require rethinking the funding and governance structure of K—12 education in California. Oversight and Accountability Nearly all individuals we interviewed argued for greater external oversight and account- ability for schools’ and districts’ use of HPSGP funds. While the funds are intended to flow to schools, some schools complained that their district controlled the funds. However, at the school level there was little accountability for how schools spent funds once they received them. Schools that were committed to a reform agenda used the funds as they proposed in their school action plans. They reviewed the action plans at least annu- ally to see what modifications were needed. In those schools, program expenditures were guided by an improvement plan. The school site council was not an effective means to ex— ercise oversight. Often the school site councils themselves had little or no knowledge of the HPSGP other than it providing a resource stream to the school. Duration of Program Funding Three years is not sufficient time for most schools, especially Decile 1 schools, to de- velop the skills and capacity to successfully implement the HPSGP. Some of those inter— viewed suggested that the $1,200 per pupil that schools received over three years should be spread over a five- or seven—year period. As noted earlier, some schools needed three years just to develop the organizational skills to learn how to conduct needs assessments, PACE W 87 identify student learning needs, determine which resources would be most affective in addressing those learning needs, establish goals and objectives for the use of those resources, measure progress to meet those goals, and make the necessary changes, as needed, if goals were not met. Faculty and staff in most schools simply lack the skills to engage in these activities with predictable rates of success. As Stanford professor and noted organizational studies expert James March has observed, most schools lack the organizational intelligence to undertake those tasks. This does not mean that they cannot develop the necessary intelligence. But, as March would argue, organizations need time to learn.13 The current policy of funding Decile 1 schools for three years assumes that whatever problems these schools face, those problems will go away after three years. However, as long as Decile 1 schools continue to serve pre— dominantly poor, non—English speaking stu— dents, half of whose parents do not possess a high school education, the problem will per— sist. In many of these schools, HPSGP funds are used to purchase supplemental services like tutoring, time for collaboration and plan— ning, teacher support, and the like. These are ongoing needs that persist beyond the three years of funding that schools are given. Redefine the Problem The recommendations in this chapter can produce marginal changes in the implemen— tation and overall effectiveness of the HPSGP as a state strategy for improving low—per— forming schools. However, the major im- pediments to change among Decile 1 schools remain factors that are not readily amenable to policy manipulation. Policy cannot compel commitment, cannot mandate organizational stability and continuity, and cannot enforce program coherence. Years of policy efforts to reform schools have shown that many of the important components of organizational cul— ture lie beyond the reach of standard policy instruments. For this reason, policymakers need to rethink their current approach to fixing low-performing schools. One recommendation is to shift attention from Decile 1 or low—performing schools to schools that serve student populations that mirror those of Decile 1 schools—those with a high percentage of low-SES students. This group of schools encompasses a unique set of policy problems: they face greater challenges and need more assistance than the average school. In addition to their need for financial and human resources, they need focused technical assistance and mentorship. The demographics of Decile 1 schools are markedly different from the average school in the state. These differences will not go away after three years. Consequently, the state’s solution to improving low—performing schools needs to go beyond fixing the HPSGP program to tackling how schools like those in Decile 1 should be funded and supported. The main policy question that needs to be addressed is not how to fix low—perform— ing schools, but what state policy can do for schools that serve large numbers of educa— tionally and economically disadvantaged students. It is a more complex and politi— cally difficult problem than simply making adjustments around the edges of the current program. Its solution touches upon the struc— ture of the system of school finance and the system of governance. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 ENDNOTES 1 The scale is from 200 to 1000. 2 The actual number of students is understated. While 94% of students in Decile 1—3 schools were tested, in some schools, mostly high schools, just over 50% of students were tested. The percent of students eligible for free lunch is probably also understated since some students, mostly in high schools, either don’t eat in the school cafeteria or don’t admit that they are eligible. 3 The HPSGP study is supported by a grant from the Bill and Melinda Gates Foundation. The study is being conducted in collaboration with American Institutes for Research, who have contracted with the California Department of Education to conduct a study of the HPSGP. The Gates supported study has a much narrower focus than the AIR study in that it is concerned primarily with how schools used HPSGP funds. The collaboration benefits greatly from AIR’s generously sharing interview protocols, study findings, and assistance in arranging for the school site visits. I am especially indebted to Jennifer Harr, Tom Parrish, and Paul Gubbins for their assistance and support. 4 Yudof, Kirp, Levin, & Moran, page. 699. 5 ibid. 6 Wirt 8r Kirst (2001). 7 Kirp (1986), page 3. 8 Principals are required to participate in AB 75, Principals Training Program, and teachers in AB 466 mathematics and reading/language arts professional development pro- gram. The focus of both programs is to align instruction with state standards and assessments. 9 Chapter 749, Statutes of 200]. 1" Scott (1992). ” March differentiates, for instance, between organiza- tional “exploitation” and organizational “exploration.” Exploitation essentially means doing what you have been doing, particularly if you are doing it successfully. Exploration, on the other hand, means finding new direc- tions and new ways of doing things in order to be suc— cessful. See March, op. cit. American Institutes for Research has contracted with the California Department of Education to conduct the state—wide assessment of thee HPSGP. The case studies of schools’ implementation of HPSGP were conducted in collaboration with this study, funded by the William and Melinda Gates Foundation. This chapter is part of a larger study for the Gates Foundation. ‘3 March (1999). EVALUATING STATE INTERVENTION: THE HIGH PRIORITY SCHOOLS GRANT PROGRAM REFERENCES Kirp, D. (1986). Introduction: The fourth R: Reading, writ— ing, ‘rithmetic—and rules, in D. Kirp, 8: D. Jensen (Eds), School days, rule days (p. 3). Philadelphia, PA: Falmer Press. March, I. G. (1999). The pursuit oforganizational intelli- gence. Malden, MA: Blackwell Business. Scott, W. R., (1992). Organizations: Rational, natural, and open systems (3rd ed.). Englewood Cliffs, NJ: Prentice Hall. Wirt, F., St Kirst, M. (2001). The political dynamics of American education. Richmond, CA: McCutchan Publishing. Yudof, M., Kirp, D., Levin, B., 8r Moran, R. (1992). Educational policy and the law. Belmont, CA: Wadsworth Group/Thompson Learning. PACE I] 89 90 CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Leslie Santee Siski N . IGH SCHOOLS HAVE DOMINATED THE NEWS ON EDUCATIONAL REFORM LATELY, IN A STEADY BOMBARDMENT OF BAD NEWS, “DISTRESSING NEWS” (ED TRUST, 2006), AND EVEN “SCARY” NEWS (DETROIT NEWS, 2006) OF HIGH DROPOUT RATES AND LOW TEST SCORES AND TOO MANY STUDENTS LEFT BEHIND. Improving high schools is an “urgent challenge,” declared Secretary of Education Margaret Spellings, speaking at a 2005 National Summit on High Schools, “one of the biggest challenges our country faces.” The US. is, Time magazine told us on its cover, a “dropout nation” (April 17, 2006), and high schools, according to the LA. Times, are a “dropout industry” (Ianuary, 2006). Oprah Winfrey (2006) ran a highly publicized series on “American Schools in Crisis,” showcasing failing high schools and an “epidemic of dropouts.” The statistics from different reports convey the same conclusion, in a remarkably consistent message of alarm: I 1.2 million students each year fail to graduate; 7,000 students drop out every school day (Alli- ance for Excellent Education, 2006); I Of every 100 students who begin ninth grade, only 68 will graduate, and only 18 will actually earn a college degree (Achieve, 2006); I 3 of 10 ninth graders do not graduate within four years; 4 of 10 African—American and 5 of 10 His- panic students (Gates Foundation, 2006). Even when students do survive the “epidemic,” research reports provide more bad news: more than half are unprepared to compete in the new global economy (Murnane 8x Levy, 1996; Education Trust— West, 2004; Gates, 2006), and unready for college (Achieve, 2006; Quint, 2006; Roderick, 2006). Test scores are low, and show little sign of improvement: even the few scattered claims of success were ex- plained away as residue of elementary school reforms (Haycock & Huang, 2001) or called into question, as in the “myth of the Texas miracle” (Haney, 2000; Klein, Hamilton, McCaffrey 8K Stecher, 2000) or the “legend of the large MCAS gains” in Massachusetts (Camilli 8t Vargas, 2006). No wonder, then, that the problem of the high school has become everyone’s problem. By 2005, 87% of the public was “extremely or very concerned” about dropouts, and 83% agreed there is “extremely” or “very urgent” need to improve high schools (Alliance for Excellent Education, 2005). Whether echoing the public or shaping their opinion, leaders inside and 91 outside of government have joined the chorus of con- cern. From the President’s Office to the Governors’ Summit to corporate boardrooms, the fate of the high school seems to be on everyone’s agenda as an ‘urgent challenge.’ When it comes to high school, Chester Finn pronounced (2005), “things are falling apart,” while Bill Gates proclaimed that the very institution of the high school itself has become “obsolete.” But are things “falling”? Is the “failure” really a matter of high schools getting worse, or of something more complicated—of the context of success itself changing? Much of the “bad” news is not, if we look a bit closer, really new: accumulating evidence below the headlines suggests not that high schools have changed for the worse, but that they have not changed for the better. Far from “falling apart,” high schools are, statistically, staying stubbornly the same. While the specific numbers of dropouts are estimates, and always contentious, the overall patterns have held steady for the past thirty years. From state to state, from TIMSS to NAEP, across studies and across tests, test scores, too, remain remarkably flat. The average reading score on the NAEP test, for example, was 285 in 1971; in 2004, after some fluctuation, it was still 285. Math scores were not much better, rising from 304 in 1973 to only 307 in 2004, thirty—one years later. Despite two decades of massive reform efforts to restructure and resize high schools, despite the introduction of standards and accountability systems that promised radical change, and even despite re- ported gains at the elementary and middle schools, on almost every measure high schools have stayed strik- FIGURE 1: NAEP Flatline in Average Reading Scale Scores for Students Age 17 (1984—2004) 300 295 290 285 280 275 270 265 . = A ‘ \e 3 =\ \e 1984 1988 1990 1992 1994 1996 1999 2004 Source: National Center for Education Statistics 92 ingly stable (Achieve, 2005; Carnoy, Elmore 8r Siskin, 2003; Education Week, 2005; National Commission on the High School Senior Year, 2001; McLaughlin & Talbert, 2001). If the problem with high schools now is not that they are falling apart, then why are we experiencing their consistent performance as a “crisis”? The prob- lem, it seems, is an inconsistency of a different kind: high schools now are being asked to do something they have never done before, and something they were not designed to do (Siskin, 2003). This is a new demand for high schools: to educate all young people through to graduation, and to ensure that they all take rigorous courses and are prepared for college. Moreover, this is not only a new standard, but tw_o standards for success (or indicators of failure) that have traditionally and organizationally been held quite separate. For the first time in the history of the American high school, aspirations for increasing attainment (raising graduation rates) are in direct conflict with expectations for increasing academic achievement (raising test scores). Cross-Purposes: Aspirations of Attainment/ Expectations of Achievement Just imagine what we could accomplish ifwe truly educated all young people well. Ifall stu- dents—regardless of their race or income—at— tended a great high school. A school that challenged them. A school that excited them. A school that made them proud. Imagine if all students had teachers who inspired them to reach beyond their grasp. If they took rigor— 0115 classes that opened their minds and their futures. Imagine ifall our young people ar— rived at their high school graduation ready to attend college now and to lead our country in thefuture. —Melinda French Gates, to National Conference of State Legislatures, July 2003 While the patterns of high school achievement and diploma attainment may not be new, what is new is the meaning attached to those figures. What Me— linda Gates imagines actually describes an emerging consensus about not only what a ‘great’ high school should be, but also what every high school should CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 provide for all students. That consensus combines two different and quite distinct demands: expectations of higher standards from the policy system, and aspira— tions for higher education for increasing numbers of students. The first is a model of academic achievement that focuses on test scores and the changing expecta— tions of rigor in the policy system; the second is a model of attainment and expanding access that fo— cuses attention on graduation rates and the changing aspirations of students. These traditionally distinct, and often competing priorities, are brought together as social and economic conditions have changed, producing simultaneous pressure to raise test scores and raise (or at least not reduce) the numbers of graduates. By the end of the twentieth century economic conditions and the connections between educa— tion and employment opportunities were changing, rapidly and dramatically. Fewer jobs were open for employees without a high school diploma, and few of those would provide salaries above the poverty level. In 1950, 60% of American jobs did not require a high school diploma (Achieve, 2006). By 1973, that figure had dropped to 32%; by 2001 it had fallen to only 9% (Carnevale & Desrochers, 2003). Even fast food and factories are increasingly demanding, and choos~ ing among, applicants with diplomas. Whether these changes actually signal ‘a nation at risk’ or not, there is little question that students increasingly are at risk of being left behind if they cannot complete high school and leave with a diploma in hand. Students have clearly gotten the attainment mes— sage, and their aspirations are also changing rapidly and dramatically. “Ask any high school student in Chicago today what he wants out of high school and the answer is almost without fail, ‘to graduate and go to collegem (Roderick, 2006). High school students across the country respond the same way: in the Na- tional Survey of High School Student Engagement (2004) 90% plan to attend college (Achieve, 2005); in another, 94% do (Alger Association, 2006). Those numbers are widespread, remarkably high and rap- idly rising: nationally, the percentage of tenth graders aspiring to college doubled between 1980 (40%) and 2002 (80%) (Roderick, 2006). Between 2004 and 2006 the college plans of African American students rose from 78% to 84%; for Hispanic students from 66% to 73%. In steadily increasing numbers, the vast majority Achievement and Attainment: The Comprehensive High School and the Problem of Reform of students enter ninth grade aspiring not only to go t_o high school, but through it to receive a diploma and access to college and a productive career. The policy system has changed its expectations of high schools too, but there the most dramatic change has been to focus on achievement—on rais- ing academic performance and improving test scores through building systems of standards—based ac- countability. By 2001, 49 states had adopted new ac- countability systems, new standards, and new tests with sanctions for schools whose scores did not rise; by 2005, 24 states, including California, had moved to make high school diplomas dependent on test scores (though several had delayed deadlines for that to take effect). Here, the policy pressure has been to ensure that no child gets through high school or receives a diploma without meeting the new standards. The potential conflict between these two differ- ent demands becomes apparent as educators con— front what a New York teacher called “a lost genera- tion”—students in a new kind of limbo who did not drop out, but cannot get a diploma because they have not passed an exit exam (Siskin, 2003c). A high school student in California expressed her frustration when facing just that situation: “I need a diploma. I want it. I deserve it. I’ve been going to school and studying. I want to have a profession.” But despite her success— ful persistence through school, despite passing all her courses, she was stopped at graduation by her failure to pass one CAHSEE test (Asimov, 2006). In 2006, the first year that diplomas depended on CAHSEE scores, reports estimated that approximately 40,000 students were in just that limbo, left behind to watch their classmates cross the stage at graduation. The challenge of the 21SK century, the ‘urgent challenge’ confronting the new accountability sys- tems, the high schools, and now the states is how to accomplish both at the same time: to achieve high standards for all students, and—at the same time—to ensure that every student can reach graduation and receive a diploma. To reach that stage, as a high school teacher put it, “a revolution would have to take place” in which “the way we teach and what is taught and how it’s taught [all] have to be changed.” That revolu- tion requires changes in every aspect of the account- ability system and the high school organization. Most fundamentally, it involves challenges of design, of as— sessment, of curriculum and of teaching. PACE a 93 94 The Challenge of Design While both goals have always existed in schools, they have tended to operate at cross purposes, histori- cally alternating as educational reform swung back and forth like a pendulum: on one side, arguments for achievement and excellence; on the other calls for ex- panding access and attainment (Angus 8r Mirel, 1999; Tyack, 1967). In times like the last “age of standards”, back in the 18905 (Mazzeo, 2001), in the reforms of the academically oriented Committee of Ten, or in the mid—century response to Sputnik, the pendulum has swung toward the ‘academic core,’ toward preparation of the college—bound, the future leaders, the talented tenth. In times aimed toward increasing access and attainment, whether due to increasing enrollments, waves of immigration, or depressed employment op- portunities, the pendulum has swung the other way. Reformers called for broadening programs to focus on “useful knowledge” rather than academic subjects like algebra that “injured the mind, destroyed the health, and wrecked the lives of thousands of chil— dren” (Tyack, 1967, p. 359). “Personal English” taught writing letters instead of reading literature; social studies replaced “dry academic subjects” like history with hygiene and “life skills” (Angus & Mirel). The one constant was the assumption that high academic standards were only for the select few (Kliebard, 1987; Ravitch, 2000). Since the middle of the twentieth century those two compelling forces have been held in delicate balance by the powerful compromise of the compre— hensive high school. The Conant Report provided the basic blueprint that could embrace both goals: a single institution that could aim to educate “all the youth living in a town, city, or district” (Conant, 1959, p. 9). The design could accommodate the “horde of heterogeneous students that has descended on our secondary schools” by bringing them together “under one roof,” but not under one curriculum (p. 602). Educators, Conant argued, would have to develop differentiated expectations: an academic track would “maintain high standards” for those of “high ability,” while other courses, in vocational and ‘general stud— ies’ would have “another standard.” To that end, high schools would have to expand in size to sort students efficiently into suitable programs, sift them into ap- propriate tracks, and supply the array of offerings that could meet their diverse needs. Indeed, the strongest recommendation of the report was that “the number of small high schools must be drastically reduced” and replaced by large comprehensive high schools with “a graduating class of at least one hundred” (p. 14). Through the end of the twentieth century, the fundamentals of that design—the function of sort- ing and sifting and the provision of something for everyone and academics for some—remained intact as high schools followed the blueprint far beyond the expectations of its authors. While not all high schools became comprehensive, that quickly became the dominant model—the kind of high schools most students go to, that most test scores come from. They grew in size, far surpassing Conant’s suggested goal for “large” to reach an average of 1,200 students, with many enrolling three or even four thousand students. They grew more and more organizationally complex, dividing and subdividing students by grades and tracks, teachers by departments and programs. The curriculum, too, grew far beyond the projected three tracks to seven, or even twelve sharply differentiated levels (Oakes, 1985; Siskin, 2003). Course offerings expanded to a “cafeteria style curriculum in which the appetizers and desserts can easily be mistaken for the main courses” (Nation at Risk, 1983). The compre- hensive high school, critics charged, had come to re— semble a “shopping mall” where “some shop at Sears, others at Woolworth or Bloomingdale’s” but where not every neighborhood would offer a Bloomingdale’s (Powell et al. 1985; Siskin, 2003). Given the changes in time, in the economy, in public aspirations and policy expectations, that fifty—year solution has become the problem for a new generation of reformers~seen as too large, too impersonal to keep students engaged (or even safe), too disconnected to adequately prepare them for college or workplace entry, and too often offering starkly inequitable opportunities both within and across schools. “Training the workforce of tomorrow with the high schools of today is like trying to teach kids about today’s computers on a 50 year old main— frame. It's the wrong tool for the times” (Bill Gates, 2005 NGA Summit on High Schools). But while the comprehensive design may be the wrong tool for the new century, the fundamentals of the design remain strong, deeply entrenched in the structures and cul— tures of high schools, solidly built into the operating systems of assessment, of curriculum, and of teaching. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Challenges of Assessment “Testing is important. Testing at high school levels will help us to become more competitive as the years go by. Testing in high schools will make sure that our children are employable for the jobs of the 21 st century. Testing will allow teachers to improve their classes. Testing will enable schools to track. Testing will make sure that a diploma is not merely a sign of endur— ance, but the mark of a young person ready to succeed.” —George W. Bush, January 2006 Testing certainly has been an important strategy in the new accountability systems—the primary tool used to push schools toward academic achievement, measured by test scores. Originally grounded in a powerful logic of ‘systemic reform,’ states were chal— lenged to establish clear standards for what all stu- dents should learn, to provide opportunities for them to do so, and to develop assessments to measure their progress (Smith & O’Day, 1991). Almost every state has adopted testing as a cornerstone of accountability, and the federal system has persistently called for more tests for more students at more grade levels—lead— ing teachers to joke that the new policy should be called No Child Left Untested. But testing is at best an imperfect tool for measuring student learning, and a weak lever for effecting change in instructional prac— tice. Those weaknesses have been particularly promi- nent, and problematic, in the effort to use testing to bring about major change in the high school. High school teachers make a sharp distinction between the idea of high standards, which they see as necessary, and the instruments of high stakes testing, which they see as inadequate, inaccurate, and inevitably tied to rising failure rates and increasing numbers of drop- outs. One challenge has been the difficulty of aligning the strategy of testing with the structure of the high school. The comprehensive design, with its large size, compartmentalized subjects and departmentalized faculty makes even delivering information to teach— ers difficult. Designing a testing system that would allow them to improve their classes seems a daunting challenge at best. At the elementary level the logic is clear: a 4‘“ grade math test, for example, should assess Achievement and Attainment: The Comprehensive High School and the Problem of Reform all 4"“ grade students on 4‘” grade standards. Each 4‘h grade teacher can adopt curriculum and adapt teach- ing strategies to align with the 4‘“ grade standards that will be assessed (though whether they should or not is a different question). At the high school level, however, that logic breaks down. A ‘high school’ math test may assess 10‘“ graders, 11‘“ and 12‘“ graders, or advanced 9m graders on a range of math topics from basic arithmetic through algebra to trigonometry. Students in different tracks will have taken algebra at different times, and from different teachers. Most states test math and reading; most high school teach— ers teach neither subject (and in many high schools, no one teaches reading). For teachers committed to the ideals of a comprehensive education that meets multiple purposes through multiple subjects, such tests are simply inadequate to assess what they should be teaching, and what students should be learning. A second challenge is that of accuracy. Even test- ing that tiny fraction of what is taught in high schools adds to the pressure on a testing industry struggling to keep up with increasing demand. There were 45 million tests in 2006, forty—nine different sets of state standards, dozens of different tests for different grade levels and subjects, and only 5 major companies to manage the task. NCLB legislation requires the use of “high-quality academic assessments” aligned with state standards, but thus far, the industry has had difficulty meeting standards of even basic accuracy (Rhoades & Madaus 2003; Toch, 2005; Viadero, 2006). Widespread reports of mistakes in test items, missing supplies, misplaced answer sheets, and mistakes in scoring have been common across states, and across companies: Harcourt mistakenly “failed” 736 students in Nevada, 355 in Connecticut; Pearson mis—scored 8000 tests in Minnesota, 2500 in Alabama. Even ETS, with decades of experience in large-scale administra- tion of high—stakes tests, reported mis—scoring five thousand tests—and then misplacing a thousand more. In Long Beach, the courier lost answer sheets for 400 sophomores—the first year that the CAHSEE scores would be used to determine graduation. At the high school level—where stakes are so high, where diplomas or college admission can depend on a single number, such problems with inaccurate assessment are particularly troubling. For teachers, they under- mine the credibility of the accountability system, and of reformers (Siskin, 2003a). PACE u 95 96 FIGURE 2: Declining High School Enrollment, Los Angeles County District 80,000 ninth tenth eleventh twelfth Source: Los Angeles County District Public Schools, 2002 Many educators, critics, and researchers are left with questions about whether high stakes tests are driving students to drop out (Carnoy et al. 2003; Haney, 2000). But no one knows for sure, because no one has been accountable for keeping accurate count. What happened to the class of 2005? asked a provocative headline in the L.A.Times, leading into its story of how a class on 1100 entering ninth grad— ers had somehow dwindled to only 521 seniors four years later (Landsberg, 2006). As the pressure of high—stakes tests, and the looming possibility of high rates of failure in many schools rises, the question is increasingly being asked. But instead of displacing the pressure toward achievement, as in earlier reform waves, concerns about attainment and graduation rates are arising alongside them, catching high schools (and policymakers) in the vise of dual expectations. Indeed, given the delays of deadlines for when di- plomas depend on test scores, the full force of the achievement effort has yet to hit most states, or most high schools. An accurate count of students as they enter, progress through, and exit schools should be a basic cornerstone in building any accountability system, and an essential component of high school reform. Instead, policymakers and educators rely on estimates from different studies that start from different kinds of data and use different methods of calculation— producing dramatically different—and politically contentious— results. In 2003, for example, the state of California reported an 87% graduation rate; federal figures put the figure at 74.1%; a Manhattan Institute study estimated a rate of only 70%. District figures, as well, can vary considerably depending on who is counting and toward what end, but dramatic declines in the numbers of students progressing from grade to grade suggest many more students are disappearing from the rolls than are appearing in dropout counts. NCLB has recently added a measure of four-year graduation rate to its accountability system, making it imperative for states, districts, and schools to keep more accurate track of students. All fifty governors have now signed the Graduation Rate Compact, pledging to develop a consistent and comprehensive system of tracking students. California has begun to develop its Longitudinal Pupil Achievement Data System, with voluntary participation of districts ex- panding to include data state—wide, but those systems are still many years and millions of dollars away from providing data that are useful to schools. Challenges of Curriculum “Maybe if they spent all day doing only Eng— lish and writing and didn’t do any math or science. But they have other things” —Teacher Keeping track of students and keeping them in school is one critical component of high school reform that data systems can help to address; keeping them on track academically to meet testing demands and college requirements involves challenges of cur- riculum capacity and development. Accountability reform is generally framed as a shift from inputs to outcomes, from counting seat time to rewarding achievement, but for comprehensive high schools that have long measured their success by how many ‘other things’ they offered, the shift to a more narrowly aca- demic focus is profound. The dilemma of how much time they should spend focused on tested subjects that prepare them for achievement ends, and how much on the ‘other things’ that engage their interests and keep them attending school has been challenging for many schools. Many schools, particularly high-needs high schools, simply lack the capacity to provide sufficient academic courses for their students. In recent stud— ies of national curricular pathways, 94% of students CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 say they plan on going to college, but only 49% are enrolled in the college prep or AP courses that would take them there (Alger Assoc, 2006). This is not be- cause students are unmotivated: 74% of minority girls report wanting to take advanced math, but only 45% have access to such classes in their schools (Achieve, 2006). Education Trust-West has developed a “Curric— ulum Opportunity Index” to assess California’s capac- ity to provide the “A-G” courses required for college: of the 371 high schools they selected, fewer than 20% of could offer the full sequence to all their students. In 19 schools, fewer than half of the students could have that opportunity (www.edtrustwest.org). States and schools are working to increase aca- demic offerings and raise academic achievement. In the new American Diploma Project, 22 states have now pledged to align “curriculum, standards, as— sessment and accountability policies with demands of college and work” (Achieve, 2006). In 1992, only six states required three science courses to gradu- ate; by 2011, 27 will. Schools and districts, too, are moving to ‘raise the bar,’ narrowing the ‘shopping mall’ curriculum as prescribed, at least on paper (though there is certainly evidence in some cases of nominal compliance—where titles have changed but what is taught, and how it is taught, remain largely untouched) (Loveless, 1999; Siskin, 2003b). Overall, however, there does seem to be a general effort to shift “from warehousing and managing student be— havior to focusing on serious student learning” (Hess & Cytyrnbaum, 2001). There are more algebra classes, and fewer offerings like “Pre—Spanish, Teen Living, Food Fundamentals, and 100 Winter Activities” that are now 90 being phased out in California 80 high schools (Education Trust, 2005). In San Jose, district officials 70 report remarkable success follow— ‘E’ 60 ing their decision to make the A-G § 50 sequence the default curriculum Cal) 40 for all students (Murray, 2005). AP courses, too, are being offered 50 at a rapidly increasing rate, rising 20 nationally from 1,220 tests in 104 10 schools in 1,956 to more than 2 0 million tests in 2004, taken by one million students in 11,000 public African American schools (College Board, 2005). California, too, reflects that shift, with the number of 11‘h and 12m graders taking exams up from 10% in 1997 to 15% in 2004. Those numbers are slightly above the national average (12%) but somewhat unevenly distributed—30% of Asian students took AP exams that year; only 6% of African American students did. One unanticipated consequence of that shift is that many schools, particularly those schools most under achievement score pressure, are not only rais- ing the floor of course offerings but lowering the ceiling. Here students are not only losing courses like cooking with chemistry but also ‘other things’ like marine biology and music, or what have been identified as 215‘ century skills” like global awareness, critical thinking and problem solving (Siskin, 2003b). Moreover, students who come in to high school with low achievement scores can lose the most—like those in a Texas high school who found themselves placed in math, remedial math, and test prep math (Carnoy et al. 2003). Such schedules leave no time for the electives that college admission committees seek and employers demand, and it is hard to imagine a less enticing schedule to keep anyone in school. On the other hand, many large low—performing high schools and new small schools have redesigned structures and curriculum in ways that show consid— erable power in retaining students, in keeping them engaged in learning, in maintaining their aspirations toward college, in raising graduation rates and college FIGURE 3: California Graduation Rates, Estimates for Class of 2003 White Asian Latino Source: Manhattan Institute (2006) PACE 15% 97 98 admissions. But those changes in structure have not ensured changes in instruction, and do not necessar- ily raise academic achievement—at least achievement measured by test scores—as many of the Gates- funded small schools have learned (Kahne et al., 2006; Carnoy et al., 2003). The one area where accountability pressures are expanding rather than constricting curriculum in the high schools is in reading—the subject most commonly tested and yet least commonly taught in high schools. Researchers estimate the number of adolescents who struggle with reading at approxi- mately 6 million (Ioftus, 2002); on NAEP tests, 64% of high school students performed below proficient and more than 25% below basic. As the dual de- mands of the new accountability require schools to keep these students from dropping out, and to pre- pare them to pass high stakes tests, literacy is proba- bly the most pressing and unmet need in high school curriculum. But there are few specialists, and few programs with successful strategies for adolescent literacy—and fewer still that can make the necessary connection to academic literacy—the skills required to read high school science or engage history texts (with the Strategic Literacy Initiative in California providing an important and promising exception) (Schoenbach et al., 2000). But without that connec— tion, teaching reading will remain what it has been: at best, the job of an isolated specialist with margin— alized students, at worst, no one’s job. FIGURE 4: AP Exam Participation in California Percent 100 90 80 7O 6O 50 4O 30 20 ’lO 0 NaUve American African American Asian Latino Source: American Diploma Project, 2005 Challenges of Teaching: Distribution and Development “Butjust because you demand that [fly a plane doesn’t mean that I’m going to be able to do it. So just to askfor something, is not a guarantee that you’re going to get better per- formance out of the person. You mightget a nervous breakdown.” —High School teacher, New York The conflicting demands of achievement and as— piration cannot be resolved by curricular change—of whatever kind—alone. Simply placing all students on the academic track, as in Los Angeles’ recent re- quirement that all students take algebra, just “triggers dropouts more than any single subject,” concluded the superintendent. “I think it is a cumulative failure of our ability to teach math adequately in the public school system” (LA Times, Ianuary, 2006). However, even as demands for performance have been escalat- ing, investments in teaching have lagged far behind the imposition of testing and sanctions. That leaves many teachers feeling helpless, asked to do something they do not know how to do, and have never seen done—something many have a hard time believing can be done at all. But under the new accountability requirements, they now have to fly the plane. To guarantee that they can, or even to make it more likely, involves challenges in the distribution of qualified teachers, and in their ongo— ing professional development. Across the country, high schools, particularly high-needs high schools, are strug— gling to recruit and retain the ‘highly qualified’ teachers that the federal accountability system now requires for all students. While there is much debate over the precise definition of‘highly qualified,’ a recent special report by Education Trust—West (2006) found an “inequitable distribution of teachers in California by every mea— surable proxy—including experience, education level, credential status, and salary.” They further found that those inequities were systematically distributed: the odds of being taught White CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 by ‘highly qualified’ teachers were lowest in high-pov- erty and high—minority schools, and in high schools. Other studies consistently find the same patterns: 44 percent of math classes in California’s high—poverty high schools are taught by teachers not certified in math (lngersoll & Education Trust—West, 2005); stu- dents in high minority schools are five times more likely to have an under—prepared teacher (Esch et al, 2004; see also Yun 8( Moreno, 2006; Rumberger & Gandara, 2004). Moreover, those inequities in teacher qualifications are directly linked to performance on the CAHSEE: in schools with the lowest passing rates, students are three times more likely to have an under— prepared teacher (Center for the Future of Teaching and Learning, 2005). Developing structures and incentives to bring qualified teachers into schools is just a first step. The next is to provide ongoing and substantive profession— al development to the teachers who are already there. Strong professional development communities (in high schools typically strong departments) can con- tribute not only to the learning of teachers, but also to the achievement of their students—including their test score performance (Little, 2002; McLaughlin 8t Talbert, 2006; Siskin & Little, 1995). Learning to teach all students to reach high standards is neither obvious nor easy; in fact, it runs counter to what many teach— ers were themselves taught about academic achieve- ment and bell curves, which is why even experienced and certified teachers can worry that they will not be able to deliver what is being asked of them—that the new demands might just produce a “nervous break— down” instead. To counter that fear, reform advocates urge that “every high school [become] a learning community for teachers” (NASSP, 1996, p. 63), that “teachers must be given more time to collaborate with one another to improve teaching and learning.” (Na- tional Governors Association, 2005, p. 10). That means time to meet together and collabo- rate with subject matter colleagues, for when high school teachers talk productively about instruction, they speak the language of their subjects. Yet while most teachers are in school, they have little time for reflection on teaching, analysis of student learning, or collaborative conversation about just how they will go about getting all students to reach high standards (Siskin, 2003b): Achievement and Attainment: The Comprehensive High School and the Problem of Reform I only 1 in 5 teachers say they “regularly meet to share ideas about lesson plans and methods of instruction” (Public Agenda, 2002) I only 30% of teachers participated in professional development that involved in-depth study in a their content area, I less than half report having a mentor to turn to in their subject area (NCES, 2000) This, like the distribution challenge, is particu— larly important in the low-performing high schools that are the primary target of the accountability movement, where many students are 4, 5, or even 9 years behind to start with, and where the demands seem particularly difficult to meet. As a Washington DC principal in just such a school explained: “We have a faculty that is somewhat needy on strategies on how to reach these students.” When teachers are asked to reach all students, to keep them engaged in school for four years, and to ensure that they are prepared to pass a high—stakes tests, every teacher becomes some- what needy. The Problem of Reform To meet the needs of those teachers and their students, high school reform itself needs to come to grips with its historically dual nature: the two long» conflicting demands of high achievement and inclu- sive attainment. Left unresolved, those demands have left high schools stubbornly the same. Accountability has asked them to meet both of these demands at the same time. But the structure of the high school—the ways that its fundamental function of sorting and sifting is built into every aspect of its form—stands in the way. The problem of reform, then, the urgent challenge, is to address all of those aspects—aspects of design, assessment, curriculum, and teaching— together. PACE K 99 102 CRUCIALISSUESIN CALIFORNIA EDUCATION 2006 Crucial Issues in Prepari- of Englis " .. ALIFORNIA HAS A HIGHER CONCENTRATION OF ENGLISH LEARNERS 'I‘HAN ANYWHERE ELSE IN THE US. THESE STUDENTS HAVE DIVERSE SOCIAL. ACADEMIC, LINGUISTIC. AND CULTURAL EACKGRDUNDS AND A WlDE RANGE OI: EDUCATIONAL NEE D8. Every day, in the great majority of California classrooms, teachers are called upon to meet the challenge of helping these students gain the breadth and depth of English language skills they need to learn the academic content that will allow them to succeed in school and beyond. Although improving how we educate California’s English language learners is a complex issue, teachers are central to the learning equation. California cannot meet its full potential until its teachers have the knowledge and skills to help these students meet theirs. .‘tudent Demographics According to the National Center for Education Statistics, in 2002, of the 3.8 million English language learners in US. public schools who were receiving EL services,1 1.5 million or 39% were in California. The state with the next greatest concentration of EL students was Texas, with about 16%, followed by Florida and New York with approximately 5% each. Nationwide, the number of limited English proficient students has been increasing rapidly—growing from approximately 2.2 to 4.4 million between 1990— 2000.2 This English learner population is becoming more dispersed, and the growth in the number of EL students has increased dramatically in some of the states that have historically enrolled few such stu— dents. Nonetheless, the five states with the largest EL populations—California, Texas, New York, Florida, and Illinois—continue to account for the largest in— crease in numbers and together were responsible for 51% of the total EL growth between 1990 and 2000 (Cortez, 2003). California made up the lion’s share of the in- crease, gaining approximately 600,000 EL students during that time period.j Over the last two decades, English learners in California public schools increased four—fold and currently represent 1.6 million of the state’s nearly 6 million students. Given the youthful— ness of this population, these students will be in Cali- fornia schools for a long time. The majority of English learners (69%) are enrolled in kindergarten through grade 6, with 31% enrolled in grades 7—12. Notwith— standing the greater numbers of K—6 English learners, the EL student population is growing most rapidly in secondary schools,4 where we have paid the least atten— tion to their needs, and where we have the least knowl- edge and instructional expertise in teaching them. 103 California has approximately one million addi- tional students who are considered English proficient, but who come from homes where English is not the primary language spoken. The educational implica- tions for these students are seldom discussed or con- sidered. Nonetheless, students who speak a language other than English at home, referred to as “language minority students”5 account for 43% of California’s K—12 school population" Although California’s EL students speak many languages, the overwhelming majority speaks Span- ish as a first language. Moreover, Spanish speakers are somewhat more concentrated in California than in the rest of the country. While 71% of all English learners in the US. speak Spanish,7 85% of EL stu— dents in California are Spanish speakers. The next largest group, Vietnamese speakers, makes up only 2.2% of the state’s English learners, followed by Hmong, Cantonese, Pilipino (Filipino or Tagalog), and Korean, each with only 1% of the state’s EL stu— dents. Together these groups account for more than 90% of California’s English learners, and no other group in the state exceeds 1% of the EL population.8 In addition, one quarter of the state’s schools have more than 10 students from two or more non-English language groups. While this great linguistic diversity among California’s students can complicate instruc— tional strategies, fortunately in a majority of cases (75%) teachers are called upon to address the needs of only one non—English language group (Figure 1). Academic Achievement of English Learners Among non-disabled students, English learners are the lowest performing of all students in Califor- nia public schools. To some extent this is an artifact of testing these students in a language they do not understand. Nonetheless, on all measures we have of student performance, even on largely non-verbal tests, EL students trail their English-speaking peers. This educational discrepancy begins early. Data collected from California students for the national Early Child— hood Longitudinal Study shows that students from non-English backgrounds begin school with far less developed cognitive skills than their English-speaking peers (Figure 2). Moreover, achievement data for older students indicate that the discrepancies persist through the grades. Figure 3 shows the steep decline in the number of EL students who are able to achieve pro- ficiency on the California Standards Test (CST) of English Language Arts (ELA)—~a test of the academic English skills necessary for success in all school sub— jects—as students progress through the grades. At second grade, a somewhat surprising 22% of stu- dents who by definition do not speak English with proficiency are nonetheless able to achieve a score of proficient or above (level 4 or 5) on the CST—ELA exam. However, as the demands of the test become greater, students’ performance falls precipitously. By 10th grade only 3% of EL students are able to meet the standard of proficiency or above on this exam. FIGURE 1: Percent of California Schools and EL Enrollment, by Number of Language Groups with Ten or More Students, 2005 70 60 50 40— 50‘ Percent of Total 20— 10— 104 Source: Rumberger, LMRI Newsletter, Spring 2006, www.|mri.ucsb.edu O ’I 2 5 (2152 schools) (4987 schools) (1504 schools) (554 schools) (257 schools) 4 5+ (185 schools) Number of Language Groups with 10 or More Students students CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 In striking contrast, 64% of 10‘h grade EL students scored as early advanced or advanced (level 4 or 5) on the California English Language Development Test (CELDT)—a test designed to evaluate the English skills of English language learners (Figure 4). This large discrepancy, particularly in the secondary grades (Figure 4), between how students achieve on the CST FIGURE 2: Cognitive Skills of California Beginning Kindergartners, by Language Background, Fall 1998 60 50" 40" Percent above 50th percentile General Knowledge I English - Non-English Note: Results are weighted (C1 CWO). Language Source: ECLS base year data for California public school kindergarteners (N=2826) and the CELDT clearly indicates that the latter test is not an adequate measure of the academic English skills that students need to succeed in school. The low level of correlation between students’ scores on these two exams should provide a caveat for educators that proficiency on the CELDT is not a reliable indicator of students’ ability to master academic work in English. Even in the area of mathematics, which is gener— ally considered to be less language dependent, English learners do not perform well. EL students’ math test scores mirror the pattern seen in language arts: as the complexity of the material increases, students’ scores decline (Figure 5). It comes as no surprise, then, that fully half of English learners from the graduating class of 2006 were unable to pass the California High School Exit Exam (CAHSEE) in both English and math by the end of the 11th grade.9 This was true even for many students who had taken and passed all required courses (Rogers et a1., 2005; EdSource, 2006). The already extraordinarily high drop—out rate for English learners (Rumberger, 2004) will almost certainly be exacerbated by this fact. In sum, the low achievement of EL students is reason for grave concern as they come to comprise a larger and larger share of the public school population. FIGURE 3: Percent English Only and English Learner Students Scoring Proficient or Above on California Standards Test, English Language Arts, 2005 7O 60 I E0 50‘ l EL 40“ 30— Source: California Department of Percent 20— Education Grade Level PACE 5:322 105 106 FIGURE 4: Comparison of English Learners Scoring Level 4 or 5 on CELDT and CST English Language Arts, 2005 Percent Early Advanced or Advanced 70 60 50 4O 30 20— 10— Grade Level 8 ’IO I Advanded or Earlv Advanced on California English Language Development Test Proficient or Advanced on California Standards Test English Language Arts Source: California Department of Education FIGURE 5: Percent of EO and EL Students Scoring Proficient or Above on 2005 CST Math Percent Proficient or Above 7O 60— 50— 40— 50— 20— ”IO— - Percent EO Percent EL Source: California Depart- ment of Education, 2005 5 Grade Level Classroom Programs for English Learners in California Schools California schools take several educational ap— proaches to teaching EL students. However, no matter what the approach, federal and state law require that EL students receive educational services designed to meet their particular learning needs. The two-fold goal of these services is that EL students receive in— struction that 1) teaches them the English language and 2) allows them to understand the classroom curriculum.10 The principle educational approach to meeting the first goal is through English language development (ELD) instruction, often via content in— struction in English as well as through explicit English language lessons designed for English learners. To meet the second goal, teachers can facilitate student understanding of the core academic curriculum using a number of specialized English language teaching strategies (typically referred to as SDAIE or Specially Designed Academic Instruction in English), using the students’ primary language, or a combination of these. The majority of California’s English learners receive instruction through ELD, SDAIE, or both (Gandara, 2000; Gandara et al., 1999; American In— stitutes of Research, 2006) in either mainstream or structured English immersion classrooms. In general, mainstream classrooms include EL students who have intermediate or advanced proficiency in English CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 together with students who are native English speak- ers. Structured English immersion (SE1), on the other hand, is intended as a one- or two-year program for EL students who have beginning or early intermedi- ate English language proficiency. SEI classrooms do not include non—EL students.” However, many class- rooms include students at both levels and incorporate both programs in the same classroom. A very small percentage of the state’s students, approximately 8%, are enrolled in bilingual programs, which include content instruction in the students’ primary language and/or SDAIE, as well as English language develop— ment instruction (Figure 6). Because of the practice of placing students in the same classroom who are in different programs, and because there is variation among schools with regard to program nomenclature—one school’s SE1 program might be another’s bilingual program—it is much more instructive to consider English learner educa- tion from the perspective of the services students receive than the program in which they are placed. English learners need and are entitled to English lan- guage development, as well as instruction designed to ensure that they have access to the core academic curriculum (SDAIE), no matter what the name of the program in which they are enrolled. Figure 7 illus- trates the education services received by California’s English learners. FIGURE 6: Instructional Programs in which English Learners Receive Educational Services Other (4%) Bilingual Instruction (8%) Mainstream Classroom (41%) Structured English Immersion (47%) Source: California Department of Education Language Census, 2005 In our survey of California teachers of English learners, we asked teachers how their EL students receive these educational services (Gandara, Max- well—lolly, and Driscoll, 2005). We learned that the EL students of more than half (55%) of the over 5,000 teachers we sampled are removed from their regular classrooms to participate in “pullout” instruction from a resource teacher or para- professional for some part of the school day. This is often because the regular classroom teacher is not adequately pre- pared to provide EL students the instruction they need. The Providing adequate numbers of teachers with the skills they need to work with English learners practice was even more mew is a critical part of. . . lent among teachers in smaller districts and those with fewer improving achievement. EL students. This is of concern because research on pullout instruction places it among the least successful strategies for teaching EL students, and experts in the field do not generally recommend it. Reasons include students’ lost opportunities to learn what their classmates are exposed to while they are out of the class, instruction that is inconsistent with that given to students who remain in the class- room, and valuable time lost in transitions (Lucas et al., 1990; Ovando and Collier, 1998). FIGURE 7: Educational Services Received by English Learners No Services (2%) ELD & Academic instruction in L1 (7%) / Other instructional Services (8%) ELD & SDAIE (51%) ELD 0an (11%) ELD & SDAIE ’ With L1 Support (21%) Source: California Department of Education Language Census, 2005 PACE 2% 107 110 Looking behind the numbers California’s teacher work force is not as prepared for teaching English learners as these numbers might suggest. Although teachers with CLAD training have many skills for teaching EL students— certainly more than teachers who have not had such train- ing— by their own account many feel unprepared to teach English Learners (Gandara, Maxwell—Jolly, and Driscoll, 2005). In a 1999—2000 survey of almost 11,000 graduates of teacher credential programs in the California State University system, one-fourth responded that they felt they were either “somewhat prepared” or “not at all prepared” to teach English learners (Office of the Chancellor, 2003). Further evidence that new teachers often do not feel prepared, is illustrated by an analysis of data from the statewide professional assignment report con— ducted by the Center for the Future of Teachers and Learning. The Center’s researchers found that almost two—thirds of the new teachers who earned their cre— dentials after content related to EL instruction was embedded in the teacher education curriculum, were not even aware that they were authorized to teach English learners. “Despite the state’s efforts to embed EL training into every new teacher’s preparation pro— gram, there is some evidence that newly credentialed teachers are unaware or unaffected by this training. In 2004—2005 only 34% of first— and second—year teach- ers reported to CDE that they are EL certified” (Esch et a1, 2005). In addition, among those with EL preparation are the 26,552 teachers authorized to teach EL students per Senate Bill 1969/395, but who have received con— siderably less training in EL methods than have CLAD teachers. Experienced teachers can earn this autho- rization by completing only 45 hours of staff devel- opment. Teachers with less experience can earn the certification after 90 hours, and are often authorized to teach either ELD or SDAIE, but not both. On the other hand, to earn CLAD certification, teachers must take 12 semester or 18 quarter units of upper division or graduate coursework—a far higher standard, but arguably still insufficient. California schools have far fewer BCLAD than CLAD teachers. There is only one BCLAD creden— tialed teacher for every 234 English learners. BCLAD teachers have more skills for working with EL stu- dents, are versed in a wider array of instructional methods than CLAD teachers, and can communicate with parents and students in their own language. Moreover, in our study of 5,300 educators of English learners in California (Gandara, Maxwell-Jolly, and Driscoll, 2005) not only did BCLAD teachers report significantly greater confidence in their own ability to teach EL students, but those non-BCLAD teachers who did not have the ability to communicate with parents and students in their home language cited this as one of their greatest challenges. We also heard from teachers, especially in small districts, about how bilingual teachers serve as the “resident experts,” often at a significant cost to their own classrooms and their personal and professional time. Because of the way data on teachers is collected in California, it is not always possible to know the capacity in which teachers with specific credentials are serving. Although we know how many teaching credentials are issued, we cannot now tell how many teachers with CLAD and BCLAD equivalent creden- tials currently work in our classrooms. Teacher data is collected by program type, rather than by credential designation. For example, a BCLAD credentialed teacher who has bilingual skills and is teaching in an all—English program (such as “mainstream” or “Struc— tured English Immersion”) is counted as a teacher with certification for using English methods—but not as a bilingual teacher. Policymakers, schools, and school districts need more accurate data if California is to use its teacher resources effectively and to under— stand how teachers’ knowledge and skills are being employed to the best advantage for students. What Teachers Say The evidence is clear that teachers’ knowledge and skill are key to student learning, but that many teachers in California are inadequately prepared for the job they are being asked to do. In 2004 we con- ducted a survey of teachers of English learners to ascertain how they viewed the challenges of teaching these students and what kinds of supports they felt they needed to do so effectively (Gandara, Maxwell— ]olly, and Driscoll, 2005). A fundamental premise of the study was that it is impossible to provide teachers with practical guidance on how best to improve their EL teaching skills without knowing what they con— sider to be their greatest challenges and needs. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 What we learned reinforces the notion that it is essential to provide California teachers with more and better preparation for working with English learn- ers. In general, we found that although teachers ex- pressed a wide range of concerns and placed different emphases on those concerns depending upon grade level taught, overall the challenges centered on a few key issues. Among the most frequently cited teacher challenges that might be alleviated through greater teacher preparation were 1) teachers’ inability to com— municate with students and their families, and 2) a lack of skills and tools necessary to meet the widely variable needs of the students in their classrooms. Teachers working in grades K—6 placed the most emphasis on their need to communicate with fami— lies, while teachers in grades 7—12 cited as their great— est challenge communicating with students, often in order to motivate and encourage them. If “chal- lenging without discouraging” students is difficult when teachers and students speak the same language, consider how much more difficult this is when they do not. Other work that we have done (Gandara et a1, 2003; Gibson, Gandara, and Koyama, 2004) has given us insight into the reasons why secondary teachers may find motivating these students such a thorny problem. These include an impoverished curriculum, limited access to the resources of the schools, and the isolation that many English learners feel in American high schools. High drop—out rates also reflect the extent to which these students recognize the limita— tions of their high school curriculum. Even students who strongly value schooling can be discouraged when they cannot follow what is going on in the class- room. Having the necessary instructional skills and tools to address widely variable student needs—both among EL students and between ELs and fluent Eng- lish speakers—was also a critical concern of teachers. Specifically, teachers sought guidance regarding how to apply several commonly used curricular packages realistically and successfully with their EL students. As California policy increasingly favors mainstreaming English learners, classroom teachers receive greater numbers of students at all grade levels with widely varying skills in English and in academic subjects. For example, in the same classroom a teacher might have orally proficient EL students who need to make signif- icant gains in their academic English skills, students CRUCIAL ISSUES iN PREPARlNG TEACHERS OF ENGLISH LEARNERS who have just entered the country and have little or no English skills, native and non-native English speakers who have superior academic preparation, and students with little formal education. Other challenges for teachers working with EL students related to the teachers’ level of preparation. For example, teachers said they had too little time to meet their students’ needs, and teachers with special- ized skills were better able to use classroom time ef- fectively than those lacking such skills. Teachers also mentioned the challenge of determining what their students know and don’t know in order to provide the best instruction. Although this has much to do with the lack of appropriate assessment instruments for EL and bilingual students, teachers can learn skills—in— cluding the primary language of their students that allow them to better diagnose their EL students’ learning needs. Teachers’ Self-Rated Ability in English Learner Instruction The importance of adequate teacher preparation to EL students’ success is underscored by research on the connection between teachers’ effectiveness and their belief in their own ability (self efficacy), and by findings regarding the relationship between teacher preparation and self-efficacy. There is a significant body of literature on the positive relationship be- tween instructional effectiveness and self-efficacy (Armor et al., 1976; Rosenholtz and Simpson, 1984; Woolfolk et al., 1990; Goddard et al., 2000; Goddard, 2001), and on self—efficacy, optimism, and the will to create change (Bandura, 1993, 1995; Farber, 1991). This led us to ask teachers to rate their ability to teach English learners effectively. We found that the more extensive the preparation of the teachers, the more confident they felt about their ability to teach English learners successfully in all subject areas.16 Furthermore, teachers who had certification designed to prepare them in the primary language of their stu- dents felt the most prepared to teach in that language. In addition, teachers at both the elementary and sec— ondary level who had received professional develop— ment related to the teaching of English learners rated themselves significantly more able to teach these stu— dents across all categories of instruction than teachers without such training. PACE $1” 111 112 FIGURE 8: Most Useful Type of EL-focused ln—Service Completed by Classroom Teachers 40 35 Elementary Source: Ga’ndara, Maxwell—Jolly, and Driscoll (2005) Teacher Participation in Professional Development Teacher preparation is critical for providing new teachers with the skills and knowledge to begin their professional teaching careers. However, formal learning should not stop there. Teachers, like other professionals, need ongoing professional develop— ment opportunities to extend and improve their job skills. This is even more critical in areas such as English learner education, because current credential programs generally do not provide teachers adequate preparation to work with EL students. In addition, while in the past many teachers did not have EL stu— dents in their classrooms, today many now find them— selves having to teach EL students. Despite what is known about the importance of teacher knowledge and skills to EL achievement, many teachers have received little professional de— velopment on how to work with English learners. For example, according to our survey results, 43% of teachers with 50% or more English learners in their classrooms had received no more than one in—service workshop that focused on the instruction of English Learners (and often none at all) during the previ— ous five years. Moreover, while this one in—service consisted of multiple hours for some teachers, for others it was limited to a single after or before school workshop and ranged, according to teacher opinion, from helpful to not at all useful. For those teachers whose classrooms had 25—50% English learners, half had received no more than one in—service designed to help them address the learning needs of these stu— dents. Furthermore, current law requires that during teachers’ first years of employment they receive pro- CRUCIAL ISSUES Secondary Linguistic Theory ELD/ESL Methods Info on 2nd Language Learning and learners Cultures of ELS Instructional Methods for academic subjects EDD- fessional development in order to augment what they learned in their credential program, and that some of this professional development focus on teaching EL students. Yet, we found that only half of the brand new teachers in the sample had done so. Given earlier studies indicating that teachers in California, on average, received only about two hours of in—service training annually that was focused on English learner issues (Gandara et al, 2003), the find— ings of this study are not surprising. It is, however, disconcerting. Teaching English learners is a compli— cated challenge that requires significant skill. Even the best teacher pre—service programs cannot cover all the knowledge and skills required of an effective EL teacher. Teachers need appropriate professional devel— opment opportunities to gain the skills necessary for teaching EL students within their particular school and classroom context. Teachers’ Views Regarding Usefulness of Professional Development Experiences Across the board, teachers in our survey most often cited purposeful professional development as having the greatest impact on their ability to teach Eng- lish learners. Specifically, they pointed to professional development that focused on 1) how to teach a second language, and 2) the learning, developmental, and other factors unique to second language learners. With regard to all other professional development subjects they considered most useful, responses varied substan— tially by elementary and secondary teaching level. Teachers also said that they learned the most through certain approaches to professional IN CALIFORNIA EDUCATION 2006 development. These included observing other more expert teachers working with EL students, partici— pating in a mentoring relationship with a more knowledgeable teacher, and working with other teachers at their own school in order to address problems across grades and subjects rather than in isolation. SUMMARY There is a critical need for more teachers with the appropriate professional preparation to work with English learners. Not only do the state’s demographics make this essential—25% of public school students are ELs—but so does current state policy, which man— dates that an increasing percentage of EL students be placed in mainstream classrooms. Estimates are that between 75—85% of all teachers have at least one Eng- lish learner in their classrooms (Esch et al, 2005). We must keep in mind that although a large number of teachers in California have some kind of official certification to teach English learners, few are well-prepared to do so. We should also consider that the “paper” figures for teachers who are prepared to work with EL students can be misleading. Caution in interpreting teacher readiness is in order, first because very few of these teachers have the most extensive preparation (BCLAD certification) to be able to com— municate effectively with students and their families, and these few are stretched too thin. Second, the other types of credentials, although generally considered equal for hiring purposes, vary in their rigor and in the amount of teacher preparation they provide. And third, teachers themselves say that even their CLAD certification, although helpful, does not provide them with many of the skills they need—and could learn— to work with their EL students. Moreover, the current approach to preparing all teachers to work with EL students, although well intentioned, is not adequate. This is due in part to a lack of teacher education fac- ulty across teacher credentialing institutions who have the expertise in teaching EL students. . CRUCIAL ISSUES IN PREPARING . TEACHERS OF ENGLISH LEARNERS POLICY IMPLICATIONS There are many potential ways to provide teachers with the skills and knowledge they need to teach EL students effectively. Based on the findings presented above, we offer the following recommendations: Districts should make structural changes to employ BCLAD teachers more efficiently. We recommend that districts find ways for BCLAD teachers to share their expertise with school/district students, parents, and their col— leagues, while meeting classroom obligations. Districts should also create more opportuni— ties for BCLAD teachers’ professional growth. A study should be conducted to determine effective strategies for districts to recruit and retain BCLAD teachers. Such a study should include exploration of what incentives and rewards for BCLAD teachers might encourage them to take on this extra work, what factors might attract more such teachers to a district, what might stimu— late more local college students to go into bi— lingual education (particularly those from the home culture of EL students), and what kinds of support would keep more new teachers with bilingual skills from leaving the field. A review should be conducted of the teacher credential curriculum that addresses EL instructional skills and, where necessary, improved an augmented. We recommend a review of the content of instruction currently embedded in multiple and single subject credential programs that is designed to prepare teachers to work with EL students. In addition, we recom— mend greater support and guidance for California’s teacher education programs in PACE $2 113 building their staffs to teach these skills to California’s prospective classroom teachers. Induction programs for new teachers should focus more on EL instructional skills. California currently has an extensive teacher induction program that provides new teach- ers with professional development and mentoring opportunities. Furthermore, all credential programs now prepare teachers to work with English learners, and part of the law requiring such preparation stipulates that new teacher induction include a focus on teaching English language learners. We recommend fulfillment of this requirement and that for all teachers without a specialist credential whose classrooms include English learners, this induction include a significant focus on teaching EL students. Professional development, in general, should focus more on EL teaching. We recommend that for all teachers without a specialist credential whose classrooms include English learners, this professional develop— ment should include a meaningful focus on teaching EL students. While the above recommendations could be implemented and begin to show results rather quickly, we also recommend longer—term strategies for increasing the numbers of EL teachers and the level of their preparation for EL instruction. To accomplish these goals, we recommend the following: Concentrate teacher recruitment efforts at the community colleges to find and train new teachers who have special knowledge of minority communities and languages, and have demonstrated an interest in teaching. Currently, we lose most of these students before they complete a bachelor’s degree (Kirst et al, 2004). Reasons for this loss include financial pressures, inadequate career counseling, and the lack of a focused academ- ic goal. Therefore, we recommend state and federal policies that foster teacher preparation programs beginning at the community col- leges. Such programs would offer EL-focused coursework and counseling, and participants would receive forgivable loans for educational expenses. At the completion of their commu- nity college coursework, participants in this program would receive a specialized Associate of Arts (AA) degree that would allow them to work as classroom aides. This would enable these students to continue their studies at a four—year college, earn sufficient income so they can forgo other work and instead focus on their teacher preparation studies, and gain valuable teaching experience. Similar support should be available for students already in the college pipeline who have special knowledge of minority communities and languages and have demonstrated an interest in teaching. They should be eligible for forgivable loans sufficient to ensure that they complete their undergraduate degrees and credentials in a timely manner, and quickly enter the teaching force. While, forgivable loan opportunities already exist, we suggest that to be maximally efficient such programs should provide stu- dents enough support to allow them to focus solely on their teacher preparation studies. Strengthen the infrastructure that support teacher educators qualified to train current and future teachers to work with EL students. As part of this effort, we recommend that schools of education, county departments of education, and other teacher preparation and inservice programs find ways to increase their capacity to prepare new and continuing teachers to teach English language learners effectively. As a means to this end, we recom— mend that teacher preparation institutions, as CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 0 I‘d well as other inservice providers, take greater advantage of the Mexican government’s program that provides teacher trainers from Mexican universities to come to the US for a period of time to help prepare California’s teacher education professionals to work with EL students. ENDNOTES Under federal law, all students classified as English learn— ers must receive services. Although states define and deliver these services differently, and inevitably some stu- dents do not receive services despite legal requirements, the majority of ELs are accounted for in these numbers. Data available at nces.ed.gov. Data available at cde.ca.gov. California Department of Education Language Census, 2005. _ These can be IFEP students, those who were initially identified as being fluent English proficient upon entry to school in California, or RFEP students, those who began school as English learners but have since become fluent English proficient. California Department of Education Language Census, 2005. Data from the National Center for Education Statistics at nces.ed.gov. California Department of Education Language Census, 2005. CAHSEE test data available at http://datal .cde.ca.gov/ dataquestl. These goals are delineated in federal law resulting from landmark court cases Lau v. Nichols 414 US. 563 (1974); Castaneda v. Pickard (5th Cir. 1981). Non—EL students might be those whose first language is English and who do not speak another language at home, students who were bilingual and already proficient when they entered school, and students who began school as English learners but have met the academic and English language criteria to be redesignated as fluent English pro— ficient (FEP). LAUSD divides its Structured English Immersion classes into two types: Model A, which is English only, and Model B, which allows some primary language support. Data are more difficult to interpret for Model A because cell sizes are smaller and the authors report a lack of con- fidence in these small numbers. CRUCIAL ISSUES lN PREPARING TEACHERS OF ENGLISH LEARNERS '3 SB 1969 was originally passed in 1994 and renewed as SB 395 in 1999. Information on current and past legislation can be found online at www.sen.ca.gov. & Data obtained via personal communication with CBEDS analysts, February 16, 2006. U! Included in this number are teachers who earned an ear- lier version of the credential, the bilingual certificate of competence. c» We asked teachers to rate their ability to teach English learners in the following content areas: English reading, English writing, math, science, social studies, English language development, and primary language reading and writing. REFERENCES American Institutes for Research and WestEd (2006). Eflects of the implementation of Proposition 227 on the education ofEnglish Learners, K—12: Findingsfrom afive—year evalu- ation. Online at http://www.wested.org/cs/we/view/rs/804 Armor, D., et al., 8r Rand Corp. (1976). Analysis of the school preferred reading program in selected Los Angeles minority schools (N0. R-2007—LAUSD). Santa Monica, CA: Rand Corporation. Bandura, A. (1993). Perceived self-efficacy in cognitive de- velopment and functioning. Educational Psychologist, 28, pp.1 17— 149. Bandura, A. (1995). Exercise of personal and collective ef— ficacy in changing societies. In A. Bandura (Ed.), Self —efficacy in changing societies. (pp. 1-45). New York: Cambridge University Press. California Department of Education (2005). Language census data. Sacramento, CA. California Department of Education. Civil Rights Project (2005). Confronting the graduation rate crisis in California. Cambridge, MA: Harvard University. Online at www.civilrightsproject.harvard.edu Cortez, A. (2003). The emerging majority: The growth of the Latino population and Latino student enrollments. IDRA Newsletter www.idra.0rg/ newslttr/ 2003/ Ian Darling—Hammond, L., Holtzman, D. 1., Gatlin, S. 1., 8: Heilig, I. V. (2005). Does teacher preparation matter? Evidence about teacher certification, Teach for America, and teacher effectiveness. Education Policy Analysis Archives, 13(42). Retrieved [2/16/06] from http://epaa. asu.edu/epaa/v13n42/. Darling-Hammond, L., Chung, R., 8c Frelow, F. (2002). Variation in Teacher Preparation: How well do different pathways prepare teachers to teach. Journal of Teacher Education, 53(4), pp. 286-302. PACE m 115 116 Darling-Hammond, L., 8( Youngs, P. (2002). Defining “highly qualified teachers”: What does “scientifically-based research” actually tell us? Educational Researcher, 31 (9), pp. 13-25. Darling-Hammond, L., Berry, 8., & Thoreson, A. (2001). Does teacher certification matter? Evaluating the evidence. Educational Evaluation and Policy Analysis, 23(1), pp. 57-77. EdSource (2006). California exit exam gets real. Online at: http://www.edsource.org/pub_abs_cahsee06.cfm Esch, C. E., Chang-Ross, C. M., Guha, R., Humphrey, D. C., Shields, P. M., Tiffany-Morales, 1. D., Wechsler, M. E., & Woodworth, K. R. (2005). The status ofthe teachingpro- fession 2005. Santa Cruz, CA: The Center for the Future of Teaching and Learning. Farber, B. (1991 ). Crisis in education: Stress and burnout in the American teacher. San Francisco: Iossey-Bass. Ferguson, R. F. (1991). Racial patterns in how school and teacher quality affect achievement and earnings. Challenge, 2(1), pp. 1—35. Ferguson, R, 8c Womack, 8. T. (1993). The impact of subject matter and education coursework on teaching perfor- mance. Iournal of Teacher Education, 44(1), pp. 55-63. Gandara, P. (2000). In the aftermath of the storm: English learners in the post—227 era. Bilingual Research Journal, 24, pp.1—13. Gandara, P. Maxwell—Iolly, 1. & Driscoll, A. (2005). Listening to teachers of English learners: A survey of California teacher challenges, experiences, and professional devel- opment needs. Santa Cruz, CA: Center for the Future of Teaching and Learning. Gandara, P., Maxwell—1olly, 1. (2000). Preparing teachers for diversity: A dilemma ofquality and quantity. Teaching and California’s Future. Santa Cruz, CA: The Center for the Future of Teaching and Learning. Gandara, P., 1. Maxwell-Jolly, 1., Garcia, E., Asato,1., Gutierrez, K., Stritikus, T., 8c Curry, 1. (1999). The effects ofProposition 227 on the instruction ofEnglish learn- ers: Year one implementation. Davis, CA: University of California, Linguistic Minority Research Institute, Education Policy Center. Gandara, P., Rumberger, R., Maxwell-1olly, 1., 8r Callahan, R. (2003). English learners in California schools: Unequal resources, unequal outcomes. Educational Evaluation and Policy Analysis, 1 1 (36). Gandara, P., & Rumberger, R. (2003). The inequitable treat- ment ofEnglish learners in California’s public schools. Santa Barbara, CA: University of California Linguistic Minority Research Institute. Gibson, M.A, Gandara, P.G, & Koyama, 1.P. (2004). School connections: U.S. Mexican youth, peers, and school achieve— ment. Margaret A. Gibson (Ed.). New York: Teacher’s College Press. Goddard, R. D., & Goddard, Y. L. (2001 ). A multilevel analy— sis of the relationship between teacher and collective ef- ficacy in urban schools. Teaching and Teacher Education, 17(7), pp. 807-818. Goddard, R. D., Hoy, W. K., 8c l-on, A. W. (2000). Collective teacher efficacy: Its meaning, measure, and impact on student achievement. American Educational Research Journal, 37(2), pp. 479—507. Gold, Y., 1996 Beginning teacher support. Attrition, men- toring, and induction. In CB. Courtney (Ed.) Review of Research in Education, pp. 548-594. Gonzalez, 1. E., L. Darling—Hammond, et al. ( 2000). Programs that prepare teachers to work effectively with students learning English. Washington, DC: ERIC Clearinghouse on Languages and Linguistics. http://www. cal.org/ericcll. Haycock, K. (1998). Good teaching matters...a lot. OAH Magazine oinstory, 13(1), pp. 61-63. Haycock, K. (2001). Closing the achievement gap. Educational Leadership, 58(6), pp. 6-11. Hayes, K. and Salazar, 1.1. (2001). Evaluation of the structured English immersion program. Final report: Year 1. Program Evaluation Branch, Los Angeles City Schools. Hayes, K. and Salazar, 1. 8r Vukovic, E. (2002). Evaluation of the structured English immersion program: Final report: Year 2. Program Evaluation Branch, Los Angeles City Schools. Hill, D. (1998). California public education on trial: Taking on the test. (The Daily Republican), Fresno edition, May 19, 1998. Downloaded from the www.dailyrepublican. com/cbest_test. Html Kirst, M., Antonio, A., 8r Conklin—Bueschel, A. (2004). Improving the transition for high school to postsecond— ary education. Stanford University: Policy Analysis for California Education working paper 04— 1. Lee, 8.1. Y Kumashiro, K.K. (2005). A report on the status of Asian—American And Pacific Islander education: Beyond the model minority stereotype. National Education Association, Washinton, D.C. Lucas, T., Henze, R., 81 Donato, R. (1990). Promoting the success of Latino language minority students: An ex- ploratory study of six high schools. Harvard Educational Review, 60, pp. 315-340. Office of the Chancellor, California State University (2003). Teacher education annual report. www.calstate.edu/ TeacherEd/archive/annual_report0304/index.shtml Ovando, C. & Collier, V. (1998). Bilingual and ESL class— rooms: Teaching in multicultural contexts. 2nd ed. Boston: McGraw—Hill. Rogers, 1., 1ellison Holme, 1. 8r Silver, D. (2005). CAHSEE results, opportunity to learn, 8c the class of 2006: More questions than answers. UCLA/IDEA. Online at http:// www.ideagseis.ucla.edu/resources/exitexam/index.html Rosenholtz, S. 1., 8c Simpson, C. (1984). The formation of ability conceptions: Developmental trend or social con- struction? Review of Educational Research, 54(1), pp. 31—63. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Rumberger, R. (2004). Why students drop out of school. In Dropouts in America, Confronting the Graduation Rate Crisis, Orfield, G. (Ed.). Cambridge, MA: Harvard Education Publishing Group. Sanders, W. L., 8i Horn, S. P. (1995). The Tennessee Value- Added assessment system (TVAAS): Mixed model meth— odology in educational assessment. In A. I. Shinkfield 8: D. Stufflebeam (Eds.), Teacher evaluation: Guide to effec- tive practice. Boston: Kluwer Academic Publishers. Sanders, W. L., & Rivers, I. C. (1996). Cumulative and resid- ual efi‘ects of teachers on future student academic achieve— ment. University of Tennessee Value—Added Research and Assessment Center. Smith T.M. 8i Ingersoll, RM. (2004). What are the effects of induction and mentoring on beginning teacher turnover? American Educational Research Journal, 41 (2). Woolfolk, A.E., Rosoff, B. & Hoy, W.K. (1990). Teachers’ sense of efficacy and their beliefs about managing stu- dents. Teaching & Teacher Education. 6 (2), pp. 137— 148. CRUCIAL ISSUES IN PREPARING TEACHERS OF ENGLISH LEARNERS PACE u 117 118 CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 TATE AND FEDERAL ACCOUNTABILITY POLICES OVER THE LAST DECADE HAVE BROUGHT MUCH-NEEDED AWARENESS OF HOW OUR STUDENTS ARE ACHIEVING ACROSS RACIAL, LINGUISTIC, AND SOCIOECONOMIC SUBGROUPS. THIS CHAPTER LOOKS AT ISSUES OF ACHIEVEMENT ACROSS SOCIOECONOMIC STATUS (SES), FOCUSING SPECIFICALLY ON LOW— INCOME STUDENTS. We consider the persistence of gaps in achievement across SES and the resulting educational and economic implications, and review recent policy and educational strategies to address such gaps. Such issues are framed within the context of California’s growing levels of poverty and increasing demands to improve student achievement. It is critical that California’s educators and poli— cymakers continue to focus their efforts on the educa- tion of low—income students. In California, nearly one in every five children lives in poverty, and 48.7% of students are eligible for free or reduced lunch (NCES, 2004). More importantly, California shows a worsen- ing trend on two key indicators of child poverty: I The percentage of children living below the pov- erty line has grown from 18.2 % in 1990 to 19.5 0/0 in 2000. I The percentage of children living in high-poverty neighborhoods has also increased, from 23.4 % in 1990 to 29.6 0/0 in 2000. California’s percentage of low—income children is higher than the corresponding national level and has risen since 1990, contrary to the national trend (Car— roll et al., 2005). Research consistently finds a negative correla— tion between socioeconomic status and a student’s ability to succeed in school (Coleman et al., 1966; Jencks et al., 1972; Sirin, 2005). This means nearly half of California’s students are considered at risk of academic failure due to their family income level, as designated by their eligibility for free or reduced meals. As discussed in Chapter 4, today’s accountabil- ity mandates demand high levels of achievement for all students, including low—income learners, with very real consequences for educators, administrators, and policymakers. Schools and districts face sanctions for missing targets of performance and proficiency, and state policymakers must grapple with the economic and practical challenges of intervention to improve student achievement. 119 required to have academic Under the Clinton administration, Title I faced yet another set of Title I funds under No Child the distribution and allocation Left Behind (NCLB) (Roza et of changes. States were al” 2005). Title I met with further crit- icism in the 19805, coinciding Standards’ define levels of with the release of A Nation at studentproficiency, and Risk in 1983, as lawmakers grew increasingly frustrated with the 6155655 aChiel/emen t- policy’s lack of clear guidelines to measure student achievement 122 (Jennings, 2000). As a result of growing concern that Title I was failing to make progress, the 1988 reauthorization included major changes and stipulations. First, states were required to define levels of achievement for designated students. Second, states had to identify schools that failed to make adequate progress towards these goals. Finally, in response to concerns regarding the flexibility of the use of funds, Title I funds were no longer solely designated for specific services, but also gave schools the flexibility to offer schoolwide programs (Jennings, 2000). Under the new schoolwide program option, Title I schools could use their funds for resources that would benefit the entire school. This increased flexibility was intended to help schools with a high percentage of high-poverty students improve the in— structional performance for the whole school. The 1988 reauthorization required a school to have 75% or more of its students in poverty in order to choose a schoolwide program. The threshold for qualifying for a schoolwide program was set deliber- ately high, as lawmakers wanted to ensure that Title I funds continued to be used to benefit the neediest students (Farkas and Hall, 2000). Schools that did not meet the required poverty threshold, or those who chose not to adopt a schoolwide program could continue to use Title I funds to specifically target the academic needs of the lowest—performing students. Under the Clinton administration, Title I faced yet another set of changes. States were required to have academic standards, define levels of student proficiency, and assess student achievement (Jen— nings, 2000). Furthermore, the poverty indicator for schoolwide programs was lowered, allowing schools to qualify for Title I funds if it had at least 50% or more of its students in poverty. This adjustment cre— ated a dramatic shift in the implementation of Title I, as a growing number of schools were able to use Title I funds to run a schoolwide program (Farkas and Hall, 2000). Current Implementation of Title I in California Title I currently provides more than $12.7 bil- lion, reaching about 15.5 million students in both public and private schools across the United States (NCES, 2004). As of May 2005, there were 5,840 Title I schools in California, representing nearly 60% of California schools (California Department of Educa- tion, May 2005). The federal government estimated that California would receive more than $3 billion in No Child Left Behind (NCLB) funding in 2005, with Title I grants representing about $2 billion of that amount (EdSource, January 2004). The reauthorization of the ESEA in 2001, com— monly known as No Child Left Behind (NCLB), dramatically altered the implementation and require— ments of Title I. As a result, the federal government’s efforts hold all Title I schools and districts account- able for increasing and maintaining student achieve— ment. Under NCLB, schools, districts, and states must meet their Annual Yearly Progress (AYP), a measure of student proficiency.5 Currently, schools and districts in California must meet the following criteria in order to make their AYP: l A 95% participation rate on statewide assess- ments in English Language Arts (ELA) and Math. l Reach an annual measurable objective (AMO). For 2005, 24.4% of students in elementary schools and elementary districts must be pro— ficient in ELA and 26.5% of students must be proficient in Math. For 2005, 22.3% of students in high schools and high school districts must be proficient in ELA and 20.9% of students must be proficient in Math. I An API score of 590 or 1 point growth. I In addition, high schools must have at least a 82.9% graduation rate, or a +0.1% one—year change, or a +02% two—year average change (California Department of Education, 2005). More importantly, under NCLB, schools that receive Title I funds and fail to meet AYP for two consecutive years are now subject to sanctions under Program Improvement (PI). Under PI, Title I schools CRUCIAL ISSUES IN CALIFORNlA EDUCATION 2006 face multiple sanctions, including setting aside addi- tional funds for professional development, providing students with school choice, restructuring the school, and replacing school staff and administration. The Impact of Title I Forty years after its implementation, Title I re— mains the largest policy effort to address the needs of low-income students in the US. Its effectiveness in improving student learning, however, is not always clear. Research on the impact of Title I reveals several common themes, including the following: I Title I is correlated with gains in mathematics and reading achievement, with a stronger impact on math (Carter, 1984; Borman and D’Agostino, 1996). I Gains in achievement are seen primarily in the elementary grades, with effects largely disappear- ing by high school (Carter, 1984; Borman and D’Agostino, 1996). I While Title I serves a great number of low— income students, there are still many children who do not receive services (Puma et al., 1993). I Teachers may hold negative views of students who are eligible for Title I services (Puma et al., 1993). While there are clearly some benefits to Title I, researchers have raised questions about the policy’s long-term effects and, ultimately, its ability to close gaps in achievement. As the authors of a comprehen— sive review of research of Title I from 1966 to 1993 conclude, “Title I alone can not be expected to serve as the great equalizer. The results do suggest, however, that without the program, children served over the last 30 years would have fallen further behind aca— demically” (Borman and D’Agostino, 1996). Any direct link between Title I funds and im- provements in student achievement remains in ques- tion today. For example, a recent national study of the implementation of Title I in its current incarnation as part of NCLB found positive trends in student achievement in Math and English Language Arts. However, the authors conclude that “it is too early to say whether these trends are attributable to NCLB, to other improvement initiatives that preceded it, or a combination of both” (Stullich et al., 2006). ADDRESSING THE NEEDS 0F LOW-INCOME STUDENTS, ‘ STATE AND LOCAL POLICIES— CATEGORICAL FUNDING AND WEIGHTED STUDENT FORMULA The federal No Child Left Behind policy, in conjunc- tion with Title I funding, provides a mandate for improving low-income student achievement. In con- junction with federal policies, state and local agencies have tried to further support low-income students with adjustments to the school funding process. A portion of California’s categorical funding focuses resources on the most needy students.6 Categorical funding, a foundation for school finance in California, was originally intended to al- leviate the burden of educating high-needs students by directing tax dollars towards specific programs or resources. However, following the Serrano decision and Proposition 13, California saw a proliferation of categorical funding programs. In recent years the categorical funding system has improved as a result of the state consolidating cat— egorical funding into six block grants. These changes represent an effort to reduce bureaucratic roadblocks and distribute funds more efficiently to districts. They also signal the state’s acknowledgement of the need for a better balance between restriction and flexibility in the ways funds are allocated to districts. Ideally, this would allow districts to better serve the neediest students. At the same time, districts have begun to experi- ment with how they distribute funds amongst their schools. One such effort is “weighted student formu- la” which decentralizes a district’s power over certain school finance decisions, and puts money directly into the hands of principals. In theory, principals have the autonomy to decide how funds will be used at their school sites, including the allocation of funds for teachers and other personnel and materials. Also known as student—based budgeting, weighted stu— dent formula allows a district to create an allocation formula based on individual student characteristics or needs. Rather than relying upon the traditional method of basing allocations on the number of stu- dents and/ or personnel, funds are allocated based upon specific characteristics of a school’s student population. This results in a focus on the needs of individual students rather than on an average student (Petko, 2005). PACE W 123 Research suggests certain school and classroom While research and imple- is important to acknowledge lOW-inCOme StudentS- Mat the consideration of weighted is most promising is the student formula in state and national discussions of school fi- consistency of findings that nance reform. Within California, emerge from recent studies this system has been implement- ed in the San Francisco Unified of eflective schools serving School District (SFUSD). Dis- low—income students. trict leaders decided to employ student-based budgeting in the 124 hopes of increasing equity and transparency in per—student funding across schools, as well as providing resources to match individual student needs. According to the district, each school “receives a budget denominated in dollars instead of positions and decides what staff and non—staff items to purchase with those dol— lars.”7 While the district office provides oversight and training, principals and their school site councils are responsible for developing their school’s budget and academic plan. The amount of funding per student is determined by a student’s grade level, English learner status, socioeconomic status, and special education status. Currently, almost 60% of the district‘s unre— stricted funds are distributed using this weighted stu- dent formula. This has resulted in additional funding ranging from $155 to $530 for English learners and low—income students (Ackerman, 2004). Currently, the Legislative Analyst’s Office (LAO) is considering weighted student formula as one of several approaches to the distribution of supplemen— tal funds, as part of a larger examination of issues of school finance. The LAO estimates that the use of a weighted student formula would translate into an ad- ditional $331 per student for low-income and English learners (P. Warren, personal communication, March 16, 2006). In order to further evaluate this funding approach, the Assembly Education Committee’s working group on school finance is recommending that, as a pilot project, the state authorize a pilot pro— gram for groups of school districts to use weighted student formula with their categorical funds (Assem— bly Education Committee, personal communication, March 22, 2006). Implementing weighted student formula will pose some serious challenges for policymakers and educators, including difficult decisions regarding which student characteristics are included in any formula, and what weight to assign to them. Further- more, weighted student formula is dependent upon leadership at the school level, particularly among principals and other site leaders. Yet, it is not clear that principals have the skills or training to undertake and successfully implement a decentralized program (EdSource, May 2004). There may also be unintended consequences of students being misidentified into categories in order to receive additional funds. State agencies and local districts must have the capacity to provide adequate professional development and support regarding the implementation and use of a weighted student formula. The concept of a weighted student formula would seem to provide educators with an effective means of addressing the specific needs of low-income students. Anecdotally and through preliminary re- search, such a system suggests potential. While united in scope, one study of three US. school districts using student—based budgeting showed promising gains in student achievement as compared with districts not employing this system (Ouchi, 2004). However, exist- ing research remains inconclusive and its implemen— tation in California is still too early and infrequent to assess any impact on student achievement. Further analyses are necessary to determine any real gains, particularly within the context of our state's educa— tional policies and practices. INSTRUCTIONAL STRATEGIES FOR LOW—INCOME STUDENTS: WHAT DOES THE RESEARCH TELL US? While it may be difficult to determine the impact of particular policies such as funding allocation formu— las on student achievement, research suggests certain school and classroom practices are effective for low" income students. What is most promising is the con— sistency of findings that emerge from recent studies of effective schools serving low-income students. Iust released this year, the “Similar Schools, Dif- ferent Results" report provides a comprehensive analysis of the policies and practices within schools serving low—income students in California today (Williams, T., Kirst, i\l., Haertel, E. et al., 2005). \Vith CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 colleagues from EdSource, Stanford University, and American Institutes for Research, we surveyed ap- proximately 5,500 teachers and 250 principals in elementary schools serving low—income students across California. The purpose of the survey was to determine those policies and practices associated with higher student achievement. The schools in our sample fell within the 25"‘—35‘h percentile on the Student Characteristics Index.8 Interestingly, within this narrow band, there is still a 250 point spread on the Academic Performance Index (API), which sug- gests that while socioeconomic status is one predictor of academic achievement, it is not the sole predictor. Thus, we sought to understand what distinguished those higher-performing schools from their lower— performing peers, given that they were all serving similarly challenged students. Extensive analysis of the survey findings revealed several practices that were more common at high— performing than at low—performing schools serving low-income students. School and district practices found to be associated with higher student achieve— ment include the following: I High expectations and a focus on student achievement across all subgroups. Teachers and principals in higher performing schools were more likely to indicate a culture of high expectations for students within their institutions. These educators reported that their schools had well defined plans for instructional improvement, with a focus on meeting API and AYP targets. Interestingly, teachers and principals in higher performing schools were also more likely to report efforts to exceed API subgroup targets. I Consistency and alignment of a standards- based curriculum and instructional program, with support and oversight from the district. Teachers in higher performing schools were more likely to indicate consistency of curriculum and instruction within and across grades, classroom instruction and curriculum materials in math and language arts aligned with state academic standards, and district support to address the instructional needs of English learner students. ADDRESSING THE NEEDS . 0F LOW-lNCOME STUDENTS Principals in these schools were more likely to report clear expectations from the district for student achievement, aligned with the district’s adopted curriculum, as well as district evaluation of the principal based on the extent to which in— struction was aligned with the curriculum. I The use of assessment data to guide instruction and improve student achievement. Educators in higher performing schools were more likely to indicate the extensive use of stu- dent assessment data by the district and the principal in an effort to inform instruction and improve student learning. Principals in these schools relied on assessment data from multiple sources, including curriculum program assess- ments, district—developed assessments, and the state’s CST and CAT/6 assessments. They used the data to evaluate teachers’ practices, to identify teachers who need instructional improvement, and to develop strategies to track the progress of struggling students. I The experience and certification of teaching staff, with support from the district for instructional and facilities resources. Teachers in higher performing schools were more likely to indicate having a regular or standard certification for teaching in California. Teachers in these schools also reported having at least five years of full-time teaching experience; principal experience was also correlated with higher stu- dent achievement. Principals at higher perform- ing schools were more likely to report that their districts provided sufficient instructional materi- als and support for supplementary instruction for struggling students, as well as support for facilities management. Our survey findings suggest that the principal and district administrators are critical to ensuring high achievement among schools serving low—income students. Principals and district administrators in higher performing schools were more likely to map out and guide a vision for school success. Alongside surveys of teachers and principals, our research team conducted interviews with 20 district PACE $2 125 superintendents and/or their key staff, in an effort to understand the role of the district in raising achieve- ment for low—income students. Superintendents were asked to discuss the three most effective strategies they employed to help improve student achievement at schools that serve high percentages of low-income students. Interestingly, there were strong parallels between the strategies mentioned in our interviews with superintendents and the common findings from our teacher and principal surveys. The five most commonly cited district strategies included the use of data and assessment to inform instruction; provid- ing professional development that helped teachers and principals understand and implement various assessments and curriculum packages; the use of a curriculum package to guide instructional coherence; the role of the principal in managing the implemen— tation of accountability; and the presence of a school culture that valued high expectations regarding stu- dent achievement. Other recent research of schools serving low- income students across California as well as Virginia, North Carolina, and Kentucky reveals similar char— acteristics among the higher achieving institutions. A joint legislative review team in Virginia, for example, found certain elements and practices common in higher performing schools, despite significant demo— graphic challenges (Ioint Legislative Audit and Review Commission, 2004). Common elements included the presence of strong and stable principal leadership, an environment conducive to learning, and an effective teaching staff. Common strategies included the use of assessment data to identify student weaknesses and teacher effectiveness, the alignment of curriculum with proper pacing and resources, the differentiation of instruction and remediation, and maximizing the amount of time spent on instruction. The Virginia study also noted the critical role of the principal in higher performing schools, in continually striving to provide the support needed for students to be suc— cessful regardless of background. Researchers in Kentucky (Kannapel and Clem- ents, 2005) found the following characteristics present in high performing, high poverty schools: I High expectations that were communicated in concrete ways I Respectful relationships among adults, between adults and students, and among students, includ- ing a culture that treated disadvantaged students in “fundamentally similar ways” as advantaged students. I Strong focus on academics, instruction, and stu- dent learning I A system in place to regularly assess the progress of individual students and to plan or change in- struction to meet the student’s needs I Leadership style varied, but all shared a collab- orative decision-making process I Strong faculty work ethic and morale I Careful and intentional manner in which teach- ers were recruited, hired, and assigned. Finally, a recent Education Trust (EdTrust, 2005) study found similar characteristics in seven “high— impact” high schools in California and North Caro- lina. These schools had both a significant population of students of color and above-average proficiency levels on state math and/ or reading assessments. Furthermore, the schools had greater than expected academic growth with previously low—performing students, and had at least an average ability to keep students in school, as well as average, smaller than average, or declining achievement gaps. Schools were compared to others with similar demographics but only average gains. EdTrust researchers found the following condi— tions to be associated with higher student achievement: I Schools exhibit a culture that is clearly focused on preparing students for college and career goals. I Educators have consistently high expectations for all students and take responsibility for students‘ success. I Schools provide extra help to ensure students stay on track with graduation and college—prep requirements and have measures in place to iden- tify students in need before it is too late. I Teacher assignments are made primarily on the basis of the needs of students. I Schools put a high priority on instructional time, arranging extra time for students who have fallen behind, and ensuring that students spend more time on task and in the classroom. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 While state and national assessment data show a strong correlation between socioeconomic status and student achievement, these recent studies suggest op- portunities to disrupt patterns of low achievement. Our research, and that of others on effective schools serving low-income students, consistently reveals the importance of a culture of high expectations, coupled with stable leadership and a vision for school improvement; the use of academically rigorous, stan- dards—based curriculum and instruction; and the use of diagnostic assessments to track student progress and identify student needs. This research also reminds us of the critical need to build capacity among educators and administra— tors at the school and district levels, in order to create the aforementioned conditions for student learning. Without such capacity, any efforts to implement and sustain reform will fail. And it is precisely at this junc— ture of capacity and reform that we find the role of the state. POLICY IMPLICATIONS As this chapter illustrates, there are a variety of reform strategies and policies designed to help low—income students. These programs recognize the link between poverty and edu— cational attainment, and hope to increase a school’s or district’s capacity to educate their most disadvantaged students. Yet as targets and sanctions increase under NCLB, Califor~ nia, like many states, faces mounting pressure to improve the achievement of all students, but particularly low—income students who consistently fail to meet academic targets. What is perhaps most notable about the recent research on high performing, high—poverty schools is the consistency of findings in terms of the characteristics and practices in these institutions, both throughout elementary and high schools, and within California as well as across the nation. These studies pro— vide insight into the strategies that work best FIGURE 4: Distribution of Schools With More Than 50% Economically Disadvantaged Students, Across 2004 API Statewide Decile Rankings 800 700 600 500 400 300 200 ’IOO 0 Number of Schools ’l 2 5 4 5 6 7 8 9 ’10 2004 API Statewide Decile Rankings Source :Tempes, West Ed (2006) for low—income students. The challenge for policymakers is to know how best to support school and district efforts to implement and sustain those strategies. Limited state resources need to continue to target the most disadvantaged students. Over the last seven years, California policy- makers have designated funds to help im- prove low—performing schools, first through the Immediate Intervention/Underperform— ing Schools Program (II/USP), and now through the High Priority Schools Grant Program (HPSGP). While the initial II/USP efforts reached schools in the lowest five deciles of performance, severe funding re— strictions have resulted in HPSGP’s focusing solely on schools in the lowest two deciles. This focusing of resources allows the state to reach a significant population of low—income and otherwise disadvantaged students within the state’s fiscal constraints. As indicated by Figure 4, schools with a majority of low- income students are disproportionately rep— resented in the lowest deciles (Tempes, 2006). API data for Decile 1 schools shows an av- erage gain of 176 points from 1999—2003, PACE 127 128 compared to the state average growth of 86 points (Bassok et al., 2004). However, while the majority of Decile 1 schools show notice- able growth, they still have a great distance to cover. For example, in 2003 the average API score for Decile 1 elementary schools was 576, a difference of over 200 points from the state target of 800 (Bassok et al., 2004). In order to help the schools with the greatest propor— tion of low—income students, the state should continue to dedicate additional resources to Decile 1 and 2 schools. The state can continue to support the use of curriculum standards and diagnostic assessments. As indicated in the studies previously dis— cussed in this chapter, there is growing evi— dence that certain school and classroom strat— egies are successful in raising achievement among low—income students. In particular, the research suggests a need to support edu— cators’ understanding and use of curriculum standards and diagnostic assessment data. California’s curriculum standards already provide educators with a clear roadmap of expectations for student learning. The state should continue to support the use of cur— riculum standards through initiatives that help teachers implement standards into their daily classroom practice. Where standards provide a roadmap for teaching and expectations for learning, diag— nostic assessments can provide a roadmap of student learning and for areas of improve— ment. However, educators often lack the ca— pacity to interpret assessment data to inform practice (Armstrong and Anthes, 2001; Noyce et al., 2000; Sharkey and Murnane, 2003). The state should identify districts and schools that have shown considerable success in creating data systems, and facilitate the sharing of ef- fective strategies to educators across the state. Additionally, the state should continue to build the individual student identifier system (CALPADS) and ensure that schools and teachers have adequate professional develop- ment to make the connections between as- sessment results and student learning. Provide professional development that meets the needs of today’s educators and students. Research tells us that when professional de- velopment provides content-based knowledge and is implemented in a sustained, consis— tent fashion, it can lead to increased student achievement (Darling-Hammond, 1998; Garet et al., 1998; Porter et al., 2000). How- ever, as discussed in Chapter 7, professional development does not always address the spe- cific needs of teachers and their students. In our own research of low—income schools, we asked teachers and principals to rank their top three priorities for additional professional development. This provided us with a critical understanding of where teachers and princi— pals felt there were gaps in their abilities, and ultimately how best to develop the capacity of educators working with economically disad— vantaged students. Interestingly, teachers in both high—perform- ing and Program Improvement (PI) schools noted similar priorities, indicating a need for professional development in instructional strategies for multiple learning styles, and English Language Arts and Math curriculum programs. Teachers in lower-performing schools also noted the need for professional development around instructional strategies for English learners. Principals in low—income schools across all performance levels in our study noted the need for professional development concern- ing the use of assessment data. Principals in the low—performing and PI schools indi— cated a need for additional training around instructional strategies for EL students; and CRUCIAL ISSUES IN CALIFORNlA EDUCATION 2006 principals in PI schools also wanted ad- ditional training in implementing a stan— dards-based curriculum as well. In addition, principals from low-performing schools also requested professional development that focused on helping them develop a school plan or shared vision as their main priorities. Finally, principals in the higher performing schools wanted additional training in evaluat- ing teachers’ instruction and addressing mul— ticultural/diversity issues at their schools. Teachers and administrators need profes— sional development that is directly appli— cable to the daily challenges they face inside their schools. Our research findings provide policymakers with a clear idea of the specific areas of professional development, in order to build capacity among educators serving low- income students. Current professional devel— opment offered by the state may not always be coherent and may not always be linked to the needs of teachers and their students. The California legislature recently passed a new education bill (AB 2109) that would require school districts receiving state professional development grant funds to analyze disag- gregated achievement data and structure their professional development programs to focus on improving achievement according to the performance of each subgroup. Such a pro— posal would provide districts with the flexi— bility to implement professional development that meets the needs of their specific teachers and students. Ultimately, any further legisla— tion for professional development needs to build upon the lessons learned from previous state—mandated efforts. California has a strong foundation of policies and programs that target low-income stu— dents. It is the role of the state to build edu— cators’ capacity to implement those policies and programs and to sustain higher levels of student achievement. While the research on high performing, low—income schools points N ADDRESSING THE NEEDS ' or LOW-INCOME sruorms to common strategies, it is important for pol- icymakers to remember that there is no one magic bullet to improve achievement for low- income students. Rather, the solution lies in finding a balance between supporting strate- gies that have proven successful across a wide range of schools, and allowing individual schools and districts the flexibility to address the needs of their specific communities. ENDNOTES A numerically significant subgroup is defined by the California Department of Education (CDE) as having “at least 100 or more students with valid STAR scores or 50 or more students with valid STAR scores who make up at least 15% of the total valid STAR scores.” http://www.cde. ca.gov/ta/ac/ay/apiaypelement504.asp. The (CDE) defines economically disadvantaged as “a student whose parents both have not received a high school diploma OR a stu— dent who participates in the free or reduced-price lunch program.” http://www.cde.ca.gov/ta/ac/ap/glossary06c.asp http://api.cde.ca.gov/datafiles.asp Data from Figures 2 and 3 includes first—time and repeat test takers, as reported by the CDE. Obtained courtesy of Iohn Woolley and Gerhard Peters at The American Presidency Project, University of California at Santa Barbara, http://www.presidency.ucsb. edu. For further discussion of California’s proficiency stan— dards, see Chapter 4. http://cde.ca.gov/fg/aa/ca/index.asp http://portal.sfusd.edu/template/default. cfm?page=initiativesformula The Student Characteristics Index (SCI) includes such factors as socioeconomic status, race or ethnicity, par- ent education level, and English learner status. Schools within the 25th—35th percentile of SCI serve California’s significantly disadvantaged students. REFERENCES Ackerman, A. (2004, August). The weighted student formula: A superintendent’s perspective. The School Administrator Web Edition. Retrieved March 28, 2006, PACE W2 l29 130 from http://www.aasa.org/publications/content. cfm?ItemNumber=1190 Armstrong, 1., 8r Anthes, K. (2001, April). Data—driven policymaking (Policy Brief). Denver, CO: Education Commission of the States. Retrieved January 11, 2005, from http://www.ecs.org/clearinghouse/30/69/3069.htm Bassok, D., Davis, D., 8r Raymond, M. (2004). Not con- demned to fail: An analysis of California’s low performing schools. Stanford, CA: Hoover Institution. Borman, G.D., & D’Agostino, IV (1996). Title I and stu- dent achievement: A meta-analysis of federal evaluation results. Educational Evaluation and Policy Analysis, 18(4), 309-326. California Department of Education. (2005). 2005 Accountability progress report. Sacramento, CA: California Department of Education. California Department of Education. (2005, May). State schools ChiefO’Connell announces Title I academic achiev- ing schools. Sacramento, CA: California Department of Education. Retrieved 1anuary 4, 2006, from httpzwww. cde.ca.gav/nr/ne/yr05/yr05re154.asp Carroll, 81., Krop, C., Arkes, 1., Morrison, P.A., & Flanagan, A. (2005). California’s K-12 public schools: How are they doing? Santa Monica, CA: RAND. Carter, L. F. (1984). The sustaining effects study of compen- satory and elementary education. Educational Researcher, 13(7), 4- l 3. Cheesman—Day, 1., 8( Newburger, EC. (2002, July). The big payoff: Educational attainments and synthetic estimates of work—life earnings (PC23-210). Washington, DC: U.S. Census Bureau. Coleman, 1.5., et al. (1966). Equality ofeducational opportu— nity. Washington, DC: U.S. Government Printing Office. Darling—Hammond, L. (1998). Teacher learning that sup- ports student learning. Educational Leadership, 55(5), 6—1 1. EdSource. (2004, January). No child left behind? The irri— pact of the federal NCLB act so far. Mountain View, CA: EdSource. Retrieved May 2, 2006, from http://www.ed— source.org/edu_nclb.cfm EdSource. (2004, May). Weighted student formula concept enlivens school finance debate. Mountain View, CA: EdSource. Education Trust, The. (2005, November). Gaining traction, gaining ground: How some high schools accelerate learning for struggling students. Washington, DC: The Education Trust. Farkas, G., & Hall, LS. (2000). Can Title I attain its goal? In D. Ravitch (Ed), Brookings papers on education policy 2000 (pp 59-123). Washington, DC: The Brookings Institution. Gandara, P., Rumberger, R., Maxwell-Iolly, 1., & Callahan, R. (2003). English learners in California schools: Unequal resources, unequal outcomes. Education Policy Analysis Archives, 11(36). Retrieved April 15, 2005, from http:// www.epaa.asu.edu/epaa/v11n36 Garet, M.S., Porter, A.C., Desimone, L., Birman, B.F., & Yoon, KS. (2001 ). What makes professional development effective? Results from a national sample of teachers. American Educational Research Journal, 38(4), 915—945. 1encks, C., et al. (1972). Inequality: A reassessment ofthe efi‘ect offamily and schooling in America. New York, NY: Basic Books, Inc. 1ennings, 1.F. (2000). Title 1: Its legislative history and its promise. Phi Delta Kappan, 81(7), 516-522. 1epsen, C., & Alth, S. (2005). English learners in California schools. San Francisco, CA: Public Policy Institute of California. Joint Legislative Audit and Review Commission. (2004). Review of factors and practices associated with school per- formance in Virginia. Richmond, VA: Author. Kannapel, P.1., & Clements, SK. (2005). Inside the black box of high-performing high-poverty schools. Lexington, KY: Prichard Committee for Academic Excellence. National Center for Educational Statistics. ( 2004). State profiles: The nation’s report card — California. Washington, DC: NCES. Retrieved Ianuary 4, 2006, from http://nces. ed.gov/nationsreportcard/states/profile.asp National Center for Educational Statistics. ( 2005). The con— dition ofeducation 2005. Washington, DC: NCES. National Commission on Excellence in Education. (1983). A Nation at Risk. Washington, DC: U.S. Government Printing Office. Noyce, P., Perda, D., & Traver, R. (2000). Creating data driv— en schools. Educational Leadership, 57(5), 52—56. Ouchi, W.G. (2004). The implementation ofa decentral- ized organization design in three large public school dis— tricts: Edmonton, Seattle and Houston. Los Angles, CA: University of California, Los Angeles. Petko, M. (2005). Weighted studentformula (WSF): What it is and how does it impact educationalprograms in large urban districts? Washington, DC: National Education Association. Porter, A.C., Garet, M.S., Desimone, L., Yoon, K.S., 8r Birman, B. (2000). Does professional development change teaching practice? Results from a three-year study (DOC #2000—04). Washington, DC: U.S. Department of Education. Puma, M. 1., et al. (1993). Prospects: The congressionally mandated study ofeducationalgrowth and opportunity. The interim report. Washington, DC: US. Government Printing Office. (ERIC Document Reproduction Service No. ED361466) Roza, M., Miller, L., & Hill, P. (2005, August). Strengthening Title I to help high poverty schools: How Title I funds fit into district allocation patterns (Working Paper). Seattle, WA: Center on Reinventing Public Education. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 San Francisco Unified School District. (2005). Weighted student formula— FAQ. San Francisco, CA: Office of the Superintendent of San Francisco Unified School District. Retrieved January 11, 2006, from http://portal.sfiisd.edu/ template/default.cfm?page=initiatives.formula Sharkey, N.S., & Murnane, R]. (2003). Learning from stu- dent assessment results. Educational Leadership, 61 (3), 77—8 1 . Sirin, SR. (2005). Socioeconomic status and academic achievement: A meta-analytic review of research. Review of Educational Research, 75(3), 417—453. Snyder, H., & Sickmund, M. (2006). Juvenile oflenders and victims: 2006 national report (NCJ 212906). Washington, DC: Department of Justice. Stullich, S., Eisner, E., McCray, J., & Roney, C. (2006). National assessment of Title I interim report. Volume 1: Implementation of Title I (NCEE 2006—4001). Washington, DC: Institute of Educational Sciences. Tafoya, S. (2002, February). The Linguistic Landscape of California Schools. San Francisco, CA: Public Policy Institute of California. Tempes, F. (2006, March). High impact: Strategic state in— tervention in California’s struggling schools and districts. Paper presented at the meeting of the PACE Legislative Seminar series, Sacramento, CA. Williams, T., Kirst, M., Haertel, E., et a1. (2005, October). Similar students, different results: Why do some school do better? Mountain View, CA: EdSource. ADDRESSING THE NEEDS OF LOW-INCOME STUDENTS PACE n 131 Preschool refers to half-day or full-day, center-based programs that aim to nurture children’s academic and social-emotional development. Increasingly, child—care centers are being called preschools, indicating their support of children’s academic achievement. Yet since the federal government interviewed directors of child care centers in 1990, the border between child care and preschool has blurred (Kisker et al., 1991 ). The old conception of”custodia|” child care versus enriched nursery school has given way to a continuum of low to high— quality center-based programs, regardless of whether they are called a center or preschool. (RAND, estimates of 2001 Census Bureau data). Given the number of mothers requiring some alternate care for their young children and evidence of preschool’s benefits, enthusi— asm has been fueled for widen- ing access for California chil— dren (Barnett, Hustedt, Robin, and Schulman, 2005; Cooper and Dukakis, 2004). Yet, children from different back- grounds enroll in preschool at discrepant levels (see Figure 3), which may have implications for their later achievement. Most recently, the in— terplay between K—12 school reform and child care activ— ists has prompted sharper debate over the basic goals of preschool. As local educators face intensifying pressure to raise test scores—pressed by Sacramento and Washington policymakers—the potential ability of preschool to advance knowledge of language and mathematical concepts has become quite attractive. Proposition 82, for example, would have authorized the state department of education to devise “content standards” for all publicly funded preschools, aligning classroom practices to the academic knowledge ad— vanced in elementary schools. While Proposition 82 failed to pass, policymak— ers moved forward on legislation aimed at expand— ing and improving preschools. As part of Governor 134 Schwarzenegger’s stated efforts to close achievement gaps by boosting preschool opportunities, he signed AB 172 in September 2006. The legislation autho— rizes preschool programs to spend $50 million a year to hire more teachers, improve staff training, and build literacy and outreach programs. Preschools qualify for program funds if they are located near elementary schools with low Academic Performance Index scores (the lowest third). The bill also in— cludes funding accountability and program evalua— tion provisions. Few disagree that child-care teachers can do more to advance pre-literacy skills. But some school dis- tricts already have adopted curricular packages that focus on instruction in pieces of knowledge, taking away from time spent engaging in more self—directed exploration of cognitively demanding tasks and social relationships (for review of these developments, Fuller, Bridges, 8r Pai, in press). 50, as policy options are considered—and com- pared to empirical evidence—we should not lose track of the basic goals of classroom and home—based child care for very young children. Government and educators have the power to either inform—or make decisions for—parents. Why Quality Child Care and Preschool Is Important Despite dramatic growth in child care and pre— school programs since the 19605 many five-year—olds enter kindergarten already behind in their basic cognitive and language skills.1 These gaps are wide, especially for African American and Latino children from low—income communities, compared with their peers from better-off families. Almost the entire dis— parity observed in fourth—graders’ reading scores can be observed in pre—reading skills and linguistic pro— ficiencies at age five (Rumberger and Arellano, 2003; Rumberger and Tran, 2006). Given that preschool enrollment varies across groups—affluent children are more likely than their economically disadvantaged peers to attend preschool (see Figure S)—expanding preschool access may counter these achievement discrepancies. Yet, a dilem— ma has arisen with respect to universal preschool: if all children benefit from preschool, and parents of af— fluent children are more likely to enroll their children in these out—of—home programs, a universal program may support or even exacerbate the disparities seen across groups (Bridges et al., 2006; Loeb et al., 2005). Figure 1 shows the magnitude of the achievement gap between whites and blacks, and between whites and Latinos, in reading and math in fourth and eighth grade in California. (The data is derived from the Na— tional Association of Educational Progress or NAEP, dubbed “The Nation’s Report Card”). This gap is large in fourth grade and remains sizeable in eighth grade, although decreases slightly (except between whites and blacks in math, where it increases slightly). CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 , EXPANDlNG AND lMPROViNG PRESCHOOL . FIGURE 1: Achievement Gaps in Reading and Math for Whites-Blacks and Whites-Latinos in 4th and 8th Grade in California (NAEP, 2005) 1.0 0.9 I White-Black g 0.8 — . . g 0.7 — - White-Latino .9 O 6 > . " 8 _ Source: NAEP data retrieved U 0‘5 June 24, 2005 from http:// L . 8 0.4 — nces.ed.gov/nationsreport- % 0.3 _ card/naepdata/ 4,.) m 0.2 “ 0.’l — 0.0 — 4th Grade Reading 8th Grade Reading 4th Grade Math 8th Grade Math The achievement gaps are represented in a unit of measurement called an effect size, which represents the difference in achievement test scores as a fraction of a standard deviation (Cohen 1988). A standard deviation (SD) from a mean test score portrays the array of children’s test scores on either side of the average. Given the classic normal curve of test scores, where most children are clustered around the average score, two-thirds of all children will score within one SD above or below the mean. By depicting a group’s fraction of a SD above or below the mean, one can compare differences in variables of interest within the same study and between different ones. One can also compare the size of the achievement gap with the effect size of interventions designed to close it. These achievement gaps—marked at fourth and eighth grade—are even present at school entry in kindergarten (Bridges, Fuller, Rumberger, and Tran, 2004). They translate into children starting kinder— garten at dramatically different levels. For example, at kindergarten entry there is a three-month discrepancy between white and (English—proficient) Latino chil- dren in pre-reading skills, and a larger discrepancy in math. Low—income children start kindergarten more than two months behind middle—class children in reading, and about six months behind children from affluent families. This underestimates the white—Latino achievement gap because the assessment procedure ex— cludes those Latino students who are not yet proficient in English—about 30% of the Latino sample. How do we boost the performance of children from low-income families? Studies of high-quality, high-cost preschools—such as Perry Preschool in Yp— silanti, Michigan or Abecedarian Preschool in Chapel Hill, North Carolina—provide evidence that they can benefit low—income children in cognitive, language, and behavioral outcomes. They have been shown to prepare children for starting school and have notable longer—term effects, including: higher test scores, lower rates of grade retention and referral to special education, as well as higher rates of high school grad— uation. What we don’t know is how to build a large network of preschool programs that can maintain high levels of quality and yield such notable benefits. Nor do we know—across an array of possible inputs and classroom practices—how to invest in quality improvements in a cost—effective manner. These intervention programs with such promising results have been conducted with small groups of chil— dren but sizeable resources. Another program, the Chi- cago Child—Parent Center (CPC), attempted to take a comprehensive intervention to scale and offered a vari— ety of services to thousands of primarily poor African— PACE W] 135 136 FIGURE 2: Developmental Gains from Participation in Center-Based Programs 0.25 Fractions of SD Above the Mean 0.20— 0.15— 0.10— 0.05 — 0.00 — Source: Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 weighted by fall kindergarten child- parent-teacher weight (C1 CPTWO). California Sample Pre-Reading Math American children in low-income neighborhoods in Chicago (Reynolds et al., 2001). This program offered preschool, home visits, health care, parenting classes, and social services to their participating families. Children who participated in the CPC programs displayed significant cognitive gains in comparison to similar children who did not attend (Karoly and Big— elow, 2005). The CPC children were less likely to be in special education programs, less likely to repeat grades, and more likely to graduate high school (Reynolds et al., 2001 ). Most notably, these significant effects were main— tained into adolescence, particularly if the children re— mained in enriched classrooms (Reynolds et al., 2001). While the scale of the CPC study shows prom— ising results, it, too, provided extensive resources beyond mere preschool participation to its partici— pants. Moreover, the sample was a distinct group: participants were largely poor, African-American chil— dren living in Chicago—a homogeneous group com— pared to the great diversity of children in California. Other Evidence on the Benefits of Preschool Recent data from the Early Childhood Longi- tudinal Study (Kindergarten Sample) supports the proposal that preschool broadly conceived—ranging in type and quality—may benefit children. Those who attended center-based programs in the year before kindergarten were, on average, one to two months ahead cognitively compared to children who did not attend such programs (controlling for other differ- ences between the groups). These benefits are notably smaller than those found in children participating in extremely enriched programs, but are significant considering they account for up to a quarter of the children’s learning in kindergarten. Moreover, these estimates refer to all children—both those with many advantages and those with none—and all types of programs—high- and low—quality. It may be that in high—quality programs, disadvantaged children expe- rience greater gains. Developmental Gains from Preschool Centers Given the apparent benefits to children, and with some 65% of mothers with children aged five and under working outside the home, preschool has become increasingly popular (Ruhm, 2004). In recent years, policymakers have addressed the tenacious problem of low achievement among economically disadvantaged youth by developing preschool pro— grams, particularly for children from poor families. Currently, Sacramento spends about $3 billion annu— ally on a variety of early care and education programs for children, with about half paid directly to state FIGURE 3: Care Arrangements the Year Before Kindergarten, by Ethnicity/Race in California 9O 8O 85 - Relative Care C _. g 70 _ 75 I Non-Relative Care g 60 65 62 _ “6 50 — [:1 Center-Based Care a) 50 Q 40 __ E _ __ SourcezEarIyChildhood Longitu- § 30 dinal Study, KindergartenCIass (“1’ 20— - ofl998—99weighted byfallkin— _ dergarten child-parent-teacher 10 _ weight (C1 CPTWO) O— Asians Blacks Latinos Whites Other Ethnicitv/Race CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 preschool and center-based programs, and the other half going to parents in the form of vouchers (Hirsch- berg, 2002). These vouchers are to be used by parents to pay a child-care center or a babysitter, relative, or other child—care provider. While there are quality regulations for state pre— school, no such guidelines are in place to ensure that the voucher money from the state that is paid to bab— ysitters, relatives, or other child—care providers is sup- porting high-quality care. Detractors claim that the state is facilitating the provision of low—quality care by subsidizing it through these voucher programs which lack quality guidelines. But unless the quality of home—based care ar— rangement improves, funding through vouchers may inadvertently place disadvantaged children further behind their peers in school readiness and achieve- ment. A variety of studies have shown that poor chil— dren are more likely to experience sustained cognitive gains after attending a center rather than a non—paren— tal home-based arrangement (Burchinal, Peisner-Fein- berg, Bryant, and Clifford, 2000; Hamre and Bridges, 2005; Howes 1997; Loeb, Fuller, Kagan, and Carrol, 2004). On the other hand, if high—quality care is avail— able, vouchers allow parents to choose the arrange- ments they prefer. Admittedly, quality is differentially defined by parents. Well over half of federal voucher dollars are used by parents for center—based programs. Additional work is needed to integrate what is known about child development with parents’ cultural values and preferences (Matthews and Ewen, 2006). Whether by choice or out of necessity, parents are placing their children in a variety of types of non- parental care in the year before kindergarten, includ- ing relative care, non-relative care, and center—based care, which includes preschools, Head Start, and other centers. The use of these arrangements varies by ethnicity/race, home language, and socioeconomic status. Relative care includes grandparents, aunts, and uncles; non—relative care refers to babysitters, family child-care providers, or any non-related arrangement. In the year before starting kindergarten, some 85% of African-American children are in center- based care arrangements, while only 50% of Latino children are. Relative care is most commonly used by Asian parents with 38% of their children being cared for by grandparents, aunts, and uncles; it is least common among whites (19%). In direct contrast, FIGURE 4: Percentage of California Children in Center-Based Care the Year Before Kindergarten, by Language Background 70— 60— 50“ 40“ 30— 20— Percentage of Children ’lO‘ English Language English Spanish Other only minority dominant dominant dominant Language Background SOURCE: Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 weighted by fall kindergarten child-parent-teacher weight (C1CPTWO) non—relative care is used by whites most (24%), and least likely to be used by Asian parents (5%). Related to the ethnic/ race break-down of child- care arrangements, the use of center-based care varies by language background. Of children whose home language is only English, 70% are in center-based care the year before kindergarten, as shown in Figure 4. While participation varies by language group, chil— dren from Spanish—dominant households are least likely to attend center—based care (47%). In spite of financial supports for children from the lowest-income groups, many families are unable to afford preschool—particularly high—quality pre— school. Child care is expensive, and the working poor are not eligible for subsidies yet they cannot pay for it themselves either. As shown in Figure 4, participation in child care is associated with family social class. Less than half (46%) of children from low—income families in California are enrolled in center—based care the year before starting kindergarten, compared with almost two—thirds of children from middle—income families, and 80% of children from high-income families. Given the benefits participation in center—based care provides to children from low-income families, dis- crepancies in access are problematic. Recent history indicates that a broad achieve— ment gap remains between children from low-income families and their counterparts in middle- and high- income families. This overlaps with ethnic/ racial 137 PACE 138 FIGURE 5: Percentage of California Children Enrolled in Center—Based Care in the Year Before Kindergarten, by Socioeconomic Status 80 7O 60 50 40 30 20 ’IO 0 Percentage of Children Low Middle High Socioeconomic Status Source: Early Childhood Longitudinal Study, Kinder- garten Class of 1998-99 weighted by fall kindergarten child-parent-teacher weight (Cl CPTWO) achievement gaps: Black and Latino children start school significantly behind their Asian and white peers. Preschool appears to boost children’s achieve— ment, and may play an important role in decreasing the achievement gap, particularly for disadvantaged children. Broadening access to high—quality preschool would benefit children who may not otherwise qualify for subsidies or have the opportunity to attend. EXPANDING ACCESS TO QUALITY CHILD-CARE OPTIONS Most California youngsters participate in preschool prior to starting kindergarten; estimates vary from 61% (using ECLS-K data) to a recent RAND estimate of 65%. While these rates vary among particular groups, given the apparent benefits to children, there is a groundswell of support for making preschool available to more children in California. Several states, including Georgia, Oklahoma, Florida, and the Abbott Districts in New Jersey have even implement— ed universal preschool—with each state designing the program to meet its specific needs. As options for preschool expansion are explored, several aspects of implementation demand further consideration. How programs would be further de— veloped depends upon decisions made on a number of related issues, including 1) access and inclusion, 2) curriculum and philosophy, 3) language and culture, 4) teacher qualifications, 5) program financing, and 6) auspice and oversight. Access and Inclusion Given the generally high rate of participation of children in center-based programs the year before they start kindergarten, one basic question regarding pre- school expansion is how it will change current enroll— ment rates. Participation is not equal across ethnicity, language, and SES subgroups—broadening access may redistribute enrollment. Family participation in pre- school typically depends on financial resources, wheth— er or not parents—in particular, mothers—are working outside the home, their neighborhood’s resources and center capacity, as well as the family‘s ability to access those centers (Burchinal and Nelson, 2000; Kisker and Maynard, 1991 ). Even if preschool is free, financially disadvantaged families may have difficulties with trans- portation, language, or working “non—business" hours which render participation challenging. Broadening access to preschool for children from low—income families, through outreach and the development of programs that meet their needs, is fundamental to preschool expansion. While preschool benefits all children, recent evidence indicates that ex- tremely poor children may benefit the most (Burchinal et al., 1997; Loeb, et al., 2006; Magnuson, Ruhm, and Waldfogel, 2004)—and they most need the achieve- ment boost. Ironically, because all children appear to gain by attending preschool, if preschool is made more available to a_ll children, including those from affluent families as a universal program would do, greater access and participation may sustain the achievement gap. Evidence is available for countering the gap by targeting low-income communities with preschool expansion. Recent findings indicate that longer—term, moderate exposure may maximize the benefits of preschool learning for children (Loeb, et al., 2005). Offering preschool to some three— and four—year—old children lies—may be one way to prolong their exposure to namely, those from disadvantaged fami— preschool learning. While the second year of atten— dance offers slightly fewer gains to children relative to the first, longer exposure was beneficial and may help to reduce the achievement gap if offered to children from low—income families (Reynolds, 1995). But how then can preschool expansion reach the most disadvantaged children? Depending on the funding and fees associated, expansion may facilitate more equal access to preschool for poor and work- ing class families who previously have not been able CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 to afford it. Those children from the most disadvan- taged families may have access to preschool through subsidized, public programs in their communities. Thus, participation appears to be relatively flat across lower social class groups, given that subsidies increase access for the most disadvantaged, and few working class families can afford expensive preschool fees. Pre— school expansion could broaden access across these social class groups. For working families, maternal employment leads to increased demand for child-care. National estimates indicate that 91% of four-year—old children whose mothers work attend preschool (National Household Education Survey, 1999). However, when faced with low-wage jobs, some mothers may opt to stay home with children, given their pay would be “cancelled out” by the high cost of potentially low- quality child care. While preschool expansion could address con— cerns of financial feasibility for some families, there is also the question of access. Low-income neighbor— hoods have significantly fewer preschools than afflu— ent neighborhoods, making it difficult for families to participate (Fuller, Waters Boots, Castilla, and Hirshberg, 2002). If preschools are not nearby, a lack of transportation may make attending preschool impractical, if not impossible. Supporting local infra- structure and developing more preschools in high- need neighborhoods is essential to expanding access. Socioeconomic status tends to overlap with eth— nicity and family home language. Latinos, who make up the largest ethnic group of California children, are least likely to use preschool. Moreover, among those Latino children whose home language is Spanish, only 47 percent attend preschool (Bridges, Rumberger, Fuller, and Tran, 2006). A host of reasons appears to explain this as noted above, including—but not limited to—family preferences related to having chil— dren cared for by family or by those who share their cultural values and language. Locating preschools in the community with staff who reflects the same background as the children served, may help Latino parents feel more comfortable about sending their children to local preschools. Neighborhood—based preschools may also be more likely to cater to the specific needs of the com- munity. For instance, many families who live in low— income neighborhoods are more likely to work long EXPANDING AND lMPROVlNG PRESCHOOL ‘ or “non-business” hours, which further complicates finding preschools that open earlier in the day or remain open later in the evening. Broadening access and inclusion is complicated as evident in states with “universal” programs. Given that participation is not mandatory, states with uni- versal programs have participation rates of 70% (Bar- nett, Hustedt, Robin, and Schulman, 2005). In Cali— fornia, some predict that a universal preschool policy would increase preschool enrollment rates from about 62% to as much as 80% (Fuller, in press). Such an in- crease in enrollment would likely to be accompanied by other dramatic changes in preschool education, such as which children are served and where, who teaches them and in what language, what they are taught, and who pays for these services. Curriculum and Philosophy Once children are brought into the preschool community, there are differing opinions about how they should spend their time and how their class- rooms should be structured. Two of the primary approaches are DAP (developmentally appropriate practice) and direct instruction. The former proposes embedding learning into engaging, child—centered activities that address children’s cognitive as well as social and emotional development. Direct instruction, in contrast, is more narrowly targeted on the skill development needed for kindergarten and beyond through structured academic activities. As teachers and schools feel the pressure to ensure that children meet the requirements set forth by the No Child Left Behind Act, the potential gains offered by starting academic skills early is tempting. In this vein, the California Department of Education has attempted to revise its pre—kindergarten quality standards to be more academically rigorous. Evidence supports both approaches as beneficial to the cognitive development of children. DAP cur- riculum seems to encourage greater enthusiasm for learning among young students, which may facilitate later learning (Dunn and Kontos, 1997; Stipek et al., 1999; Stipek, 2005). Direct instruction appears to achieve slightly higher cognitive gains, but also increases children’s stress levels (Stipek, 2005). Others who have examined whether a blended ap— proach—making direct instruction more develop- mentally appropriate—could maximize gains found PACE V] 139 140 that either approach implemented more purely was more effective than blending the two (Stipek, 2005). However, there was some evidence that a child—cen- tered, developmentally appropriate practice curricu— lum better sustained benefits to children over time (Marcon, 2002). Sustained effects are important in boosting achievement, and the benefits of many programs appear to dissipate by third grade (Rumberger and Tran, 2006). This phenomenon may be different for children in particularly high—quality preschool programs if they subsequently enter supportive and high-quality elementary schools. Consistency in cur- ricular approach and philosophy has been found to be a driving factor in children’s maintaining cogni- tive and social benefits gleaned in preschool (Stipek et al., 1999). Language and Culture In order for the preschool curriculum and phi— losophy to be effective with California’s children, special consideration regarding language and culture is essential (Matthews and Ewen, 2006). The im— portance of continuity in language and culture has wide—reaching implications for California’s diverse children. Given the high percentage of California preschoolers who live in non—English-speaking homes, many children arrive at preschool for their first substantive exposure to English (Bridges, Rum— berger, Fuller, and Tran, 2006). Supporting children’s home language development—whatever it is—has been shown to facilitate the acquisition of language and literacy in English (Snow, Burns, and Griffin, 1998). Teachers who are able to communicate with their students—preferably in the student’s home lan— guage—and provide continuity with family expecta— tions in the social and behavioral demands they place on children, may be an important part of program quality (Wishard, Shivers, Howes, and Ritchie, 2003). These teachers may support their students in cultur— ally appropriate ways that children consistently recog— nize and to which they respond. Less is known about how different cultural expectations and practices be— tween home and preschool affect children, and their later academic achievement and social—emotional well-being. Teacher Qualifications Highly qualified teachers are needed to teach our preschool children—but what type of teacher education and training benefit children most? Gen— erally speaking, teachers with more formal education are likely to provide the highest quality care to chil— dren (NICHD ECCRN, 2000). Yet, recent evidence indicates that teachers who have bachelor’s degrees do not provide higher—quality care than teachers without, and their education is not consistently as- sociated with any academic gains for their young students (Early et al., 2006). Careful examination of results from multiple programs in many different states supports this finding. Some maintain that since B.A. degrees are re- quired of other K- 12 teachers, similar standards should be maintained for preschool teachers, re— gardless of whether a two—year degree would suffice. However, the preschool workforce is dramatically under—qualified for such requirements. Currently, only about one—third of preschool teachers pos— sess bachelor’s degrees, and that third tends to be concentrated in more affluent communities (White— book, Burton, and Young, 2002). Increasing the training of the rest of the workforce would require tremendous resources, and the community college infrastructure for training them does not currently exist. This fact was made clear when California counties recently implemented incentive programs to increase the training and retention of the early care and education work force, and coursework was not available or accessible to some teachers (Hamre et al., 2004). Such formal education requirements attached to preschool expansion would necessitate tremen— dous preparation of teachers for B.A. degrees (Bellm, Burton, Whitebook, Broatch and Young, 2002). Most likely a majority of those preschool educators pursu- ing a B.A. would need to work full—time, and could only take a couple of classes per year. Recent evidence showed that a financial incentive program which paid preschool teachers to take unit—bearing course— work gained the average teacher about six units—or two courses—a year (Hamre et al., 2004). It would take most preschool workers years to complete their degree. Decidedly this is a slow route to developing a larger preschool work force. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 The focus on teachers’ formal education has eclipsed the most essential aspect of preschool quality: positive teacher—child interactions. Less is known about how to cultivate teachers who interact in highly responsive and nurturing ways that most benefit children. It appears that specialized training in child development—regardless of formal education— supports more sensitive and responsive care—giving among teachers and their young students (Burchinal, et al., 2002; Howes et al., 1992). This training may foster teacher’s ability to provide consistent emotional support to preschoolers through responsive, caring behavior with regular feedback, teacher skills associ— ated with the most gains for children (Pianta et al., in press; Peisner-Feinberg et al., 2006). Financing Preschool Expansion The funding of preschool expansion could be conducted through the state education budget, as proposed by the governor. Preschool—as most fami- lies know—is expensive and targeting its expansion to particular children and communities may make it not only more feasible, but make it a viable proposal. With limited resources, some propose that preschool should be expanded and offered to families on a slid— ing scale, subsidizing only those families who are unable to pay. This is how Georgia—the nation’s pioneer in providing universal preschool—began its program. Texas has aggressively expanded preschool and still requires better-off families to pay fees. The money could be focused on extending the preschool care and services offered to children from disadvan— taged families who stand to gain the most. Other states with more extensive preschool pro- grams have taken different approaches, although most have adopted policies that address the neediest children first and/or offer them more resources. The universal preschool program in Georgia offers preschool on a sliding scale. Oklahoma provides higher per child re— imbursement for children with special needs and Eng— lish-language learners which, in effect, provides more resources to lower—income preschool programs. Including middle- and upper-income families in preschool program expansion may facilitate the development of high—quality preschool programs, but creative solutions are needed to avoid disenfranchis- ing low-income families. Otherwise, this well-mean— ing program could exacerbate the very problem it is EXPANDING AND lMPROVlNG PRESCHOOL designed to address—the existing achievement gap between low—income and other students. Policymakers should not forget that state and federal agencies already spend over $3.7 billion an- nually on public child care and preschool, including California’s share of the federal Head Start program (Fuller, Bridges, and Livas, 2005; LAO, 2005). On top of this maze of programs, several counties are now building “universal” preschool programs. Vari— ous child—care groups might come to see their shared interest in consolidating—and simplifying points of entry for parents—this array of funding streams as they seek additional funding. Who Should Run Preschools? Debate persists about where preschool should be housed. The public school system may hold appeal in terms of centralizing and coordinating education for children, regardless of their age, offering economies of scale, and having an infrastructure for employing the necessary staff. Yet many community—based organiza— tions have effectively developed preschool programs over the past century throughout California. About 70% of founyear-olds presently attending preschool attend a community program, not a preschool situ- ated in the public schools. Moreover, three-fifths of all California preschoolers attend a center that’s fully supported through parental fees, not public funding (PACE, 2006). Most research indicates that children appear to reap commensurate benefits regardless of where a preschool program is housed or who runs it—wheth— er it is a school district or a nonprofit. For example, in Georgia’s UPK program, children attending non- school district programs actually performed better than their peers participating in school—district pre— schools (Henry and Gordon, 2006). In New Jersey, researchers found no significant differences between children’s preschool gains, based on where the pro- gram was housed (Resnick et al., 2002). The key to designating the location of preschool programs may lie in establishing quality standards that support the development of a mixed system of quality preschools that reach the communities they intend to serve. Such a system could provide the flex— ibility that many parents need, while not sacrificing the benefits preschools can offer children. PACE a 141 ADDRESSING LONG-TERM ISSUES While the benefit of preschool for all children is no- table, reducing the achievement gap is fundamental. Preschool can only reduce the achievement gap if more low—income children participate in high—quality preschool programs, or if they receive greater benefits than middle—income children. While the differential benefits of preschool across income groups may bear out—with low-income children exhibiting greater cognitive gains than their more affluent peers—there are other ways to improve the school readiness and performance of low—income children. Some children might be enrolled in kindergarten one year later, giving them a chance to mature cognitively, socially, and emotionally. Preschool programs could be offered to three—year—olds or extended to full—day programs to give children a longer “exposure” to preschool. Extend— ing programs during the summer may provide more substantial benefits to children, providing a preschool experience that is consistent, moderate, and sustained. RECOMMENDATIONS Based on the most current research, we offer the following recommendations for implemen— tation of preschool expansion in California: I Simplify the existing $3.7 billion in current programs. One approach would be for Sacra— mento to fund one parent voucher effort and one high«quality preschool program. Regulato- ry relief around the federal Head Start program could allow further integration of these centers into a single state—coordinated network. I Target public dollars first on those children who we know benefit most from quality child care and preschool: youngsters from low— income families. At the same time, we must recognize that thousands of lower middle— class families confront the most scarce availability of affordable preschool centers, and need assistance in participating, too. I Extend care to low-income three-year-olds for long-term exposure. Provide full-day pro- grams with longer hours for families who need them. Develop new preschools in the neigh— borhoods where low-income families live. I Conduct additional research to identify what preschool inputs, teacher qualities, and classroom practices are most effective in ad- vancing quality and child development across diverse populations. I Develop preschool programs that truly respect and involve parents in defining the goals of local child care options, including preschools. Pre—literacy skills and the under~ standing of mathematical concepts are essen- tial in terms of success in elementary school. But the broader motivation and eager desire to learn in group settings require inventive, developmentally appropriate classroom prac- tices with well—prepared teachers. Early edu~ cators must also think more carefully about the continuity that children do, or fail to, ex— perience between their homes and preschools. I Provide better training for diverse, qualified teachers. Recent evidence points to the utility of requiring child—care providers to obtain a two— year degree in child development. The state, if serious about improving quality, must expand training incentives and more flexible college programs, building on the lessons learned in earlier First 5 California training experiments. I Develop sustained and adequate financing of child care and preschool options, with a stronger and carefully crafted role for Sac— ramento. Including preschool expansion in wider legislation to reform the financing of the K- 12 school system is advisable. At the same time, the state department of education has learned much about how to fund both community— and school—based preschool programs. The integration of myriad pro— grams can be aided by county—level agencies, but the legislature and governor could lead CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 this consolidation and simplification effort, moving toward a single child care voucher program and a single preschool center program. REFERENCES August, D., 8< Hakuta, K., Eds. (1997). Improving Schooling for Language Minority Children: A Research Agenda. Washington, DC: National Academy Press. Barnett, S. (1995). Long—term effects of early childhood programs on cognitive and school outcomes. The Future ofChildren, 5, 25—50. Barnett, W. S., Hustedt, 1. T., Robin, K. B., 8t Schulman, K. L. (2005). The state of preschool: 2005 state preschool yearbook. The National Institute for Early Education Research. Bates, 1. E., Marvinney, D., Kelly, T., Dodge, K. A., Bennett, D. S., 81 Pettit, G. S. (1994). Child care history and kin- dergarten adjustment. Developmental Psychology, 30, 690-700. Belsky, 1. (2001). Developmental risks (still) associated with early child care. Journal of Child Psychology and Psychiatry and Allied Disciplines, 42, 845-859. Burchinal, M. R., Campbell, F. A., Bryant, D. B., Wasik, B. H., & Ramey, C. T. (1997). Early intervention and mediating processes in cognitive performance of chil— dren of low—income African—American families. Child Development, 68 (5), 935-954. Burchinal, M.R., 8r Nelson, L. (2000). Early Childhood Research Quarterly,. 15, 385412. Bridges, M., Fuller, B. C., Rumberger, R., 8: Tran, L. (2004). Preschool for California’s children: Promising benefits, unequal access. PACE Policy Brief, 04-3, Berkeley, CA: Policy Analysis for California Education (PACE). Brooks—Gunn, 1. (2003). Do you believe in magic? What we can expect from early childhood intervention programs. SRCD Policy Briefi 17. Campbell, F.A., 8: Ramey, CT. (1994). Effects of early in— tervention on intellectual and academic achievement: A follow—up study of children from low—income families. Child Development 65, 684-698. Campbell, F. A., & Ramey, C. T. (1995). Cognitive and school outcomes for high—risk African-American student at middle adolescence: Positive effects of early interven- tion. American Educational Research Journal, 32, 743-772. Campbell, F. A., Ramey, C. T., Pungello, E. P., Miller— Iohnson, 5., 8r Sparling, 1. 1. (2002). Early childhood education: Young adult outcomes from the Abecedarian Project. Applied Developmental Science, 6, 42-57. EXPANDING AND lMPROVlNG PRESCHOOL Cooper, S., & Dukakis, K. (2004). Kids can’t wait to learn: Achieving voluntary preschool for all in California. Oakland: Preschool for All. Currie, 1. (2001). Early childhood intervention programs: What do we know? Journal of Economic Perspectives, 15, 213-238. Currie, 1., St Thomas, D. (2000). School quality and the longer—term effects of Head Start. Journal of Human Resources, 35, 755-774. Currie, 1., 8( Thomas, D. (2001). Early test scores, socio- economic status, school quality and future outcomes. Research in Labor Economics, 20, 103—132. Early, D.M., Bryant, D.M., Pianta, R.C., Clifford, R.M., Burchinal, M.R., Ritchie, S., Howes, C., 8: Barbarin, O. (2006). Are teacher’s education, major, and credentials re- lated to classroom quality and children’s academic gains in pre—kindergarten? Early Childhood Research Quarterly, 2 I , 174- 195. Fryer, R., & Levitt, S (2004). “Understanding the Black- White Test Score Gap in the First Two Years of School,” The Review of Economics and Statistics, May. Fuller, B., with Bridges, M., & Pai, S. (in press) Standardized Childhood: The Political and Cultural Struggle over Early Education. Stanford University Press. Fuller, B., & Huang, D. (2003). Targeting investments for universal preschool: Which families to serve first? Who will respond? PACE Working Paper, 03—1, Berkeley, California: Policy Analysis for California Education (PACE). Fuller, B., Livas, A., 8( Bridges, M. (2005). How to expand and improve preschool in California: Ideals, evidence, and policy options. PACE Working Paper 05—1. Berkeley, California: Policy Analysis for California Education (PACE). Garces, E., Thomas, D., & Currie, 1. (2002). Longer term effects of Head Start. The American Economic Review, 92, 999— 1012. Hamre, B., & Bridges, M. (in review). Early care and educa— tion staff preparation, quality, and child development: A review of the literature. Hamre, B., Grove, R., O’Brien, B.C., Lu, F., Bridges, M., & Pai, S. (2004). Program Evaluation: Matching Fundsfor Retention Incentives for Early Care and Education Stafif: Final Report: 2001 —2004. Berkeley, CA: University of California, Policy Analysis for California Education. Heckman, 1ames (2000). “Policies to Foster Human Capital,” Research in Economics, 54(1), 3—56 Heckman,1.,lchimura,H.,& Todd, P. (1998), ‘Matching as an econometric evaluation estimator’, Review ofEconomic Studies 65, 261—294. Hirschberg, D. (2002). CDSS—PACE Child care planning project—Final report. Berkeley: University of California, Policy Analysis for California Education (PACE). PACE I] 143 144 Hofferth, S., Shauman, K., Henke, R., & West, I. (1995). Characteristics of Children’s Early Care and Education Programs: Data from the 1995 National Household Education Survey. Washington, DC: National Center for Education Statistics. Howes, C. (1997). Children’s experiences in center-based child care as a function of teacher background and adult: child ratio. Merrill-Palmer Quarterly, 43, 404—425. Iencks, C., 8( Phillips, M. (1998). The Black-White Test Score Gap. Washington, DC: Brookings Institution. Kisker, E., Hofferth, 8., Phillips, D., & Farquhar, E. (1991 ). Profile ofChild Care Settings: Early Education and Care in 1990. Washington, DC: United States Department of Education. Kisker, E. & Maynard, R. (1991). Quality, Cost, and Parental Choice in Child Care.’ In Blau, David (ed) The Economics of Child Care. New York: Russell Sage Foundation. Liang, X., Fuller, B., & Singer, 1. (2000). Ethnic differences in child care selection: The influence of family structure, parenting practices, and home language. Early Childhood Research Quarterly, 15, 357-384. Loeb, S., Bridges, M., Bassok, D., Fuller, B. C., & Rumberger, R. (2006). How much is too much? The effects of dura— tion and intensity of child—care experiences on children’s social and cognitive development. Economics of Education Review. Loeb, S., Fuller, B., Kagan, S. L., & Carrol, B. (2004). Child care in poor communities: Early learning effects of type, quality, and stability. Child Development, 75, 47-65. Magnuson, K. A., Ruhm, C. 1., & Waldfogel, I. (2004). Does pre-kindergarten improve school preparation and per— formance? National Bureau ofEconomic Research Working Paper 10452. Cambridge, MA: NBER. National Center for Educational Statistics (1998). User’s Manualfor the ECLS-K Base Year Public— Use Data. Washington, D. C.: US Department of Education, Office of Educational Research and Improvement. NICHD Early Child Care Research Network. (2003). Does the amount of time spent in child care predict socioemo- tional adjustment during the transition to kindergarten? Child Development, 74, 976— 1005. O’Brien-Strain, M., Moye, L., 8( Sonenstein, F. L. (2003). Arranging and Payingfor Child Care. San Francisco, California: Public Policy Institute of California. Reardon, S. (2003). Sources of educational inequality: The growth of racial/ethnic and socioeconomic test score gaps in kindergarten and first grade. State College, PA: Penn State University (Manuscript). Reynolds, A. 1., Temple, J. A., Robertson, D. L, & Mann, E. A. (2001 ). Long-term effects of an early childhood interven- tion on education achievement and juvenile arrest: A 15 year follow—up of low—income children in public schools. Journal ofthe American Medical Association, 285, 2339- 2346. Ruhm, C. I. (2004). How well do parents with young chil- dren combine work and family life. In S. M. Bianchi, L. M. Casper, and R. B. King (eds.), Work, Family, Health, and Well-Being. Mahwah, NI: Lawrence Erlbaum Associates. Rumberger, R. W., & Arellano, B. ( 2003). Understanding and Addressing the Latino Achievement Gap in California. Draft Report submitted to the Latino Policy Institute, California Policy Research Center, University of California. Shonkoff, I. P., 8( Phillips, D. A., Eds. (2000). From Neurons to Neighborhoods: The Science of Early Child Development. Washington, DC: National Academy Press. Smith, T., Kleiner, A., Parsad, B., Farris, E., 8( Green, B. (2003). Pre-kindergarten in the US. Public Schools. Washington, DC: U. S. Department of Education, National Center for Education Statistics. Snow, C. E., Burns, M. S., & Griffin, P. (Eds). (1998). Preventing Reading Difi‘iculties in Young Children. Washington, DC : National Academy Press. U. S. Department of Education, National Center for Education Statistics. (2003). The Condition of Education. Washington, DC: US. Government Printing Office. Waldfogel, I. (2002). Child care, women’s employment, and child outcomes. ]ournal ofPopulation Economics, 15, 527- 548. Wertheimer, R., Croan, T., Moore, K. A., & Hair, E. C. (2003). Attending kindergarten already behind: A statisti- cal portrait of vulnerable young children. Child Trends Research Brief, 2003-20. Washington, DC: Child Trends CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 ALIFORNTA EDUCATORS AND POLICYMAKERS D ESERVE HEARTY ERAISE EOR WHAT THEY HAVE ACCOMPLISHED IN RECENT YEARS. AS OUR CONTRIBUTORS HAVE DETAILED, SACRAMENTO HAS ESTABLISHED CLEAR LEARNING OEIECTIVES, NOW ALICN ED WITH INSTRUCTIONAL MATERIALS AND STANDARDlZED TESTS. Despite the ups and downs of school funding, student learning has accelerated in elementary grades. Modest yet discernible responses to standards-based accountability have been felt in middle schools across the state, especially in students’ math performance. But earlier chapters also point to aspects of the state’s public education system where much work re— mains. Test score growth has faded over the past two years, achievement gaps between children from poor and middle—class families have failed to narrow, and almost one third of California teenagers will never receive a high school diploma (Greene & Winters, 2006). Sacramento has been eager to erect a high— standards, high—stakes educational system. But state policymakers have shown less determination to ad- dress the feeble structure of funding streams that props up the public schools. This cuts into the morale and efficacy of local school boards and principals who are struggling to serve an evolving mix of children and families. California is the 11['1 richest state in the nation in terms of per capita income; we rank 315‘ when it comes to spending per pupil. Perhaps the most crucial issue facing the Califor— nia schools is whether the current state reform Strat— egy will sustain the achievement growth we have seen since 1999 in the elementary grades and ever reach the high school. The pressing tasks in coming years include a careful rethinking of school finance and a more motivating approach to accountability. The timing couldn’t be better. The California economy has bounced back, notwithstanding the state’s structural deficit. A new gubernatorial term is getting under way, and a fresh legislature will soon begin work. The governor’s high—ranking education commission will be reporting out more than a dozen studies on efficiency and finance adequacy, along with a Sketch for long—term institutional reform. The Con- gress and President will be reviewing, and perhaps improving, the No Child Left Behind Act. 145 146 Regaining Momentum to Motivate Learning To rekindle the spirit and local efficacy of school reform, the analyses you have just read suggest that policymakers and educators might pursue four basic priorities: I Reorient school accountability and funding to improve schools that serve the lowest-performing students. i Develop a coherent strategy for raising the achievement of English learners. - Make school finance simple, transparent, and adequate for a high-standards school system. I Decentralize authority by building district ca— pacity, freeing principals to deploy resources to instruction, and focusing state responsibility on tracking and rewarding performance. The original policy theory behind standards— based reform—back in the 19808—was that states would articulate clear learning objectives and finance a fully professional teaching force. Then, local districts and school principals would be given greater freedom to redeploy resources to instructional improvement efforts. In California, however, policymakers micro— manage: limiting the textbooks and materials teachers can use; narrowing the curriculum; tying up almost one third of school spending on highly regulated cat- egorical aid. No private firm would centralize every operation and technology in this way—then expect that their best and brightest employees would remain in such an un—professionalized workplace. Our four—point reform agenda emphasizes a pair of conceptual thrusts. First, an accountability and fi— nance system that fails to close achievement gaps will make California’s society even less fair, by awarding opportunities only to some. Second, the state’s heavy reliance on rules, set in Sacramento (and Washing— ton), will fail to create motivating and professional workplaces for local educators. Tough love may yield short—term gains. But the long-term challenge is to attract and retain quality teachers. POLICY SHIFT 1 Reorient School Accountability and Finance to Boost Low-Performing Students Under Sacramento’s current reform program, achievement gaps have failed to narrow. This is a formula for creating a divided society, with a gaping rift that separates rich Californians from the poor. Policymakers have made attempts to fix the problem. Two intervention programs, described in prior chap- ters, have targeted dollars to low—performing schools. Initial evaluation results are not encouraging: districts often fail to allocate dollars to instructional improve- ments; this type of short-term aid is inadequate to attract (and retain) a stronger teaching force; and school site councils scatter dollars based on multiple demands. One structural problem—worsened by NCLB’s problematic way of determining which schools are deemed “failing”—-is that hundreds of schools and 152 districts are beginning to be hit by sanctions, as student growth targets don’t keep pace. Parents and policymakers do need to know which districts are progressing and which are not. But the current accountability systems are long on sanctions and short on the means to build district capacity. As our authors have suggested, improvement efforts would more likely take hold if interventions targeted a select number of low—performing schools and stretched funding and technical assistance over a longer period of time. We also see the need for more careful state—local strategies. This will require a thoughtful revision of NCLB in Washington to free Sacramento to focus more carefully. And, if we are serious about helping these schools, more resources will be needed, espe— cially to attract more able school managers and stron— ger teachers. Assembly Bill 172, backed by Governor Schwarzenegger, exemplifies legislation that targets interventions on children most in need. The law will focus new preschool dollars on communities with low—performing schools. Unlike the original PSAA policies, the Legislature has shown growing interest in progressively target— ing dollars on students most in need. In the summer of 2006 lawmakers— backed by major education groups—decided to focus an additional $2.9 billion CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 on low-performing schools.‘ It was a mistake to tie up most of the dollars in reducing class sizes—which yields achievement gains only under certain condi— tions. Alternative strategies should be tried and their effectiveness compared.2 But the progressive focus on schools serving low-performing students is an impor- tant precedent. Superintendent Iack O’Connell is right to lobby Washington to change NCLB so that schools are re- warded when they spark achievement growth. NCLB, as presently structured, unfairly penalized schools serving poor children, largely because they start toward the bottom of the staircase leading to “uni— versal proficiency,” mandated by NCLB. Schools in better-off communities started more than half way up the staircase, and the steps they must climb each year are less steep. When a public policy allows children in well—off families to pull farther and farther ahead of the most disadvantaged children, it discourages op- portunity and motivation. POLICY SHIFT 2 Develop a Coherent Strategy for Raising the Achievement of English Learners As Thomas Timar points out in Chapter 5, “The main policy question. . .is not how to fix low—performing schools, but what state policy can do for schools that serve large numbers of educationally and economical— ly disadvantaged students.” Remember that one in five California students lives in a low—income household. Closely related is the fact that more than a quar- ter of California’s students are English learners (1.6 million). Providing them with an education that meets their learning needs requires, at minimum, qualified teachers and an effective curriculum (in— cluding materials). Policymakers have advanced a number of com— mendable measures to support English learners. The California English Language Development Test (CELDT) was first administered in 2001—02; Eng- lish language development standards were adopted in 1999; and California teachers may pursue two supplemental credentials certifying expertise in EL instruction. Rekindling Reform However, research from the field, including ex- tensive teacher surveys, reveal a chasm between what our current teacher workforce can provide and what English learners need to reach proficiency targets. As the authors of Chapter 7 point out, policies aimed at better preparing teachers for diverse learners must consider recruitment strategies and the limited capac— ity of teacher training colleges. While results of the CELDT show increasing numbers of English learners are achieving fluency, these results do not correlate to success on the Cali— fornia Standards Test in English Language Arts. Fur— thermore, research affirms that “one size does not fit all” in educating our diverse student population. This year, in a heated debate over instructional materi— als for English learners (SB 1769), lawmakers argued over requiring textbook publishers to provide supple— mentary material for English learners while giving districts choice in purchasing materials. A recent study by the American Institutes for Research finds that districts benefit from more options in addressing their EL needs.3 The bill was vetoed on grounds that it would further “segregate” Californians. We encourage policies that support our schools in supplementing core curriculum to meet the needs of language learners, in building teacher capacity to reach California’s unique student body, and in track- ing demographic shifts. Students living in poverty and those learning English are not isolated to urban cen— ters. We must also study the effects of the California High School Exit Exam on student motivation and drop out rates among English learners and students living in poverty. Earlier chapters detailed the widening linguis- tic diversity of California’s children and how many enter kindergarten already behind in their cognitive and language development. Rising public interest in expanding preschool opportunities —especially for youngsters from poor and blue—collar families—is one encouraging sign. More broadly, the principle here is to enrich young children’s learning environ- ments, and this necessarily involves adequate teacher training, more effective pedagogical practices, and more respectful, more engaging relationships between parents and educators over time. PACE fl 148 POLICY SHIFT 3 Make School Finance Simple, Transparent, and Adequate for a High-Standards School System It‘s certainly cheaper to push local educators to work harder or teach with a simpler curriculum than to fi— nance stronger teacher salaries, attract stronger rookie teachers, or enrich instructional materials. Again, the fundamental policy dilemma is that Sacramento has succeeded in creating a high-standards, high-stakes accountability system while failing to provide suffi— cient resources to ensure that every classroom delivers an equal opportunity to learn. Early next year the governor’s Committee on Ed— ucation Excellence will deliver an array of new studies that delve into aspects of school finance, the efficiency of public schools, and improving the ways and means by which more than $50 billion is spent on K—12 edu- cation each year. We are hesitant to put forward spe— cific recommendations in this arena beyond urging policymakers to carefully examine the reports. The analyses offered in this volume, however, do suggest two key principals might guide school finance reform. California’s method for financing educa- tion and distributing dollars to districts, and then to schools, should become fully transparent and simpler. There‘s a not—so—funny joke in Sacramento that only about five people in the state capital fully understand the thick statutes that structure the Byzantine funding system. In contrast, states like Oklahoma and Texas have fairly simple allocation formulae, weighting stu— dents by specified needs and associated costs. A curi— ous citizen or journalist can quickly learn how dollars flow to local districts. In California, even veteran legislators don’t un— derstand how school finance works. This places them at the mercy of special interests that want to tinker to a get a bit more equalization aid or a few more text— books. Public support could only grow for a reformed finance system that was easily explained and which seemed more fair. Progress is being made to be able to track the learning patterns of students over time, even when they move from one school to another.4 This is an analytic piece to a fundamental puzzle: figuring out which education investments actually yield growth in achievement. Despite the billions of dollars spent each year for Sacramento’s more than 60 categorical aid programs, no way exists to evaluate the relative benefits of these proliferating programs. Even after fighting to consolidate these funding channels into simpler block grants, say, for low-performing students or for teacher development, there are lawmakers newly advancing a bevy of specialized programs. Here again, silver bullets are cast in Sacramento, yielding unknown effects for students and further restrict— ing the capacity and flexibility of local educators and elected school boards. POLICY SHIFT 4 Focus the State’s Responsibility, Deregulate with Accountability Local educators continue to be pressed to raise stu- dent achievement with diminishing control over fungible resources inside schools. Sacramento should revisit the original model of school accountability borrowed from innovative firms in the 1980s. A grow- ing number of companies were pushing to focus cen— tral managers on defining quality and character of the end product, then granting local managers and pro— fessionals more discretion to figure out how to apply resources to boost performance. This new view of management which emphasizes centrally determined outcomes but locally crafted remedies provided the cornerstone of what came to be called systemic or standards—based reform (Cohen, 1990; Smith & O’Day, 1991; Fuller et al., 2006). But somewhere along the way Sacramento decided to regulate student outcomes and the inputs and teach— ing tools that local educators are now required to use. And the rules around accountability were layered on top of the rules that come with categorical aid programs. So, urban school principals are forced to become bookkeepers and rule keepers, rather than instructional leaders. And for all the rhetoric from Sacramento about aiding districts in the quest to strategically strengthen teaching and learning, the proliferation of rule—bound categorical aid has again increased in recent years. Sacramento is beginning to focus on the capac— ity of local districts to address achievement shortfalls. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 This is an encouraging trend. We applaud the state department of education’s initial strategy for helping districts with growing numbers of low—performing schools.5 School districts and county offices of educa— tion play a key role in leveraging reforms to improve student achievement. District capacity varies when it comes to interpreting and implementing policy. A new PACE report details how three diverse districts are effectively closing achievement gaps, crafting quite different strategies (Wood, et al., 2006). Researchers at the American Institutes for Research are leading a new district collaborative, conven— ing state policymakers, researchers, and innovative district leaders. The legislature also is focusing on district capacity, for example, through AB 2109, enacted this year, which requires districts receiving professional development funds to focus those learn- ing opportunities on closing achievement gaps. Many school districts have successfully closed disparities in the share of children who reach proficiency in read— ing and mathematics. For instance, 52% of California districts, reporting complete data between 2002 and 2006, did narrow the disparity in the share of Latino versus white fourth—graders deemed proficient in lan— guage arts, closing the gap by almost three percentage points on average. School principals have lost considerable au— thority over resources. Yet centralization has failed to close achievement gaps and may even threaten the everyday commitment and motivation of local educators. Under the original model of state—led ac— countability, principals were to gain more authority, so that fungible resources could be deployed based on a grassroots understanding of the barriers to instruc- tional improvement. Simple and transparent school finance mechanisms could move dollars to principals in accountable ways. All the regulations and rules surrounding categorical aid has created thousands of jobs in district offices with little evidence of effec— tiveness inside schools. A better way can be found to free-up principals to lead with more controllable re- sources, while still holding them accountable to raise student performance. Careful financing of innovative forms of school— ing is advisable, along with more rigorous evalua- tion of what’s working and for which students. Early results from charter schools are mixed yet positive overall. The RAND evaluation found that students in Rekindling Reform charter schools keep pace with pupils in neighbor- ing public schools at a lower cost. The jury is still out on “small schools.” Little is known about the rising public investment in career academies and other in- novative structures within high schools. But govern- ment should continue to experiment, adequately fund these innovations, and then rigorously assess their comparative performance. The Risk of Complacency One thing that Sacramento policymakers should not do is rest on their laurels. It’s natural that poli— cymakers who fought hard to create the present ac— countability system would continue to defend it. Other equally forceful groups continue to nibble on the edges of these policies, hoping to reduce the amount of student testing or lower the bar for what we are expecting students to master. But California’s present accountability system is yielding diminishing gains. We shouldn’t throw out the baby. But the bath water is looking quite murky. The analysis contained in early chapters, on bal— ance, suggests that the basic contours of Sacramento’s accountability program are serving students well. But it would be unfortunate if the architects of the Public School Accountability Act remained stuck in 1999. The challenge is to build from recent policy success. The real and pressing risks are that achievement gaps will continue to grow, achievement will plateau, and high school students will not benefit from reform, jeopardizing California’s future. In this context, the defenders of current policy might come together with its critics in the new legis— lative session—seeking to craft a thinner set of cen- tralized rules without lowering expectations or stan- dards. Both sides might ask what policy conditions could better attract bright, idealistic college graduates into workplaces that are becoming more professional, less hog tied from above. How we adjust the balance between rules and resources to enrich the work lives of teachers is a pressing issue. Tough love may have its place in politics. But nurturing the motivation and efficacy of teachers and students alike is the long—term challenge facing policymakers. A system that’s now long on rules and short on resources is unlikely to sustain motivating reform. PACE a 149 150 ENDNOTES SB1133 was signed after an adjustment to funding. See class size reduction in Chapter 1. The status of the teaching profession 2005, published by the Center for Teaching and Learning, details the continuing shortage of quality teachers. Further reductions in class size will require an even greater supply. 3 Parish, T., Linquanti, R., & Merickel, A. (2006) Eflects of the implementation of Proposition 227 on the education of English Learners. Menlo Park: American Institutes for Research. The California Longitudinal Pupil Achievement Data System (CALPADS) is scheduled to be in place by 2008. The California Department of Education’s improvement guidelines, the Nine Essential Program Components, and its District Intervention and Assistant Team (DAIT) program are examples of how Sacramento can provide leadership in partnering with districts to strengthen Iow- performing schools. REFERENCES Center For Teaching and Learning. (2006). The status of the teaching profession 2005. Santa Cruz, CA: Center for Teaching and Learning. Civil Rights Projects. (2005). Confronting the graduation rate crisis in California. Cambridge, MA: Harvard University. Online at www.civilrightsproject.harvard.edu. Cohen, M. (1990). Key issues confronting state policymak— ers. In R. Elmore (Ed.), Restructuring schools: The next generation ofeducational reform (pp. 251—288). San Francisco: Iossey Bass. Fuller, B., Gesicki, K., Kang, E., & Wright, I. (2006). Is the No Child Left Behind Act Working? The reliability ofhow states track achievement. PACE Working Paper Series No. 06— 1. Berkeley, CA: Policy Analysis for California Education (PACE). Gandara, P., 8r Maxwell-Jolly, 1., 8: Driscoll, A. (2005). Listening to teachers of English learners: A survey of California teacher challenges, experiences, and professional development needs. Santa Cruz, CA: Center for Teaching and Learning. Greene, I, 8x Winters, M. (2006). Leaving boys behind: Public high school graduation rates. Civic Report No. 48. New York, NY: Manhattan Institute for Policy Research. Smith, M., St O’Day, I. (1991). Systemic school reform. In S. Fuhrman & B. Malen. (Eds), The politics ofcurriculum and testing: The 1990 yearbook ofthe politics ofeducation association (pp. 233-267). Woody, E., Bae, S., Park, 8., & Russell, 1. (2006). Snapshots of reform: District efforts to raise student achievement across diverse communities. Berkeley, CA: Policy Analysis for California Education (PACE). CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 CATHERINE BITTER is a Senior Research Analyst at the American Institutes for Research in Palo Alto, California. Ms. Bitter served as the Project Manager and Project Director respectively for the independent Evaluation Studies of the Immediate Intervention/Underperforming Schools Program (II/USP) of the Public Schools Accountability Act (PSAA) of 1999 for the California Department of Education, completed in 2003 and 2005. She is also assisting with AIR’s independent evaluation of California’s High Priority Schools Grant Program (HPSGP). MARGARET BRIDGES is the research director of Child Development Projects at Policy Analysis for California Education (PACE). Currently, she is examining the implications of expanding preschool access, and assessing the ramifications for Latino preschool children and their families. Dr. Bridges completed her doctoral and postdoctoral training in child development, studying family, school, and neighborhood effects on children’s outcomes. JAY G. CHAMBERS is a senior research fellow and a managing director of the Education Finance Business Devel— opment Group at the American Institutes for Research (AIR). Dr. Chambers is an expert in school finance and edu- cation cost analysis. Over the past 30 years, he has directed and been involved in many national-, state—, and local studies related to programmatic and resource costs in education, and has written and conducted projects with local, state, and federal partners to improve the quality of fiscal and cost information for school decision making. BRUCE FULLER is professor of education and public policy at the University of California, Berkeley. Trained in sociology and education policy at Stanford University, Fuller helped to staff the Assembly Education Commit- tee and later served as an education adviser to Governor Jerry Brown. He worked as a research sociologist at the World Bank before teaching at Harvard University. He currently co-directs Policy Analysis for California Education (PACE). His new book, Standardized Childhood: The Political and Cultural Contest over Early Education, will be published by Stanford University Press in 2007. PATRICIA GANDARA has directed education research in the California Legislature (1985—87), and has been a Commissioner for Postsecondary Education in California (1981—86). She currently is Professor of Education at the University of California, Davis where she is also Director of CAP—Ed (Center for Applied Policy in Education) and Associate Director of the UC Linguistic Minority Research Institute. She has authored and co—authored a study of 5300 teaches in CA Listening to the Voice of Teachers of English Learners, as well as English Learners in California Schools, Unequal Resources, Unequal Outcomes and Learning English in California: Guideposts for the Nation. HALEH HATAMI is a writer and has conducted education policy research and provided technical assistance to schools in Oregon, North Carolina, and California. She has worked on local, state, and federal school improvement initiatives, including school needs assessments, and has co-authored articles on school—based budgeting, leadership coaching, and improvement planning. At PACE she coordinates a seminar series for legislators and staff on a range of education policy topics. She served as managing editor for Crucial Issues 2006. MELISSA HENNE served as Assistant Research Scientist at Policy Analysis for California Education (PACE) where she worked in the Elementary/ Secondary Education division. Henne has been involved in several projects, includ— ing case studies of effective districts and educators’ responses to state accountability. Prior to coming to PACE, Henne was a project coordinator at the New York State Education Department, were she assisted in the implemen- tation of the School Quality Review Initiative (SQRI). Previously, she taught eight-grade English and Social Studies in New York City. She is currently studying at University of California at Berkeley Graduate School of Education. "151* 152 JESSE D. LEVIN is a Senior Research Scientist for the Education Program at the American Institutes for Research (AIR). He conducts research on educational production, finance and adequacy. He has served as Deputy Project Di- rector and Principal Analyst on the New York Adequacy Study. Currently, he serves as Project Director for The Cost of Comprehensive High School Reform. JULIE MAXWELL-JOLLY began her career as a Spanish bilingual teacher. Since leaving the classroom more than twenty years ago she has continued to work on English learner and immigrant student issues. She was the first Lin- guistic Minority Research Institute’s post-doctoral fellow in 2001 and currently works as a Senior Research in the UC Davis School of Education. Most recently she has participated in conducting and designing research on the prepa- ration of teachers for working with English language learners, the experiences of Latino students in the California Community College system, and an in-depth district level study of programs and outcomes for English learner students. JENNIFER O'DAY is a Managing Research Scientist and policy analyst in the Education Program of the American Institutes for Research. Dr. O’Day has carried out research and written extensively in the areas of systemic standards- based reform, educational equity, and capacity—building strategies. Dr. O’Day’s has served as Principal Investigator/ Project Director for a series of national, state, and district studies, covering such topics as Title I accountability under both IASA and NCLB, the implementation and effects of California’s Public School Accountability Act of 1999, and district interventions in low performing schools in Chicago and San Diego. Dr. O’Day is currently leading a four—year investigation of the effects of site-based leadership and professional learning opportunities on instructional practices and student achievement in San Diego City Schools. LAWRENCE O. PICUS is a Professor at the University of Southern California Rossier School of Education. He also serves as the director of the Center for Research in Education Finance (CREF), a school finance research center housed at the Rossier School of Education. His current research interests focus on adequacy and equity in school fi— nance as well as efficiency and productivity in the provision of educational programs for K-12 school children. Picus is past—president of the American Education Finance Association. In addition to authoring numerous publications on the subject, he has consulted extensively on school finance issues with more than 20 states, and conducted adequacy studies in Arkansas, Arizona, Kentucky, Wyoming, and Washington. LESLIE SANTEE SISKIN is a Research Professor at NYU’s Steinhardt School of Education, who conducts work in the sociology of organizations and organizational change with particular focus on high schools. Working with the Consortium for Policy Research in Education, she has been a Principal Investigator for two recent national studies of high school reform efforts: Accountability and the High Schools and Comprehensive School Reform Designs in Second- ary Schools. She has also published numerous articles and three books, including Realms of Knowledge, The Subjects in Question, and The New Accountability. THOMAS TIMAR is Professor of Education Policy in the School of Education at UC Davis, Associate Director of the Center for Applied Policy in Education, and Senior Associate, PACE. His areas of specialization are education finance and governance, education policy and politics. ELISABETH WOODY was Principal Research Scientist at PACE, where she directed public school accountability and effective schools projects. She has spent much of her career examining the impact of public school policies and re— forms on students’ and teachers’ daily lives. Her current work focuses on school and district efforts to address inequi— ties in achievement, and an analysis of national trends in achievement by gender, across race, socioeconomic status, and language. EEEEEEEEEEEEEEEEE FIGURE 1: California—Percentage of Fourth—Graders Proficient or Above in Reading and Math, According to State and National Assessment of Educational Progress (NAEP) testing Reading 100 a 90 s g: 80 g a 70 :3 E’ 60 E’ .0.) .‘2 LE 50 e O O at 40 a ‘f 30 ‘5 a 20 a 0. C1. 10 o waueco’xeosorxwgcs Math 6’) 'L ’5 v 6) (o a Q) o) o N "L 0) o) 0) ave goo fags? -0— Fourth grade results, SAT—9 (percentage children above the national norm) Fourth grade results, CST English-Language Arts (percentage children proficient or above) --A-— Fourth grade results, federal NAEP tests (percentage children proficient or above) Source: National Center for Education Statistics for a rather unprofessional workplace. Almost one— third of all California teenagers never graduate from high school (Greene 8( Winters, 2006). Regaining Momentum We must figure out how to rekindle earlier prog- ress. Californians will back a school system that’s showing results and helping to equalize opportunities for its graduates. One first step is to understand what has worked and to scrap policies that are not working. This new edition of Crucial Issues in California Educa— tion identifies which reforms have proven to be effec- tive and provides guidelines for how we can regain our momentum. Overall, the chapters in this volume stem from the assumption that we must build from what’s work— ing, while being candid about the flagging effects of current school accountability and finance systems. The chapters that follow, authored by leading researchers with a keen interest in California edu- cation, tackle these crucial issues facing educators and policymakers. Pulled together by PACE, these analysts detail the moving parts of California’s ac- countability and finance systems, along with their interplay with the federal government’s parallel ac— countability system, and implications for the state’s diverse students. One crucial issue facing policymakers is whether state and federal policies are thoughtfully balancing rules with resources. A second pressing question is why Sacramento’s reform strategy is failing to close achievement gaps, failing to help make our society more just (Figure 2). Much of the reform rhetoric since the late 19905 has emphasized a top—down, tough love approach to schools—that a clearer curric- ulum, prescribed textbooks, transparent achievement data, and stiff sanctions aimed at local educators would motivate stronger performance. This strategy has yielded important results. But our contributors question whether rules and punish- ment alone will continue to motivate local educa— tors. Nurturing growth, a spirit of cooperation, and respect for local context are required as well. This edition of Crucial Issues points to how policymakers can rethink their reliance on rules and sanctions, and on the proper balance with providing local educators adequate and aligned resources. The Evolving Policy Context Before moving into particular pieces of Califor— nia’s school reform puzzle, this chapter describes the overall landscape, the context and terrain in which improvements to accountability and finance may be pursued. This policy context is ever evolving, with political and popular demand for change often 2 CRUCIALISSUESIN CALIFORNIA EDUCATION 2006 out—stripping local educators’ capacity to carefully implement all that’s thrown at them. This context includes four key elements described in this chapter: I The evolving demographics of California’s plu— ralistic society, one that continues to manifest inequality of opportunity and rising numbers of non—English speaking children. I The standards—based accountability system that has coalesced since 1999, under the Public Schools Accountability Act (PSAA), complicated by the parallel No Child Left Behind (NCLB) ac- countability program. I The unreliable flow of dollars into local schools and how these dollars often fail to reinforce in- structional improvements. I Multiple nodes of public authority and agencies that attempt to govern and improve local schools. A policy’s efficacy in raising the motivation of students or teachers, even raising test scores, must be judged within this often-turbulent context. Future re— forms will gain credibility or fade away based on their capacity to fit these surrounding forces and advance improvements locally. PACE’s previous edition of Crucial Issues, re— leased in 2000, asked whether the panoply of reform policies, many enacted the previous year, were fitting together to advance a coherent set of learning aims, implemented within a sensible governance system. Since that time, Sacramento has successfully put in place several of these puzzle pieces, the core ele- ments of an aligned system: clear learning standards, instructional materials that cover standards, and stu— dent assessments that match up. What remains missing from the puzzle are an equally robust and aligned finance system and a clear strategy for improving the everyday workplaces in which teachers labor. Pressing more rules and pre— scribed pedagogies, alone, may prove to erode moti— vation inside schools, not enhance it. Let’s start with the evolving context in which contemporary reforms are attempted and new initia- tives will be situated. Today’s crucial issues in educa— tion stem from California’s unique history, one that includes remarkable demographic and economic change, along with recurring efforts to improve our schools. V‘Riiles' and Resources : . ’ the? EvolvingContéxt for School Reform FIGURE 2: 8th Grade Achievement Gap, by Economic Status English Language Arts CST /" Percent Proficient and Above 2003 2004 2005 2006 Algebra 1 CST Percent Proficient and Above 2003 2004 2005 2006 -l '— Middle-class, affluent students + Economically disadvantaged students Source: California Standardized Testing and Reporting (STAR) Program THE ORIGINS OF STRONG STATE ACCOUNTABILITY The challenges faced by our local schools find their roots in a half—century of demographic change and recurring efforts at school reform.2 The 1960s and 19705 spurred a variety of state initiatives aimed at equalizing the quality of schools serving California’s communities, be they rich or poor. In the wake of the civil rights movement, Sacramento created its own Early Childhood Education program (paralleling fed- eral Head Start), the Miller—Unruh reading effort, and a pioneering bilingual education initiative. PACE W 3 12 Governor's Advisory Committee on Education Excellence The committee will focus on four issues: The distribution and adequacy of education funding; The functioning and effectiveness of governance structures; Teacher recruitment and training; and, The preparation and retention of school administrators. Source: Office of the Governor P-16 Council State Superintendent Jack O’Connell formed a 52- member action group covering preschool to higher education to: I Improve student achievement at all grade levels, including eliminating the K-1 2 achievement gap; I Link all education levels, from preschool through higher education, to create a comprehensive, seam- less system of student learning; I Ensure all students have access to caring and quali— fied teachers; and, I Increase public awareness of the link between an educated citizenry and a healthy economy. Source: California Department of Education CHAPTER SUMMARIES CHAPTER 2, Funding California’s Schools, Part1: Past, Present, and Future?, lays out the complex compo- nents of and past fixes to the state’s education finance system. According to national reports, California con— sistently ranks low in its effort to fund schools. The author reveals what state—to—state comparisons mea- sure, how much California spends on its schools, and what it buys. A history of the state’s school finance in recent years focuses on issues of equity, productivity and adequacy and how the Williams case fits in, set— ting the context for the debates over adequacy in the subsequent Funding California’s Schools, Part1]. CHAPTER 3, Funding California’s Schools, Part1]: Resource Adequacy and Efficiency, explores the broad and complex question of“how much" we need to spend to get the achievement results we want. How do we align dollars to Sacramento’s learning standards? The Williams Settlement established the state’s re— sponsibility for providing minimal levels of qualified teachers, facilities, and textbooks. But what must be spent to offer truly sufficient opportunities to learn? How much is “adequate”? The authors sketch the im- plications and challenges put forth by the adequacy movement for the state of California, describing dif- ferent approaches to defining adequacy and how each addresses common problems in traditional school funding systems. Adequacy studies, they argue, move beyond offering a single figure or formula. They clar- ify the tangled strands of money flow and governance and offer policy options. CHAPTER 4, California’s Accountability System, outlines several key issues confronting the state as it continues to refine its school improvement initia- tives. As accountability in California encompasses a wide range of reforms at the federal, state, and local levels, this chapter focuses primarily on the evolu— tion of the state accountability system and its current interaction with the federal NCLB. The overlap of the two systems, and the lack of promising findings on the benefits of intervention programs like II/USP, forces California to rethink its accountability system. Authors show how the “theory” plays out in practice and offer concrete recommendations at a time when policymakers have a window of opportunity to nego— tiate positive changes to the current model. CHAPTER 5, Evaluating State Intervention: The High Priority Schools Grant Program, examines state ef- forts to improve instruction in the lowest performing schools—those receiving HPSGP funds. The chapter assesses past strategies to address the problems of persistent low achievement in schools, a problem that is most acute among schools that serve large numbers of non—English—speaking, minority students from disadvantaged socio-economic backgrounds. It then discusses preliminary findings from an ongoing study of schools in the HPSG program. The paper identifies some of the strengths and weaknesses in the program and makes recommendations for its improvement. CHAPTER 6, Achievement and Attainment: The Com— prehensive High School and the Problem of Reform, questions how we frame the current “crisis” in our high schools. What makes high schools so different and high school reform so difficult? The problem, the author argues, is an inconsistency of a different kind: for the first time in the history of the American high school, the aspirations of the public (more students CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 - Rules and Resources : . The Evolving Context for School Reform The Birth of a Reform: Lessons from Class Size Reduction The story of California’s "class size reduction" in the late 1990’s offers a valuable lesson on policy in practice. Budget surplus in the late 19905 allowed sweeping changes to class size, creating a ripple effect across the state's education infrastructure. Senate Bill 1777, under Governor Pete Wilson, legislated class size reduction (CSR) in K-3 classrooms statewide. Six years ago, Crucial Issues in California Education 2000 predicted the impact. While hailed as one of the most popular policies in years, the effects of CSR on student achievement was inconclusive by 2002* With the increase in the number of classrooms, schools required a massive infusion of teachers at grades K-3. The number of uncredentialed teachers rose 10.7 percent in the second year of implementation. The rulemaking implications of CSR’s rapid implementation resound today. The policy "lessons learned” can be applied to current efforts to reform California’s schools through the standards and accountability system, including: I Analysis of state capacity to support statewide reform must take place prior to implementation. I Sustainability of reform depends on a predictable and efficient flow of financial and human resources. What we have learned about the state's in— tervention in low-performing schools illustrates the unsustainable nature of high-cost/short-term initiatives (Chap— ters 4 and 5). I Implementation varies with demo- graphic context. Schools with a major— ity of high— poverty and English Learner students had less success in imple- menting improvement plans. * CSR Research Consortium, (2002). wanting higher levels of education) are in increasingly direct conflict with the expectations of the policy system (higher standards and test scores). This is a new set of demands for high schools—something they were not designed to do: to educate all young people through to graduation, to ensure that they all take rigorous courses and are prepared for college. The final three chapters of Crucial Issues 2006 look at the ways accountability and finance interact with California’s specific demographic context, with a focus on language, poverty, and early intervention. CHAPTER 7, Crucial Issues in Preparing Teachers of English Learners, makes the case for recruiting more and better—prepared teachers with specialized skills for teaching English learners (ELs)—students whose first language is not English, and who are not yet English proficient. California has the highest concentration of ELs in the nation and the numbers continue to grow. The authors review and critique current approaches to preparing teachers to meet the academic needs of these students. They report on what teachers them- selves said in a statewide survey about their current level of preparation and professional development. Among other urgent concerns for this population: without effective instruction, many ELs will fail the high school exit exam and fail to graduate. CHAPTER 8, Addressing the Needs of Low-Income Students, while recognizing recent state and federal policy attention on our most economically challenged students, points to a persistent achievement gap. The socio—economic implications of these dispari- ties include diminishing school participation, fewer high school graduates, and a less-educated work force. Authors tease out the link between poverty and achievement and policies aimed at compensating for economic disadvantage. Included in these fixes are programs like federal Title I funding. Finally, authors turn to the classroom and examine research on school and classroom strategies for addressing the needs of students in California’s low—income communities. CHAPTER 9, Expanding and Improving Preschool, brings the book to its conclusion—at the earliest stages of schooling. Research has shown that gaps in learning appear between different groups of children in their early language and cognitive development prior to entering kindergarten. This gap does not close, and in fact, widens for some over 12 years of public schooling. One strategy for improving school readiness and reducing initial achievement differences among children is increasing access to preschool. Yet poor and minority children are less likely to par— ticipate in preschool, which may contribute to the education achievement gap. This may be particularly problematic in California because of its large and PACE & 13 10% of total K—12 revenues in 1979—80. In the two decades that followed, the share of federal revenues has decreased slightly, with state revenues outpacing local revenues by up to 5% over most of the period. Today, the state plays the largest role, on average, in school funding, providing about half of the revenues, while local taxes contribute about 43%. The federal government provides the remaining 6—8%, a pro— portionate contribution which has remained in this range for the past 20 or 30 years. While California has followed the national trend of an increasing state and decreasing local responsibility in school funding, the California experience has been far more pronounced. Figure 2 illustrates and contrasts the federal, state, and local revenue trends of the nation as a whole and California from 1974—75 through 2001—02. While the federal share of total California revenues has for the most part followed that of the nation, state and local revenues have differed dramatically. For instance, the largest difference between the state and local shares of total US educational rev- enue over the 28-year period was a mere 6.3% (in 2000—01) compared to a 47.7% state/local difference for California (in 1987—88). The well—known ex- planation for the exaggerated trend experienced by California was the notorious court case of Serrano v. Priest and subsequent passing of Proposition 13, the latter of which, by capping property tax rates, effectively limited the amount of local tax revenue that could be collected. The end result was a major FIGURE 1: US Public School Income From Local, State and Federal Sources, 1929-2002 90% ' I Rapid Increase/ Decrease . . Focus on Ade u c A1 in State/Local Roles I FOCUS on Equnty ! q a V 80% I l 70% g i I I Standards-Based Reform (NCLB) and State-Set Goals 600/ I I <1) 0 I I E’ a I CK 50% i ' -—‘7 T9 :2 I | Q— l o 400/ — e O l l (U a l l 30% I I 20% I Rapid Increase in l I Federal Role I 10% l O%_ l l l I I I T I | I IQI b‘l [Cb] I I [4)] [<0] I/\I [0,] IN! [’5] | I I’\I lql Ix] I43I [<0] I/\l lo)l IKI Q Q v Q) "L Co r\ fix , ,V ,V 50 50 f0 ,‘o f0 A f\ f\ f\ f\ 93 Sb 9) 93 99 ,0) 9 .0) ,0) ,0) ,0 NO) 49 v49 6‘ <0“ <30 0?) <33 <8 «9 a” a” «‘0 «Cb Q9 <23” 9;“ 950 6233 <59 03* 0;“ 0§0 0) oQ Year “-0- Federal Share of US Total State Share of US Total + Local Share of US Total Notes: Local Revenues Include a relatively small amount from nongovernmental private sources (gifts and tuition and transportation fees from patrons), Source: US. Department of Education, National Center for Education Statistics, Biennial Survey of Education in the United States, 1919-20 through 1955— 56; Statistics of State School Systems, 1957-58 through 1969—70; Revenues and Expenditures for Public Elementary and Secondary Education, 1970-71 through 1986-87; and The NCES Common Core of Data (CCD), ”National Public Education Financial Survey,” 1987—88 through 2001-02 28 CRUCIALISSUESIN CALIFORNIA EDUCATION 2006 shift of support for school finance from local to state revenue sources or, as stated in Downes (1992): The primary effect of Serrano II and Proposi— tion 13 was thus to create what was effectively a state—financed system of public education. Clearly, with the growing role of the states in school finance from the 19605 to the 19705, the way in which states distributed revenues to local school sys— tems became more significant and drew more atten— tion. The growing state role was largely an outgrowth of the realization of the interconnectedness between local communities and the potential impact of the quality of education in one community on the labor market and well-being of its neighbors. Prior to the recent adequacy movement, the amount of money devoted by states to education and the mechanisms used to distribute funds to local school districts was, for the most part, a result of political bargaining and had little to do with student needs. With the growing state role, the focus of the funding debate in this new era of school finance turned to equity or, more precisely, to what the school finance literature referred to as horizontal equity. Under this concept, students and taxpayers across districts should be treated similarly with respect to the resources they receive and taxes they are required to pay, respectively (see Coons et al., 1970). Under this kind of framework, the equitable provision of FIGURE 2: US and California Public School Income From Local, State and Federal Sources, 1974—75 to 1999-00 80% 70% E1— [3 1:] [:1 E] E] El El El E] D D D State Share California '3 . 60% D g 43.13. 3.11.8. D D g A A A A E State Share 550% w ’==\ ________—’=§~ > , & Local Share _ / \ __ 1' (U ‘5 40%—-l:t—13 El 1- D A “<5 A A Loci Share California (1) g 30% A A A A MA. m A A A A A A A A A 20% Federal Share California 10%“ O O 7 Federal Share 0% l l l | l l l l l l l T l l l l | l l l | l l l l l l 8‘0’\Q>CBQ'\(L’5bCbQ'\’L43V°)b’\‘b<>)QNW «Va «6A 401‘ «m «<2? «(25% <23st (99> <29?) (8)9: (va2: (8)9: (8052» 999’ Q3399 99>?) 09,0; ($9 099 (£39 0);) 0&9 039 09,0) 0339 qqsb 0Q, 9&9 Year + Federal Share of US Total —- — State Share of US Total + Local Share of US Total 0 Federal Share of California Total El State Share of California Total A Local Share of- California Total Notes: Local Revenues Include a relatively small amount from nongovernmental private sources (gifts and tuition and transportation fees from patrons). Source: US. Department of Education, National Center for Education Statistics, Biennial Survey of Education in the United States, 191 9-20 through 1955— 56; Statistics of State School Systems, 1957—58 through 1969—70; Revenues and Expenditures for Public Elementary and Secondary Education, 1970—71 through 1986—87; and The NCES Common Core of Data (CCD), "National Public Education Financial Survey,"1987—88 through 2001 -02 PACE 29 36 learning disabilities who account for about half of the population of students with disabilities (Hanushek et al., 2002). There have even been studies showing the benefits of higher spending (e.g., Card and Krueger, 1992; Ferguson, 1991; Hedges et al., 1994; Betts, 1996; and, Greenwald et al., 1996). But there have been other studies that reveal inconsistent effects of par- ticular variables and raise the specter of ambiguity in what works (e.g. Coleman, 1966; and, Hanushek, 1986; 1996). In conclusion, the current research on educa- tional effectiveness has not come far enough that it can tell us definitively what practices and resources are most effective in promoting student success. Moreover, the current research is hardly exhaustive in terms of the types of student outcomes that have been addressed. This is not to say that research does not provide some basis for analyzing educational costs. As will be seen below, the educational effectiveness litera— ture serves as a key element in one of the four main approaches to modeling adequacy. However, with the narrow set of student outcomes used in most educa— tion research and the ambiguity in the literature as to what works in schools, it would seem that focusing strictly on educational research is not likely to pro— vide a clear, unambiguous answer on how to design a school funding system. California's Specific Need for Adequacy Research Putting forth the argument that money does not matter is not a credible contention. At the same time, to say we could not do better with what we have does not make much sense either. But it seems clear from the data that we, as a state, could do significantly better for our children. One thing is certain: the California schools that children attend today are not the same schools children attended in the 19505 and early 19605. The recent report by the Rand Corpora- tion (Carroll et al., 2005) and funded by the Hewlett Foundation shows California ranking very low among the 50 states in just about every dimension of educa— tion that one could imagine: real spending levels, stu— dent achievement scores, NAEP test score results, high school graduation rates, drop—out rates, and the rate of teen pregnancies. Stephen Carroll, the lead author of the report, states: A lot of people have expressed concern about the state of K—12 education in Califor- nia. We found that those concerns are well placed. California schools are lagging behind most other states and these findings suggest policymakers need to make major changes in order to repair the problems. Despite some improvements, the state has a long way to go to reclaim its standing as a national leader in K—12 education.23 In 2006 California enrolls more than one in eight of the nation’s children (12.8%), almost half (45%) of whom are Hispanic, one in eight of whom are Asian and Pacific Islander ( 12%), and one in twelve of whom are black (8%). About one of every five children in California lives in a family whose income is below federally established poverty thresholds. Almost four of every ten families (39.7%) headed by a single mother also lives in poverty, and almost one in every three (29.6%) of children live in high-poverty neighborhoods. In the early to mid-19705, California spent about the same share of its personal income on public education as the rest of the country—about 4.5%—but this has declined to about 3.6% today.24 APPROACHES TO MEASURING ADEQUACY Where then do we turn to measure the cost of an adequate education? How do we define adequacy in school funding? How do we ensure that all students have access to similar educational opportunities? How do we determine what additional resources might be required to address the special needs of certain student populations? How do we ensure that com- munities in different parts of a state have a reasonable access to schooling services? Be it a researcher, practitioner, parent, policy— maker, or other member of the general public who has looked at resource data or had the opportunity to personally visit school campuses in any state, he or she is sure to have observed vast differences in the access to resources (e.g., teachers, educational special- ists, textbooks, computers, classroom facilities, audi- toriums, or playgrounds) enjoyed by different com— munities. This wide variation in access to resources CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 has been well documented in earlier studies (Carroll et al., 2005; Chambers et al., 1993, 2000). Four alternative approaches have emerged to es— timate the amount of the investment required to offer an adequate education: 1) Professional Judgment, 2) Successful Schools, 3) Evidence-Based Approach (also known as the Expert Judgment), and 4) Cost Func— tion Analysis. All of these approaches have one thing in common: in line with the concept of vertical equity, they all begin from the premise that school funding formulas should be based on the cost of meeting the needs of the students served. As mentioned above, while this notion seems an obvious starting point for thinking about school funding today, two decades ago such decisions were simply based on revenues avail— able, with little or no systematic assessment of goals and objectives and the cost of achieving them. Professional Judgment Model (PJM) Drs. Jay G. Chambers and Thomas B. Parrish of the American Institutes for Research (AIR) first pio— neered systematic approaches to addressing adequacy and equity in school funding more than 20 years ago in two major school finance reform studies in Illinois (1982) and Alaska (1984). With the notions of de— veloping cost—based school funding, the researchers adapted the Resource Cost Model (RCM) as a method for estimating the cost of an “appropriate” educa- tion for all children. The RCM had been previously referred to as the ingredients approach, and was used by Levin (1983) for cost—effectiveness analysis and by Hartman (1979) to project the cost of implementing the Education for All Handicapped Children’s Act (predecessor of today’s Individuals with Disabilities Education Act, or IDEA) originally passed in 1975. In the Illinois and Alaska studies of the early 19805, Chambers and Parrish developed an educator-based decision-making process along with the RCM, and that methodology has evolved into what is currently referred to in the literature as the professional judg- ment approach. While the early model of Chambers and Par— rish was a more input-based model, applications of the Professional Judgment Model (PJM) in the past decade have been output-based. The model has evolved in light of the standards-based reform move- ment and the emphasis placed on accountability. The original application of the PJM uses panels of F NDJNGtALlFO-RNJA’SSCHOOLS,PARTll:F ‘VRESOIIRCE Aesop/icy Am) EFFlClENCY . educators to design and develop resource specifica- tions of delivery systems for educational services that will provide an adequate school program.25 Different research teams have applied this model using various methods to select panelists and involve non-education stakeholder panels. In addition, different assumptions have been used about what is included or excluded from the panel modeling exercises (i.e., specification of district versus school-level resources). However, in most applications the panels have included a wide variety of educators, including superintendents, prin— cipals, school business officers, and teachers. One of the most recent applications of the PJM by Chambers et al. (2004) involved designing a base set of resources or prototype for a school with a spe— cific demographic configuration of students, and then modifying the design to fit varying compositions of students.26 This approach not only provides a total cost estimate for an adequate education, but it shows how this cost varies in relation to pupil needs. Adding supporting cost studies on geographic cost differences and other cost factors for central ad- ministration and maintenance and operations services, one can develop statewide projections of costs as well as projections by school and by district based on the varying demographic con— figurations of students. Some critics have suggested that the professional judgment approach can lead to rather high cost estimates for adequacy due to the relative freedom of the panels to specify resources without being bound to a budget constraint (Taylor, Baker, and Vedlitz, 2005; Hanushek, 2005).27 To address this potential MEASURING ADEQUACY: Professional Judgment Model Successful Schools Model Evidence-Based Approach criticism, the original approach I Cost Function developed by Chambers and . Parrish incorporated checks and AnalySIS balances by including stakehold- er anels of non-educators re — . p p I Multiple Model/ resenting various policy constit- uencies (e.g., local school board members, taxpayers, parents, Hybrid Approach PACE FF] 37 and the program failed to deliver the hoped for improvements in student outcomes.32 The bottom line is that changes in funding need to be phased into the system to provide state and local officials with sufficient time to plan on how best to allocate new dollars or cut old programs, to meet existing standards, and to improve the quality of instructional programs. Adequacy as a Policy Process It is important to recognize that “costing out” adequacy in education is not a science. It is a process of organizing the best data available to arrive at a reasonable estimation of what it might cost to provide an “adequate” educa- tion. To be sure, sophisticated analytic tools can be brought to bear upon the process, but the estimation of the costs of an “adequate” education is more of a quest than an end- point on a journey. Thus, it is inappropriate for courts or policymakers to seize upon any particular estimate as the only one worthy of being “adequate.” There remains a role, an appropriate zone of local discretion, for policymakers to weave their judgments into the amalgam of costing out facts and research outcomes. These studies of adequacy will arrive at al— ternative instructional designs and models by which schools can achieve adequacy or excellence in education. Researchers will then be able to impute costs to these instructional models. However, it would be a mistake to translate whatever instructional model might support the cost analyses into a mandate for local districts and schools to spend revenues. However insightful these instructional de— signs may seem, and however persuasive a case is made for their effectiveness, education continues to be more of an art than a science. Thus, there is a need to harness continu— ally the creativity and experience of local educators in any state in order to achieve the higher levels of student outcomes that virtu- ally everyone desires. Because local educa— tors remain central to achieving educational adequacy, they must be given a wide swath of discretion to determine how money should be used in order to achieve a wide range of stu- dent outcomes, including higher achievement test performance, improved attendance rates, higher graduation rates, increased participa- tion in extra curricular activities or student government, and decreased drop-out rates. In many instances, adequacy studies are conducted to provide a comprehensive cost estimate for achieving educational adequacy within a state. To do this, one has to conduct the analysis by ignoring current revenue streams from the federal, state, and local governments. Here, the goal is to determine the total cost for a given outcome level with- out regard to where the funds might come from to support this effort. What this means is that analysts must address how these cost estimates will later interact with existing federal funding programs such as Title I, the Individuals with Disabilities Education Act (IDEA), and other funding streams. Analysts will also need to take into account the fact that many of these federal programs do not supplant existing funding streams, but instead require that states and local school districts contribute their fair share to supplement the awarded funds. Schools Can’t Do it Alone The success of schools also depends on other individuals and institutions to provide the health and nutritional services, the intel— lectual stimulus, and family supports upon which public school systems can build. Schools cannot and do not perform their role in a vacuum, and this is an important quali— fication for all conclusions reached in any study of adequacy in education. As Rothstein (2004) points out: CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 I Closing the gaps (between the achievement of lower—class and middle—class children) re- quires not only better schools, although those are certainly needed, but also reform in the social and economic institutions that pres- ently prepare children to learn in radically different ways. It will not be cheap. I Raising the achievement of lower~class children requires abandoning the illusion that school reform alone can save us from having to make the difficult economic and political decisions that the goal of equality inevitably entails. However, even with the narrow focus on schools themselves, the types of adequacy studies described above offer no guarantees of improved performance. The success of schools depends not only on providing the necessary resources to achieve the results, but on the effective use of those resources by local school systems. This means developing ap— propriate incentive and accountability struc— tures to increase the likelihood that schools will use resources productively. Current Studies of School Finance Adequacy in California With respect to furthering the investigation of educational adequacy in the California public school system, we are in the midst of an excit— ing and historic time. Four foundations are currently funding a suite of studies to address school finance reform in the state of Califor— nia. The goal is not only to provide alternative estimates of the costs of adequacy, but also to help increase the likelihood that whatever resources reach the schools will have the de- sired impact.” The collection of studies being done for the State of California includes the following: I defining adequacy I costing out studies using alternative ap— proaches ‘ NDING turtdamksscnoms, PART u: 'RESOURCEADEQUACYAND EFFICIENCY teacher cost differences studies of English language learners lessons from other states alternative sources of funds structures of governance contracts and teacher assignment fiscal substitution across non-education government spending business office personnel and policies state teacher policy I data availability and transparency. Few states have the opportunity or the fiscal resources to conduct a collection of studies encompassing all of the components listed above. While each of the studies will be done by separate teams of researchers, the founda— tions have provided for a centralized struc- ture to coordinate and mange the distribution of research funds for the studies, and will be responsible for synthesizing the results of these studies and then disseminating them to policymakers. Moreover, the studies are being conducted at the request of various constituencies of the state government, but at the same time are under the control of the foundations. This provides for a significant amount of independence, shields the studies from political pressure, and increases the like— lihood that the findings will be objective. In most states, adequacy cost studies are conducted quite independently of other types of studies, and minimal thought is given to some of the other significant issues such as governance, which is so critical to success when implementing policy changes. Moreover, the kinds of resources required to conduct thorough costing out studies are substantial. While it is rare for states to invest even as much as $500,000 in educational studies, a significantly larger investment is required to complete more comprehensive studies with a broader set of dimensions, public engagement, and various checks and balances built into the analysis. PACE $2 43 48 schools in the United States. Journal of Political Economy, 100(1), 1-40. Carroll, S. I., Krop, C., Arkes, I., Morrison, P. A. 8( Flanagan, A. (2005). California K—12 public schools: How are they doing? RAND Research Report. Santa Monica, CA: RAND. Chambers, I. G. (1978). Educational cost differentials and the allocation of state aid for elementary/secondary edu- cation. The Journal ofHuman Resources, 13(4). Chambers, I. G. (1980). The development of a cost of edu- cation index: Some empirical estimates and policy issues. Journal ofEducation Finance, 5(3), 262—281. Chambers, I. G. (1981a). Cost and price level adjustments to state aid for education: A theoretical and empirical re- view. In Forbis Iordan, K. (Ed.), Perspectives in state school supportprograms, second annualyearbook of the American Educational Finance Association. Cambridge: Ballinger Publishing Co. Chambers, I. G. (1981b). The hedonic wage technique as a tool for estimating the costs of school personnel: A theo— retical exposition with implications for empirical analy- sis. Journal ofEducation Finance, 6(3), 330—354. Chambers, I. G. & Fowler, WI (1995). Public school teacher cost differences across the United States (Report No. 95— 758). Washington, DC: US. Department of Education, National Center for Education Statistics. Chambers, I.G., Parrish, T.B., Levin, I.D., Smith, I.R., Guthrie, I.W., Seder, R.C. & Taylor, L. (2004, March). The New York Adequacy Study: Determining the cost ofprovid- ing all children in New York an adequate education —final report. Palo Alto, CA: American Institutes for Research. Chambers, I., I. Levin, & Parrish, T. (Forthcoming, 2006). Examining the relationship between educational out- comes and gaps in funding: An extension of the New York Adequacy Study. Peabody Journal of Education. Chambers, I. G. 8r Parrish, T. (l982a). The issue ofadequacy in thefinancing ofpublic education: How much is enough? (Report No. 82—Al9). Stanford, CA: Stanford University, Institute for Research on Educational Finance and Governance. Chambers, I. G. & Parrish, T. (1982b). The development ofa resource cost modelfunding basefor education finance in Illinois, Volume I B executive summary, and Volume II B technical report). Stanford, CA: Associates for Education Finance and Planning. Prepared for the Illinois State Board of Education. Chambers, I. G. & Parrish, T. (1983). Adequacy and equity in state schoolfinance and planning: A resource cost model approach. Stanford, CA: Stanford University, Institute for Research on Educational Finance and Governance. Chambers, I. G. & Parrish, T. (1984). The development ofa program cost model and a cost—of-education indexfor the state ofAlaska: Final report. Volumes I—IV. Stanford, CA: Associates for Education Finance and Planning, Inc. Chambers, I.G. 8r Parrish, T. (1986a). The AEFP/RCM docu- mentation for Version 86.4 (prepared to support users of the AEFP/RCM System Computer Simulation Programs). Mountain View, CA: Associates for Education Finance and Planning, Inc. Chambers, I. & Parrish, T. (1986b). The RCM as a decision making process. Prepared for the Institute for Research on Educational Finance and Governance under a grant from the National Institute of Education. Chambers, I.G. 8( Parrish, T. (1985a). The RCM: A resource management and program budgeting approach for state and local educational agencies. Stanford, CA: Stanford University, Institute for Research on Educational Finance and Governance. Chambers, I.G. & Parrish, T. (1985b). The RCM system: Plans for marketing and dissemination. Stanford, CA: Stanford University, Institute for Research on Educational Finance and Governance. Chambers, I., Parrish, T., Levin, I., Guthrie, I., Smith, I. & Seder, R. ( 2004). The New York Adequacy Study: Determining the cost of providing all children in new york an adequate education. American Institutes for Research (AIR)/Management Analysis and Planning (MAP). Prepared for the Campaign for Fiscal Equity (CFE). Chambers, I.G., Goertz, M., Marder, C., Padilla, C. & Parrish, T. (1993). Translating dollars into services: Chapter 1 resources in the context ofstate and local re- sources for education. Palo Alto, CA: American Institutes for Research. Prepared for the US. Department of Education. Chambers, I.G., Stullich, S., Lieberman, I., Parrish, T., Kaleba, D. & Van Campen, I. (2000). Study ofeduca- tion resources and federal funding: Final report, executive summary, and technical appendix. Washington, DC: US. Department of Education, Office of the Under Secretary. Clune, W. H. (1994). The shift from equity to adequacy in school finance. Educational Policy, 8(4), 376-394. Coleman, I. (1966). Equality ofeducational opportunity, Prepared for US Department of Health, Education, and Welfare, Office of Education, Washington DC. Coons, I. E., Clune, W. H., 8( Sugarman, S. D. (1970). Private wealth and public education, Cambridge, MA: Harvard University Press. Darling—Hammond, L. (2004). Inequality and the right to learn: Access to qualified teachers in California’s public schools. Teachers College Record, 1936-1966. Downes, T. A. (1992). Evaluating the impact of school fi- nance reform on the provision of public education: The California case. National Tax Journal, 45(4), 405—419. Duncombe, W. D. & Yinger, I.M. (1999). Performance stan— dards and education cost indexes: you can‘t have one without the other. In Ladd, H.F., Chalk, R. 8( Hansen, I.S. (Eds), Equity and adequacy in educationfinance: Issues and perspectives (pp. 260—297). Washington, DC: National Academy Press. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Duncombe, W. D., Lukemeyer, A. & Yinger, I. (2003). Financing an adequate education: A cost study of New York. In Fowler, W.F. Ir. (Ed.), Developments in school finance 2001 -02 (pp. 127-154). Washington, DC: U.S. Department of Education. Evans, W. N. & Schwab, R. M. (1995). Finishing high school and starting college: Do Catholic schools make a differ- ence? Quarterly Journal of Economics, 110, 941-974. Ferguson, R. F. (1991). Paying for public education: New evidence on how and why money matters. Harvard Journal on Legislation, 28 (2), 465—498. Firm, 1. D. & Achilles, C. (1990). Answers and questions about class size: A statewide experiment. American Educational Research Journal, 27(3), 557—77. Greenwald, R., Hedges, L.V. & Laine, R. D. (1996). The effect of school resources on student achievement. Review of Educational Research, 66(3), 361—396. Grissmer, D. (1999). Class size effects: Assessing the evi- dence, its policy implications, and future research agenda. Educational Evaluation and Policy Analysis, 21, 231—248. Grubb, W. N. 8( Hyman, I. (1975). Constructing teacher cost indices: Methodological explorations with California uni— fied school district. In Selected Papers in School Finance, 1975. Washington, DC: U.S. Department of Health, Office of Education, and Welfare. Guthrie, J. W., Hayward, G. C., Smith, I. R., Rothstein, R., Bennett, R. W., Koppich, J. E., Bowman, E., DeLapp, L., Brandes, B. 81 Clark, S. (1997). A proposed cost— based block grant model for wyoming school finance. Sacramento, CA: Management Analysis and Planning Associates, L.L.C. Hanushek, E. A. (1986). The economics of schooling: Production and efficiency in public schools. Journal of Economic Literature, 49(3), 1141-1178. Hanushek, E. A. (1996). School resources and student performance. In Burtless, G. (Ed.), Does money matter? The ejfect of school resources on student achievement and adult success (pp. 43-73). Washington, DC: Brookings Institution. Hanushek, E. A. (2002). Teacher quality. In Izumi, L & Everson, W. (Eds.), Teacher quality (pp. 1—12). Stanford, CA: Hoover Institution Press, pp. 1—12. Hanushek, E. A. (2005, October). The alchemy of ‘costing out’ an adequate education. Paper prepared for Conference on Adequacy Lawsuits, Harvard University. Hanushek, E. A., Kain, I. F. & Rivkin, S. G. (2002). Inferring program effects for special populations: Does special ed— ucation raise achievement for students with disabilities? Review ofEconomics and Statistics, 84(4), 584-599. Henderson ,A. T. (1998). Parents are a school’s best friends. Phi Delta Kappan, 70, 148—153. Hedges, L. V., Laine, R. D. 8( Greenwald, R. (1994). Does money matter? A meta—analysis of studies of the ef— EUNDl-NG CALIFORNIA’S SCHOOLS, PART ll: RESOURCE ADEQUACY AND EFFICIENCY fects of differential school inputs on student outcomes. Educational Researcher, 23(3), 5-14. Iepsen, C., & Rivkin, S. (2002). Class size reduction, teacher quality, and academic achievement in California public elementary schools. San Francisco, CA: Public Policy Institute of California. Karoly, L. A. & Bigelow, I. H. (2005). The economics of invest- ing in universal preschool education in California. RAND Research Report. Santa Monica, CA: RAND. Koski, W. S. & Hillary A. W. (2004). What educational re- sources do students need to meet California’s educational content standards? A textual analysis of California’s educational content standards and their implications for basic educational conditions and resources. Teachers College Record, 106, 1907-1934. Kenny, L., Denslow, D. & Goffman, I. (1975). Determination of teacher cost differentials among school districts in the state of Florida. In Tron, E. (Ed.) Selected Papers in School Finance, 1975. Washington, DC: U.S. Department of Health, Education, and Welfare. Karoly, L., Greenwood, P., Everingham, S., Hoube, 1., Kilburn, M. R., Rydell, C.P., Sanders, M. & Chiesa, J. (1998). Investing in our children: What we know and don’t know about the costs and benefits ofearly childhood inter— ventions. Santa Monica, CA: RAND Corporation. Krueger, AB. (1997). Experimental estimates of education production functions. Quarterly Journal of Economics, 114(2), 497-532. Krueger, A.B. & Whitmore, D.M. (2001). The effect of at- tending a small class in the early grades on college test taking and middle school test results: evidence from proj— ect STAR. The Economic Journal, 111, 1—28. Lee, V.E., Smith, J.B. & Croninger, RC. (1997). How high school organization influences the equitable distribu— tion of learning in mathematics and science. Sociology of Education, 70, 128—150. Lee, VB. and Smith, J. (1997). High school size: Which works best, and for whom? Educational Evaluation and Policy Analysis, 19, 205-228. Miles, K.H., Odden, A., Archibald, S. 8( Fermanich, M. (2002). An analysis of professional development spend- ing in four districts using a new cost framework. Working Paper, Madison: University of Wisconsin—Madison, Wisconsin Center for Educational Research, Consortium for Policy Research in Education. Mishel, L. St Rothstein, R., (Eds). (2002). The class size de— bate. Washington, DC: Economic Policy Institute. Murphy, J., Beck, L., Crawford, M., Hodges, A. & McGaughy, C. (2001). The productive high school: Creatingpersonal- ized academic communities. Thousand Oaks, CA: Corwin Press. McKinley, S. K. (2005a). The journey to adequacy: The DeRolph saga. Journal ofEducation Finance, 30 (3), 288— 312. PACE g 49 per pupil in California was more than $600 below the national average (Carroll et al., 2005). In addition, when adjusted for purchasing power, teacher salaries in California lag behind the other four most populous states (Carroll et al., 2005).22 Pupil-teacher ratios reflect these limited resources: despite the class-size reduction proposition passed in July 1996, California continued to have the second highest pupil—teacher ratio in the US. as of 1999—2000 (Carroll et al., 2005). In addition, the schools that are struggling the most tend to have the most challenging student populations and the fewest resources to address them. The correlation between school poverty and II/USP participation is clear, with an average school poverty rate in California of 50%, compared to an 80% poverty rate for II/USP schools. In addition, according to a recent study by the Center for the Future of Teaching and Learning, 21% of California teachers in the lowest achieving schools were considered underprepared and/or novice (in their first or second year of teach- ing) in 2004—05. In high—achieving schools, only 11% of teachers were given this designation. Similar differences were seen between schools serving high minority populations compared to schools with low proportions of minority students (Esch et al., 2005). District Influence on Capacity. An additional key aspect of capacity is the infrastructure to support and sustain improvement efforts. Both of the II/USP evaluations found a substantial district influence in all aspects of II/ USP implementation and impact. In particular, school and student outcome data indicated that achievement trends differ significantly and sys— tematically across the large urban districts, and these differences overshadowed any effects of II/ USP. For example, Figure 3 depicts the estimated reading scores among lower decile schools (Deciles 1-3) in the seven largest districts in California from 1999 through 2004. Large differences in growth patterns among districts can be seen. Qualitative data from the II/USP studies indi- cated that the districts’ influence on outcomes varied by the extent and nature of district actions and strate- gies. One area in which district actions were key was capacity building. For example, some districts played FIGURE 3: Estimated Reading Scores (SAT—9, CAT/6)a in Decile 1—3b Elementary Schools (Grades 2-5) from California’s Seven Largest Districts,c Controlling for Differences in School DemographicsOI n U m Standard Deviations from State Mean -O.2 .0 b 9'3 m .C'> 00 l _\ _\ N 4.4 -’|.6 District A District B District D . =..———i% ’4 + — + District C —-k— - e: ' DistrictE —0— District F 4 ~ ‘ ’ ‘ N §~ \ ’ - __ El — — ’ H \Q/b—J \o’//0 —0— District 0 A Note: The trends of Districts A, and B overlap perfectly. 1000 7000 7001 7007 700? 700A . Using SAT-9 scores from 1999 through 2002 and CAT/6 scores for 2003—2004. Scale scores are standardized against the state mean scale score in each year. . Based on API state rank of 1999 and 2000. . Including Fresno, Los Angeles, Long Beach, Oakland, Sacramento, San Diego, and San Francisco d. The analysis models control for differences among districts in school—level demographics (student ethnicity, percentage of students eligible for free/re— duced price lunch, percentage of English Learners). Source: CDE (star.cde.ca.gov), Official School-level Data 64 CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 an active role in selecting External Evaluators, in some cases narrowing the approved list of providers, and in other cases actually making the selection for each (or all) schools. In addition, some districts set up or re- quired supports during the implementation of II/USP, including external assistance, professional develop- ment, and monitoring (O’Day and Bitter, 2003; Bitter et al., 2005). Beyond direct II/USP participation, district central offices also affected the capacity of all low— performing schools in their jurisdiction through instructionally related policies. For example, some districts implemented a common specified approach to literacy instruction, frequent monitoring of stu- dent progress, or aligned professional development. Others implemented mandated curricular packages. And others targeted additional resources (both mon- etary and personnel) towards their lower—performing schools. Where the district took a strong role in curricu~ lum and instruction, it tended to overpower or direct any independent effects of II/USP. Some districts, however, lack either the capacity or will to provide such leadership or support. This variation can stem from the inequitable distribution of funds across dis- tricts, as well the fact that some districts have higher needs than others (e.g., high concentrations of Eng— lish Learners or students in poverty). The PSAA did not address this variation in district capacity, nor the role that districts can play in schools’ strategies for improvement. CALIFORNIA’S ACCOUNTABILITY SYSTEM MOVING FORWARD California’s accountability system has suc— ceeded in increasing the attention of educa— tors to student outcomes and low-perform- ing schools. In addition, it has established incentives and supports that aim to motivate educators and provide additional capacity to schools undertaking improvement efforts. However, findings above point to important limitations of the system. The limited effects of the intervention programs associated with 5 CALIFORNIA’S AccouNiAsiuiv SYSTEM. this system and the implementation of the federal NCLB now raise questions about what next steps California should take to make the system more effective and coherent. Based on the evidence to-date, we have iden- tified three major goals for strengthening California’s education accountability system: I Align the federal and state accountability systems: NCLB has instituted an additional layer of targets and criteria for interven— tions. Resolving differences between the two systems will be a key challenge in upcoming years. I Allocate sufficient additional resources and assistance for low-performing districts and schools, and take appropriate actions when improvement is not forthcoming: California has already established a system of interven- tions for low-performing schools that pro- vides resources, assistance, and consequences. However, the limited effect found from these programs raises questions about how to im- prove the design and implementation of this system. I Build a long-term infrastructure to sup- port the continuous improvement of schools throughout the state. Building a strong infra— structure will enable the continuous improve- ment of all schools across the state. GOAL 1: Align the federal and state accountability systems. The introduction of PSAA, including the API, succeeded in focusing the attention of educa— tors and the public on student achievement and low—performing schools. However, over— lapping targets and expectations more recently introduced through the federal accountability system have diffused this attention and in- creased confusion around targets and expected consequences. Schools are now being identified for interventions based on two sets of criteria that sometimes, but not always, overlap. PACE $2 65 100 REFERENCES Achieve, Inc. and the National Governors Association. (2005). An action agenda for improving America’s high schools. Washington, DC: National Governors Association Center for Best Practices. Asimov, N. (2006). San Francisco Chronicle Ian 7. http:// sfgate.com/cgi-bin/article.cgi?file=/c/a/2006/01/07/ MNGCRG]VOD1.DTL Camili, G., St Vargas, S. (2006). The legend ofthe large MCAS gains of 2000-2001 . Education Policy Analysis Archives 14(4). http://epaa.asu/epaa/v14n4/. Carnoy, M., Elmore, R., 8( Siskin, L. S. (2003). The new ac- countability: High schools and high—stakes testing. New York: Routledge. Carnevale, A., & Desrochers, D. (2003). Standards for what?: The economic roots ofK- l6 reform. Princeton, NJ: Educational Testing Service. Center for the Future of Teaching and Learning. (2005). The status of the teaching profession, 2005. Santa Cruz, CA: Author. Conant, I. B. (1959). The American high school today: Afirst report to interested citizens. New York: McGraw Hill. Education Trust—West. (2004). Are California high schools ready for the 21-" century? Oakland, CA.: Author. Education Trust-West. (2006). Too little thought, too little action: California’s teacher equity plan falls short. Oakland, CA: Author. Education Week.(2005). Quality Counts, 24 (17). Esch, C.E, Change-Ross, C.M., Tiffany-Morales, 1., & Shields, PM. (2004). California’s teaching force 2004: Key issues and trends. Santa Cruz, CA: The Center for the Future of Teaching and Learning. Finn, CE. (2005). Things are falling apart. Education Next. Palo Alto, CA: Hoover Institution. www.educationnext. org/20061/27.html. Haney, W. (2000). The myth of the Texas miracle in educa- tion. Educational Evaluation and Policy Analysis Archives. [http://epaa.asu.edu/epaa/v8n4I/ [8, 41]. Haycock, K., & Huang, S. (2001 ). Are today’s high school graduates ready? Thinking K—I6: Publication of the Education Trust, 5(1), 3—17. Hess, A., & Cytrynbaum, S. (2002). The effort to redesign Chicago high schools: Effects on schools and achieve- ment. In V. Lee (Ed), Research on high school reform ef— forts in Chicago. Chicago: Consortium on Chicago School Research. Ingersoll, R. M. (2005). Analysis for California’s hidden teacher spending gap: How state and district budgeting practices shortchange poor and minority students and their schools.” Oakland, CA : Education Trust-West. Kahne, 1., Sporte, S., 8( de la Torre, M. (2006) Small high schools on a larger scale: The first three years of the Chicago High School Redesign Initiative. Chicago: Consortium on Chicago School Research at the University of Chicago. Klein, S.P., Hamilton, L.S., McCaffrey, D., St Stecher, B. (2000). What do test scores in Texas tell us? Education Policy Analysis Archives, 8(49), 14. http://epaa.asu.edu/ epaa/v8n49 Landsberg, M. (2006, January 29). Back to basics: Why does high school fail so many? Los Angeles Times. Little, ].W. (2002). Locating learning in teachers’ com- munities of practice: Opening up problems of analysis in records of everyday practice. Teaching and Teacher Education, 18, 917-946. Little, J.W., 8t Dorph, R. (1998). Lessons about Comprehensive School Reform: California’s School Restructuring Demonstration Program. Berkeley, CA: SB 1274 School Restructuring Study. Loveless, T. (1999). The tracking wars: State reform meets school policy. Washington, DC: Brookings. Mazzeo, C. (2001). Frameworks of state: Assessment policy in historical perspective. Teachers College Record, 103(3), 367-397. McLaughlin M., & Talbert, I. (2001). Professional com- munities and the work ofhigh school teaching. Chicago: University of Chicago Press. Murnane, R., & Levy, F. (1996). Teaching the new basic skills. New York: The Free Press. Murray, L. (2005). Graduating college-ready, work-ready students. The Education Trust-West, Superintendent in residence report. www.EdTrustWest.org National Association of Secondary School Principals. (1996). Breaking ranks: Changing an American institution. Reston, VA: Author. National Commission on Excellence in Education. (1983). A nation at risk: The imperativefor educational reform. Washington, DC: Author. National Commission on Teaching and America’s Future. (2005). What matters most: TeachingforAmerica’sfitture. Washington DC: Author. Oberman, I. with C. Arbeit, C. Praglin, & Goldsteen, S. (2006). Challenged schools, remarkable results: Three les— sonsfrom California’s highest achieving high schools. San Francisco, CA: Springboard Research. Olson, L. (2001). A quiet crisis: Unprepared for high stakes. Education Week on the Web. [http://www.edweek.org/ April 18, 2001]. Presley, ].B., 8( Gong, Y. (2005). The demographics and academics of college readiness in Illinois. Edwardsville IL: Illinois Education Research Council. http://ierc__siue.edu Rhoades, K., 8( Madaus, G. (2003). Errors in standardized tests: A systemic problem. Boston: National Board on Educational Testing & Public Policy. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Roderick, M. (2006). Closing the aspirations-achievement gap: Implications for high school reform. New York: MDRC. Rumberger, R., & Gandara, P. (2004). Seeking equity in the education of California’s English learners. Teachers College Record, 106(10), 2032—2056. Schoenbach, R., Greenleaf, CL., Christine C., & Hurwitz, L. (2000). Reading for understanding: A guide to improv- ing reading in middle and high school classrooms. San Francisco: Iossey-Bass. Siskin, LS. (1994). Realms of knowledge: Academic depart- ments in secondary schools. London and New York: The Falmer Press. Siskin, L.S. (2003a). Colleagues and ‘Yutzesz’ Accountability inside schools. Voices in Urban Education. Spring 2003: 24—3 1 . Siskin, L.S. (2003b). Outside the core: Accountability in tested and untested subjects. In M. Carnoy, R. Elmore, St L. S. Siskin (Eds.). The new accountability: High schools and high-stakes testing (pp. 87-98). New York: Routledge. Siskin, L.S. (2003c). When an irresistible force meets an immovable object: Core lessons about high schools and accountability. In M. Carnoy, R. Elmore, 8r L. S. Siskin (Eds.). The new accountability: High schools and high— stakes testing (pp. 175- 194). New York: Routledge. Siskin, L.S., 8t Little, I. W. (1995). The subjects in question: Departmental organization and the high school. New York: Teachers College Press. Smith, M., 8( O’Day, I. (1991). Systemic school reform. In S. H. Fuhrman 8t B. Malen (Eds), The politics of curriculum and testing (pp. 233—267). New York: Palmer Press. Yun, I.T., & Moreno, LP. (2006). College access, K—12 con- centrated disadvantage, and the next 25 years of educa- tion research. Educational Researcher, 35(1), 12- 19. Achievement and Attainment: The Comprehensive High School and the Problem of Reform PACE u 1'01, 108 TABLE 1: Actual and Adjusted Gains, by Teacher Authorization; Grade 2, Selected Schools, LAUSD ‘ Language 5 , BCLAD 1.8 (n=142) 1.5 (n=142) 4.1 (n=148) 2.4 (n=148) CLAD/LBS f .2.D(n=32) 2.7 (n=32) 1.0(n=34) 0.4 (n=34) $31969 * * * * ALevel@ ‘ i 1.8 (@155) 1.6 (n=155) 0.3 (n=155) -1.5(n=155) No Authorization -2.4 (n=74) -2.9 (n=74) 0.5 (n-93) -1.8 (n=93) *Actual and adjusted gains were not reported here due to the small sample size. @ LAUSD certifies language competencies of its teachers if they do not already hold a BCLAD; A Level indicates fluent bilingual. Source: Hayes and Salazar (2001), page 36 Teacher Impact on Student Achievement Providing adequate numbers of teachers with the skills they need to work with English learners is a critical part of the formula for improving the achievement of these students. Although we cannot assign teachers either all the credit or all the blame for student achievement, teachers play a central role in students‘ education. This is particularly true for youth who are extra vulnerable, such as English learn- ers. Thus our discussion of teacher preparation for English learners warrants mention of the growing research on the critical impact that teachers have on student learning overall. Briefly, this research finds that teachers with good preparation and pedagogical skills (Haycock, 1998; 2001 ), a rigorous postsecondary education (Ferguson, 1991; Ferguson and Womack, 1993), and full certification in their field (Darling— Hammond and Youngs, 2002; Darling-Hammond, Chung, and Frelow, 2002; Darling—Hammond et al., 2005) make a significant difference in student achievement. Furthermore, these gains in student learning appear to be stable over time. The converse is true as well. It takes students a significant amount of time with good teachers to overcome the effects of one who is ineffective (Sanders and Rivers, 1996; Sanders and Horn, 1995). The evidence suggests that this is equally true for teachers of English learners. That is, teachers with a greater amount of specialized preparation in working with English learners are more effective in increasing student achievement among these students. As a case in point: a study was conducted in the Los Angeles Unified School District (LAUSD), where 20% ofthe state’s English learners attend school. It investigated the relationship between English learner student achievement gains and the credential held by the teachers who taught them. The study sample included 177 classrooms in 29 schools with large numbers of EL students (Hayes and Salazar, 2001). Researchers found that “state/ district authoriza- tion of teachers does have an impact on student out- come. For example, students of teachers holding no state or district authorization achieved largely nega- tive or very small positive...adjusted gains in reading and language” (pp. 37—38; Table 1). '2 A follow-up of 1‘“[ through 3rd grade classrooms in the same schools during the subsequent school year again found that “students of credentialed teachers outperformed students of emergency permitted teachers” (Hayes, Salazar, and Vukovic, 2002, p. 90). Unfortunately, research on the distribution of such highly qualified teachers in California indicates that the shortage of these teachers is particularly acute in the schools with the most EL students (Gandara and Rumberger, 2003; Esch et al, 2005). Teacher Certification for EL Classrooms English learners require significant expertise beyond that expected of teachers who do not have EL students in their classrooms (Gandara and Max— well-lolly, 2000; Gonzalez and Darling-Hammond, 2000; Darling—Hammond et al., 2001). This is true no matter what the classroom instructional program. In an attempt to provide teachers with this expertise, California has developed two supplemental creden— tials designed specifically for teachers of EL students. The credential most teachers have pursued has been the CLAD (Culture, Language, and Academic Devel— opment) credential, which focuses on EL teaching methods. It is granted to teachers who have taken a prescribed course of study, which usually includes a class in cultural diversity, another in first and second language acquisition, and a third in pedagogi- cal strategies for teaching EL students in English. CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 Teachers may also opt out of any of these require- ments by passing an examination demonstrating competency in these areas. Many California teachers actually hold a “CLAD equivalent” authorization (SB 1969)13 that requires far less preparation than the CLAD credential or certificate. Less common has been the BCLAD (Bilingual, Culture, Language, and Academic Development) cre- dential, which in addition to the CLAD coursework, includes a class on specific bilingual pedagogical strat— egies and requires that teachers demonstrate cultural and linguistic competence in a non-English culture and language. The BCLAD can also be earned via ex- amination. Far fewer teachers have pursued this cre- dential—in part because it requires more coursework and competencies. Many teachers of EL students come from the language groups in which they are certified to teach bilingually. However, most are native English speakers who have acquired their non—English language outside the home. One reason for this is that the college— going pipeline of Latinos and some Southeast Asian language groups is exceptionally weak, with many students dropping out of high school and few go on to college (Civil Rights Project, 2005; Lee and Ku— mashiro, 2005). Additionally, a disproportionate per— centage of students from these backgrounds have had difficulty passing the teaching examinations, usually due to issues related to English as a second language (Hill, 1998). Finally, students from these groups who are bilingual and who hold college degrees often have many other attractive job opportunities. The majority of teachers currently in the work- force has been credentialed under the above condi- tions. However, this is changing. Beginning in the spring of 2002, the approach to preparing teachers for EL instruction has been to embed the required competencies into the standard teacher credential curriculum. The legislation mandating this approach, AB 1059 (Ducheny, 1998), laudably aimed to ensure that every California teacher has some knowledge of teaching EL students. However, the law did not man- date that teacher education faculty have the knowl— edge and skills to embed this content, it did not set standards of performance or content, and, it did not add any instructional time for credential students to gain this expertise. CRUCIAL ISSUES IN PREPARING TEACHERS OF ENGLISH LEARNERS An important additional aspect of the Califor— nia teaching credential process is that it requires two years of induction, during which time a new teacher is to receive mentoring support from an experienced teacher. This approach addresses the problem of send- ing inexperienced teachers into the classroom to sink or swim on their own, and is supported . . . there are not enough by a body of research that has shown its effectiveness (Smith and Ingersoll, 2004; Gold, 1996). However, even this ef— fective practice suffers from the experienced EL teachers to serve as mentors to new teachers. problem mentioned previously, namely that California lacks sufficient numbers of teachers prepared to meet the educational needs of English learners. Consequently, there are not enough experi- enced EL teachers to serve as mentors to new teachers. Thus, the opportunity to strengthen teachers’ skills in working with English learners is often forfeited. Looking at the numbers Currently California school districts employ thousands of teachers who hold CLAD certification, or who hold a similar certification indicating their preparation to teach English language development and/or to use instructional methods for English learners. Of the 306,548 teachers the state employed in 2004-2005 per California Basic Education Data System (CBEDS) data, 41% reported that they are prepared and authorized to teach English Language Development (ELD) and 31% said that they are authorized to teach Specially Designed Academic Instruction in English (SDAIE)”. The figures include all BCLAD teachers, as they are certified to teach both ELD and SDAIE. California teachers are far less likely to earn BCLAD, the more comprehensive credential, and according to 2004—2005 CBEDS data, 12% of the state’s teachers have BCLAD15 authorization. The con— siderable overlap among teachers who report having SDAIE and ELD training leads the Center for the Future of Teaching and Learning to estimate that only 48% of teachers in 2005 held one or more of those designations (Esch, et al., 2005). PACE u 109 120 UNDERSTANDING THE RELATIONSHIP BETWEEN POVERTY AND STUDENT ACHIEVEMENT The link between poverty and lower student achieve- ment is hardly news. Beginning with the Coleman Report in 1966, researchers have understood the impact of a student’s background on his or her chanc— es for success in school. The news, however, is that an achievement gap based on socioeconomic status per- sists, both nationally and within California, despite policy and educational efforts to address the needs of low-income students. The National Center for Education Statistics (NCES) consistently reports a negative correlation be- tween poverty and student achievement, as measured by mathematics and reading performance on the Na- tional Assessment of Educational Progress, or NAEP as it is commonly known (NCES, 2005). A compre— hensive review of research on socioeconomic status and academic achievement between 1990—2000 finds that “parents’ location in the socioeconomic structure has a strong impact on students’ academic achieve- ment” (Sirin, 2005, p. 438). Family socioeconomic status determines such factors as a child’s access to resources at school and at home, and the quality of a child’s school and classroom environment. FIGURE 1: The Impact of Economic Disadvantage, English Language Arts Test Scorces, CST 2005 Percent Proficient and Above 70 2 3 4 5 6 7 8 9 10 11 Grade — — Non-Economically Disadvantaged Economically Disadvantaged Source: Cailfornia Department of Education (2006) FIGURE 2: The Impact of Economic Disadvantage, CAHSEE English Language Arts (ELA) Passage Rates, 2001 -2006* ’100 Percent Passing, ELA A l\) LN 01 \I O O 2001 2002 2005 2004 2005 2006 Non—Economically Disadvantaged — Economically Disadvantaged Source: Cailfornia Department of Education (2006) * See Endnote #3. California mirrors national trends, with signifi- cant gaps between the achievement of “economically disadvantaged” and “non—economically disadvan- taged” students on the state’s assessment of English Language Arts and Mathematics, as measured by the California Standards Test (CST). Low—income stu- dents are on average 30% less likely to achieve profi- ciency in these core subjects. Figure 1 demonstrates the persistent gap between lower and higher income students on the state’s assessment of English Lan- guage Arts proficiency. Such gaps in performance have real implica- tions for California’s schools. The state’s Academic Performance Index (API) includes “socioeconomi- cally disadvantaged” as a significant subgroup.1 In 2005, 68 0/o of schools in California enrolled enough low—income students to constitute a significant sub- group.2 Thus, two thirds of California’s schools are required to meet growth targets for a socioeconomi— cally disadvantaged subgroup under the state’s ac— countability system. Of further concern is the relationship between poverty and English learner (EL) students. For in- stance, in the 2004-05 school year 32% of Title 1 students in California were also identified as English learners (KW. Ashley, personal communication, May 31, 2006). Historically, California has seen an increase in its percentage of EL students; “Although California’s total public school enrollment has in- creased by 50 percent since 1981, the number of English learners has risen by nearly 300 percent.” CRUCIAL ISSUES IN CALIFORNIA EDUCATION 2006 FIGURE 3: The Impact of Economic Disadvantage, CAHSEE Mathematics Passage Rates, 2001 —2006* ’100 Percent Passing, Math 2001 2003 2002 2004 2005 2006 Non—Economically Disadvantaged — Economically Disadvantaged Source: Cailfornia Department of Education (2006) * See Endnote #3. (Tafoya, 2002). This poses a unique challenge for our schools, as a student’s inability to understand Eng- lish is negatively correlated with academic achieve— ment (Jepsen and Alth, 2005; Gandara et al., 2003). Results of the California High School Exit Exam (CAHSEE) show similar gaps as well. Approximately 25% fewer low-income students pass the exam as compared to their wealthier peers, and there are no signs of the gap narrowing since the exam’s introduc— tion in 2001 (Figures 2 and 3).3 Even when we see an overall increase in passing rates, low-income students are consistently less likely to pass the CAHSEE, as compared with their non—ec— onomically disadvantaged peers. Lower passing rates have real implications for students; beginning with the class of 2006, the exam is now required to receive a high school diploma. The value of a high school diploma is unequivo— cal. High school dropouts, for example, have higher rates of involvement in the juvenile justice system and in incarceration, and consistently earn less than their more educated peers (Snyder and Sickmund, 2006). Recent data shows that average earnings ranged from $18,900 for high-school dropouts, to $25,900 for high school graduates, and to $45,400 for college graduates (Cheesman-Day and Newburg- er, 2002). Educators and policymakers simply can not afford to ignore discrepancies in achievement across socioeconomic status. ADDRESSING THE NEEDS OF LOW~INCOME STUDENTS FEDERAL AND STATE POLICY EFFORTS TO ADDRESS THE NEEDS OF LOW—INCOME STUDENTS Federal Policies —Title I and NCLB Over the last few decades, several federal and state policies have sought to address the needs of low- income students. Most notable is Title I, a centerpiece of the federal Elementary and Secondary Education Act (ESEA), first signed into law in 1965. Originally designed to improve the academic achievement of poor and disadvantaged students, Title I distributes federal aid to needier schools. Upon signing the bill, President Johnson, who helped usher the bill through both houses of Congress, noted the bill’s importance, remarking that, “By passing this bill, we bridge the gap between helplessness and hope for more than 5 million educationally deprived children....We rekindle the revolution—the revolution of the spirit against the tyranny of ignorance.”4 While Title I funds were directed towards dis- advantaged children, the federal government did not specify the types of services or programs that should be provided. Instead, states and school districts were in charge of distributing more than $1 billion in fed— eral aid to local schools (Jennings, 2000). Originally, Title I was primarily used to fund supplementary pro— grams, allowing schools to implement specific strate- gies to benefit the lowest—performing students (Farkas and Hall, 2000). The program fell under sharp attack in the late 19605 when a report from the NAACP revealed that Title I funds were being used for a wide range of non—educational activities (Jennings, 2000). As a result of mounting evidence that states and districts were not using federal funds as intended by the law, federal lawmakers required two provi- sions——“c0mparability” and “supplement, not sup- plant”—to govern the distribution and allocation of Title I funds. The first provision requires that Title I schools receive funding that is comparable to the funding received by other schools, essentially man- dating the equalization of funds before Title I money is distributed (Roza et al., 2005). The second provi— sion requires that Title I funds do not replace state and local funding, ensuring that Title I provides extra resources for disadvantaged children. These provi- sions have remained in place and continue to govern PACE 71’ 121 , (“"6“ ,u' a A ‘ . . w " ' A ‘ . ‘ Ji' , . 9“ \ , x z , , 1 olicy Analfiis for California Education OVER THE PAST 1141111347121 '_ '2 xiii 11:14: C 1711321} REE B131"; r5115 153,5. 1} 1.5L G A 11‘; S 11101" E Fifi #1141} LC? 1 CH 1 12134 ‘1: 1; P110 L #3 S. State and federal agencies now spend over $3.7 billion annually on a variety of child care programs, ranging from vouchers for care provided by a grandparent, to formal preschool efforts (LAO, 2005). Research evidence continues to mount that de- tails how high—quality centers and family child care homes yield strong cognitive and linguistic benefits to children from poor families. Several counties have embarked on ambitious efforts to serve these chil— dren through such programs. The evidence is less clear about the sustainable benefits for children from middle-class and affluent families (Loeb et al., 2006; Magnuson, Ruhm, and Waldfogel, 2004). Filmmaker Rob Reiner’s Proposition 82—defeat— ed in the June 2006 election—sparked much debate over how to advance the development of preschool— age children. Among the middle—class, preschool—0r nursery school as it was called—offered a means for socializing children before they started kindergarten. Children would go to nursery “school” several days a week for a few hours, learning to share and play well with their classmates. Early educators and their professional associations—backed by most develop— mental researchers—encouraged learning through structured play, allowing preschoolers to engage in a variety of activities and learn to form friendships. This purpose was broadened with the advent of Head Start in the 19605—as policymakers defined preschool as a mechanism for advancing the early cognitive growth and social skills necessary for doing better in kindergarten and elementary school. Re— search has indicated that preschool can have broad and sustained benefits for disadvantaged children (Campbell et al., 2002; Heckman, 2000; Loeb et al., 2004; Reynolds, Temple, Robertson, and Mann, 2001; Shonkoff and Phillips, 2000). In response to the achievement gap between ethnic and socioeconomic groups—evident at kindergarten entry—preschool has been targeted as a potentially potent early inter— vention (Reardon, 2003; Rumberger and Arellano, 2003; Rumberger and Tran, 2006). The dramatic rise in the share of mothers with young children who work outside the home—climb— ing from 15% in the 19505 to almost two—thirds today—represents another major force in increased demand for child care. About two—thirds of Califor— nia four—year-olds already attend preschool centers