,: éfiéfi afifiqafifi . .35. 42. FIVE YEARS LATER MAY 2012 V INSTITUTE OF GOVERNMENTAL STUDIES LIBRARY FEB 1 6 2017 UNIVERSITY OF CALIFORNIA L31306;~11 G a”? 101% \9 BL 0613 ii Acknowledgments ’ his report commemorates the fifth anniversary of the Getting Down to Facts project, which sought to provide a thorough and reliable analysis of the critical challenges facing California’s education system as the necessary basis for an informed discussion of policy changes aimed at improving the performance of California schools and students. The report focuses on the four key issues that received emphasis in the Getting Down to Facts studies: governance, finance, personnel, and data systems. The authors review what has changed and what has not in the five years since the original studies were completed, and reaflirm the importance of a long—term agenda for reform in California’s education system that is guided by solid evidence and rigorous analysis. The report was produced with support from PACE’s core funders, including the California Education Policy Fund (a project of Rockefeller Philanthropy Advisors), the Dirk and Charlene Kabcenell Foundation, the James Irvine Foundation, and the Stuart Foundation, and we would like to express our gratitude to them. We would also like to thank Chris Reimann, who provided excellent editorial support; and Melinda Nichols, PACE’s Administrator, who cheerfully and efficiently managed the publication and production process from beginning to end. The views expressed in the report are those of the authors, and not necessarily those of PACE or our funders. Policy Analysis for California Education (PACE) is an independent, non—partisan research center based at Stanford University, the University of California Berkeley, and the University of Southern California. PACE seeks to define and sustain a long-term strategy for comprehensive policy reform and continuous improvement in performance at all levels of California’s education system, from early childhood to postsecondary education and training. To accomplish this goal, PACE bridges the gap between research and policy, working with scholars from California’s leading universities and with state and local policy—makers to increase the impact of academic research on educational policy in California. Founded in 1983, PACE: I Publishes policy briefs, research reports, and working papers that address key policy issues in California’s education system. I Convenes seminars and briefings that make current research accessible to policy audiences throughout California. I Provides expert testimony on educational issues to legislative committees and other policy audiences. l Works with local school districts and professional associations on projects aimed at supporting policy innovation, data use, and rigorous evaluation. Table 0fcontents _ ., II. III. IV. Getting Down to Facts - Five Years Later Recent Developments in California’s Educational Governance: 2007-2011 Financing California’s Public Schools: Toward a Weighted Student Formula Teachers and Leaders for California Schools Data, Policy Learning, and Continuous Improvement 16 25 32 ver an 18—month period from September 2005 to March 2007, the Getting Down to Facts Project brought together scholars from 32 institutions with diverse expertise and policy orientations. Our goal was to synthesize what we knew about effective school finance and governance in general, and about California’s school finance and governance systems in particular. We hoped that this information would serve as a base for productive public conversations about what we should do to improve K-12 education in California. The project was commissioned at the request of a bipartisan group of California leaders, including the governor’s Advisory Committee on Educational Excellence, the President Pro Tem of the California Senate, the Speaker of the California Assembly, the Superintendent of Public Instruction, and the Governor’s Secretary of Education. We produced 23 reports on topics ranging from the costs of achieving student outcome goals to governance complexity to data availability and use. Overall, the reports documented that good things were happening in some districts, schools, and classrooms, but that the flaws in California’s school finance and governance systems were likely hindering the quality of education available to many students in the State. Most importantly, education governance in California was unusually complex and restrictive which kept schools and districts from responding effectively to state standards and accountability. Similarly, the finance system was overly complex, restrictive, and irrational. Differences in spending across districts had little to do with differences in costs, needs or interests, but were instead historical artifacts with no current rationale. California stood out from other states in a number of areas including the low number of educators (fewer teachers per student, administrators per student, and professional support staff per student) and the poor quality of information available on student learning and the effectiveness of programs in the state. While the project did not draw direct links between these policies and student achievement, the policies were likely at least part of the reason that Califomia’s students did (and continue to do) so poorly on national achievement tests relative to students in other states. We hoped that the Getting Down to Facts reports would help with the development of policies to streamline governance and to simplify and rationalize school finance, in addition to improving information and ensuring that schools had sufficient and excellent staff. Our initial optimism was clearly unwarranted. As discussed in greater detail in the chapters that follow, the past five years have seen only small improvements on the problems identified in Getting Down to Facts, though the issues raised in the reports have penetrated policy discussions and there are indications that greater improvements may come. Economic challenges, not structural reforms of finance and governance, have dominated education policy in California during the past five years. The economic crisis has altered the context of schooling in Califomia. Nonetheless, many of the conclusions from the earlier reports continue to have relevance today. This report seeks to provide an update on the finance and governance ofCalifornia schools, noting progress towards reaching the goals set out in the original reports and highlighting areas for continued focus. Getting Down to Facts identified areas of concern in school governance structures, school finance, personnel, and data availability. This update picks up each of these themes. Governance In the area of governance, the initial project found that California placed substantial restrictions on schools’ and districts’ use of resources. These restrictions imposed compliance costs and made it difficult for local actors to respond to incentives in the accountability system. Regulatory requirements in an education code with 500 chapters and more than 1,250 articles appeared to stifle local innovation. These regulations also imposed needless constraints on local school administrators, causing them to focus on compliance and its attendant paperwork rather than on meeting teaching and learning goals. While other states with strong accountability systems had reduced regulations to enable local improvement initiatives—Florida and Connecticut, for example—Califomia had not. Instead of encouraging flexibility and innovation at the local level, many of California’s state policies constrained local actors, forcing very similar policies upon districts regardless of either local needs or capacities. Flexibility to allow for innovation, simplicity and transparency were three of five characteristics of effective governance that the original project identified. The complexity and rigidity of governance reflected in the education code conflicted with each of these goals. In addition the earlier report posited that effective systems are often characterized by accountability and stability. While California had set challenging standards for its schools, the system of parallel public reporting of school performance under the federal No Child Left Behind Act and Califomia’s own Public School Accountability Act sent mixed signals to parents and educators. Surveys of superintendents and principals also revealed that constant changes in state-level policy hindered their own planning. In, “Recent Developments in California’s Educational Governance: 2007—2011,” Richard Welsh and Dominic Brewer update these original findings. They find that the State continues to have a governance system that is weak in each of the five characteristics of good governance systems: stability, accountability, flexibility, transparency, and simplicity. Most stakeholders see little change in the system since the initial Getting Down to Facts reports. However, little change is not the same as no change at all. Local authorities have gained more flexibility over resources from a consolidation of categorical grants. The elimination of the office of the Secretary of Education under Governor Brown has somewhat streamlined state-level decision making. There have been fewer policy fluctuations, though the precipitous decline in funding has overshadowed this greater policy stability. The Federal government has taken a greater role in education through its Race to the Top initiative and other stimulus funds incentivizing innovation. There is also some evidence of new local initiatives increasing accountability, flexibility and transparency, such as those in the Los Angeles Unified School District. Finance School governance is linked with school finance. The failures of governance are due, at least in part, to the complexity, opacity, and rigidity of the school finance system. In California, district spending levels are set at the state level, with only minor exceptions, and a higher proportion of funds come from state revenues than in most other states. This degree of state control stems from Proposition 13, which limits the local property tax and leaves districts with limited capacity to raise local funds for school operations. The state also controls other aspects of school finance policy, imposing more restrictions on how districts spend their revenue than do other states. The initial Getting Down to Facts reports identified a number of problems with the current finance system. First was the lack of transparency. The number of dollars available to each school district was largely a historical artifact of spending in the 1970s combined with confusing categorical grant programs. As a result, similar districts received substantially different revenues per pupil, and differences in student needs across districts were not systematically accounted for in determining revenue levels. Second, in addition .to the opacity of the system, excessive reliance on multiple categorical funding streams imposed costly compliance burdens on school districts. Third, the system produced inequitable funding levels. Predating the implementation of modern accountability policies, the finance system had not been updated to align with the state’s accountability system nor redesigned to help local officials meet student performance goals. Differences in spending across California districts were substantial and not systematically tied to costs, needs, or demands. Despite a court-ordered school finance equalization plan spending across California school districts varied widely, with measures of cost differences such as district poverty level, racial and ethnic makeup, urban status, and district grade span explaining little of the variation in spending. Fifth, the sources of school funding were unstable in terms of both revenue fluctuations and delays in the budgeting process. Stock price volatility and the state’s relatively progressive personal income tax produced years of boom and bust for California schools. Finally, the overall level of school funding was low relative to other states, particularly when adjusted for the high cost of college educated workers needed to teach and to lead schools. “Financing Califomia’s Public Schools: Toward a Weighted Student Formula,” by Heather Rose finds little system change over the past five years, except for a move toward fewer categorical programs that has somewhat increased spending flexibility at the district level. However, two other substantial changes are worth noting. The first is the economic downturn which reduced spending across the State. State general fund spending was 15 percent lower by the end of the decade than it was at its peak in 2007-2008, though federal stimulus dollars reduced spending cuts at the district level. The second change is the emergence of a series of proposals to switch the current school finance system to a weighed student funding formula. Implementing a weighted student funding system would increase transparency and reduce the complexity, rigidity, and inequality of California’s current system for funding schools. Personnel In addition to assessing Califomia’s school governance and finance systems broadly, the Getting Down to Facts project identified some critical areas for deeper analysis. In particular, the project analyzed systems for personnel management and also analyzed the availability and use of information for school improvement. The personnel reports compared California’s teacher policies to policies in other states, assessed the impact of collective bargaining laws on the teacher work force, and examined school leadership, including the development of leaders and the job of the principal. The project concluded that California did not have a coherent system for supporting _ the entry, development, and retention of quality teachers and administrators. In particular, State policies on teacher recruitment and professional development were weak, due process rules combined with weak evaluation systems kept schools and districts from dismissing ineffective teachers, and salary schedules did not help districts achieve their goals by for example paying more for teachers in difficult to staff fields. One of the more striking findings from the initial report stems directly from the school finance system and the relatively low levels of spending across the state: California has dramatically fewer teachers and school leaders per student than most other states, including similar states such as Florida, New York and Texas. Jennifer Imazeki’s chapter in this volume, “Teachers and Leaders for California Schools,” finds that cuts in state budgets have led to layoffs and cuts in professional development. Layoffs worsened the already low number of adults per student in the state’s schools. The consolidation of categorical programs described above provided more local flexibility on school spending, but many districts used funds formerly spent on professional development for teachers to help offset overall cuts and maintain staffing levels and instructional activities. On the positive side of the ledger the California Commission on Teacher Credentialing has increased its oversight of teacher education programs, and efforts by the US. Department of Education to encourage reforms in teacher development, compensation, evaluation, and retention policies have shifted policy discussions both locally and at the state level. Data In order to improve schools, district leaders and other policy—makers need to learn from their experiences and the experiences of other educators. The original Getting Down to Facts reports found that California lagged other states in the development of a longitudinal student and teacher data system, and that the State had not developed sufficient analytical capacity to make good use of educational data, both of which hindered cycles of improvement. Instead, California had many unconnected data collections within the Department of Education, while other important data, particularly concerning teachers, were collected by agencies other than the Department. The lack of a comprehensive educational data system made it difficult to link key data elements, preventing systematic analysis of what was working and wasn’t working in classrooms, schools, and school districts. David Plank’s chapter, “Data, Policy Learning, and Continuous Improvement,” provides a new assessment of information use for California schools. He finds some progress on this front, most notably that California’s longitudinal student data system (C ALPADS) has been running for two years, and will soon be fully operational. C ALPADS allows for detailed analyses of student learning as well as simple descriptions that weren’t available before such as accurate measures of drop—out and graduation rates. These data are a step forward but they could be even more useful if linked to information on teachers, the most important resource in schools. Governor Brown vetoed state funding for the state—wide database on teachers, and chose not to apply for federal funding to link K—12, higher education and workforce data, thus limiting the development of a useful data system for California schools. Five Years Later The economic downturn and the consequent reductions in funding have dominated school policy during the five years since the release of the Getting Down to Facts reports. In spite of these challenges, however, at least in our optimistic eyes, there are signs of progress towards the goals laid out in the initial reports. In particular, the weighted student formula for funding schools currently under discussion would increase the simplicity, transparency, flexibility, equity, and ultimately the level of educational funding. The new CALPADS system provides information on student learning and educational attainment, previously unavailable, which allows policy- makers, educators and the public to see what students are learning and, potentially, which programs are working. On the other hand, many of the issues identified in Getting Down to Facts that were hindering education then, still apply today. The system is heavy on compliance and complexity and light on information and related opportunities for improvement. A number of districts are breaking this mold, creating their own systems for improving education quality; State policies, however, often hinder rather than help their progress. It is our hope as part of the original Getting Down to Facts project and this update that the next five years will see fruitful changes to state policy so that California can provide the educational opportunities that its students deserve. overnance refers to the institutions, organizations and individuals involved in the educational decision making and delivery systems. Five years ago the Getting Down to Facts (GDTF) research project examined California’s governance system in an attempt to help frame the education policy conversation. The key findings addressed the manner in which California’s governance system works and how the system is perceived by key stakeholders. GDTF highlighted the complexities of California’s multidimensional educational governance structure. The system is characterized by many organizational entities -- schools, districts, and county, state and federal agencies -- that have overlapping responsibilities across executive, legislative, and judicial jurisdictions (Brewer and Smith, 2006). (See Figure 1.) The GDTF analysis concluded that California’s governance system offers multiple opportunities to impede change, but few opportunities to lead and implement change (Brewer, Pelayo and Ahn, 2008). Though more money is needed to improve education in California, increased funding without a sustained effort to improve governance is unlikely to lead to substantial improvement in student learning. FIGURE 1. Major Institutions in California’s Educational Governance . State Board Legislature of Education Superintendent of Public instruction California Department of Education Local District Board of Education County Board of Education Local District Superintendent County Superintendent District Central Office County Office of Education Schools School Site Councils Teachers Students and Parents Governor _ .Other State Agencies Unions and Other Interest Groups Other CoUnty and CityAgencies _ - Brewer and Smith (2008) developed an analytic framework of governance effectiveness with five specific indicators: I Stability; I Accountability; I Innovation, Flexibility and Responsiveness; I Openness and Transparency; I Simplicity and Efficiency. These indicators are defined in Table 1. Based on a comprehensive review of the research literature and historical documents, as well as on interviews with leading academics and key stakeholders, Brewer and Smith found the system in California lacking on each of the five indicators of effective governance. For example, interviewees noted that TABLE 1. Five Characteristics of Good Governance revenue fluctuations and staff turnover at all levels are common. Increased use of categorical funding over recent decades coupled with frequent and prescriptive policy changes has led to frustration at the local level. Many of those interviewed felt that compliance took precedence over creativity, state regulations were overly burdensome, and special interests were overly influential. The structure was highly fragmented; stakeholders agreed that there was a need to clarify the role of institutions in the system, especially at the state level. There was also a strong desire to give districts greater autonomy over decisions. GDTF recommended relaxing state regulation, allowing more local control and increasing capacity to evaluate policies. Characteristic Definition and Rationale Stable learning. Accountable spending. Innovative, Flexible and Responsive Transparent and Open Simple and Efficient minimized. A stable governance structure is one in which policy is planned, made and implemented as far in advance as possible. Revenue is known in advance for planning. Policies are given an opportunity to work, with few major changes of direction or new initiatives introduced suddenly. Leaders have tenures that allow for knowledge development and on—the-job A governance structure with strong accountability has clear lines of authority between the various parts of the system, with limited duplication of functions. There are consequences for good or bad behavior and outcomes. Accountability gives the right incentives for actors within the system to accomplish their goals. There is alignment between revenue and An innovative, flexible, and responsive governance structure can adapt to changing context and is able to respond appropriately to new short— and long-term external demands. New approaches are encouraged; many ideas are generated and spread throughout the system. In a transparent and open governance system, it is clear to the public and all stakeholders how decisions are made and who makes them; participation is encouraged at every level. Transparency allows for the exchange of information between the different levels of the system. An open and transparent system is less likely to be ”captured” by special interests or have corruption and bribery, and most likely to encourage public engagement and support of schools. There is an open flow of information, including monitoring and evaluation of data and mechanisms to communicate performance to citizens. In a simple and efficient governance structure, decisions are made in a timely manner with minimal overlap or confusion among entities. Decision making is located where knowledge is greatest. Policy is coherent and decisions across multiple domains and levels are coordinated so that there is minimal duplication and waste. The decision making and implementation structure is not burdensome to stakeholders in the system. Costs are Source: Updated from Brewer and Smith (2008) L Recent Developments in California’s Educational Governance: 2007-2011 Data from a new round of interviews conducted in November 2011 with state-level policy leaders in education are summarized in Table 2. Most interviewees continue to rate the state as weak on every indicator. These findings bring to light four overarching trends over the past five years: fiscal instability; some simplification in educational governance in Sacramento; a changing federal role; and the local dynamic, including parental activism. TABLE 2. Governance Findings for California in 2011 Compared to GDTF Findings from 2008 Characteristic Stable Policy Then: - Policy fluctuates frequently. Now- Funding Then: Now projections fluctuate widely. State-level decision making Then: knowledge and expertise. Now: Leadership Then: Now: Accountable Lines of authority Then: was responsible for what tasks. Now: . No change. - Policy fluctuations are less frequent. - A Democratic legislature and Democratic Governor provide a foundation for consensus, but their agendas remain somewhat unclear. Findings from Interviews Then (2008) and Now (201 1) - Frequent adjustments in the areas of student assessment and curriculum lead to premature changes in requirements and implementation. ~ Funds fluctuate according to economic trends. - Lower levels of overall funding, increased reliance on categorical funding, lateness of budget, and inability to raise funding locally lead to unpredictable financial planning. . Levels of overall funding are lower, budgets are late (except in 201 0—1 1), and revenue - Reduction of staff in state—level agencies and shorter term limits reduce long—term - Multiple agencies serving different bosses hinder coherent decision making. . Lack of student data system hinders effective decision making. - The Office of Secretary of Education under Governor Brown has been eliminated. - Student data system (CALPADS) launched but funding for CALTIDES was cut. - Turnover of state officials, school boards, and superintendents is high. . High turnover leads to lack of continuity and stability of programs. . Seven new SBE members were appointed in 2011. - Few interviewees knew who was in charge of different aspects of the system and who TABLE 2. Governance Findings for California in 2011 Compared to GDTF Findings from 2008 (Continued) Clarity of responsibilities Then: - Numerous local, regional, and state-level entities have overlapping responsibilities. - The responsibilities of each stakeholder in the system are unclear. Now: ~ No change. Innovative, Then: Flexible and Innovation Responsive - Charter schools are one example of a relatively successful attempt at local autonomy and innovation. Flexibility . Stakeholders sense that the system is highly bureaucratic and more concerned about compliance with regulations than about innovation. . Local entities do not have autonomy to make decisions or attempt innovative strategies. Responsiveness - State decision-makers have preferred one-size-fits-all solutions, such as class size reduction. Now: - Local authorities have more flexibility over resources due to reduction in categorical funding streams. - Bolder reforms are underway in major districts like LAUSD. Transparent Transparency and Open Then: - Little concern about transparency was found among interviewees; no widespread evidence of unethical actions or corruption was evident. - No evidence was offered that California is any worse than other states in public participation and voter turnout. Now: - No change. Special interests Then: - Interviewees showed great concern over the role of special interests in state—level decisions. - Particular concern was directed toward employee unions and their influence on system decisions. Now: - No change. Simple and Simplicity Efficient Then: - Instability, confusing lines of authority and unclear responsibilities lead to an overly complex system. Now: - No change. Efficiency Then: - Rigid, prescriptive state legislation leads to wasted effort to comply with a multitude of mandates. - Interviewees expressed a need for more local authority and flexibility in resource allocation. Now: . Changes in categorical funding have led to more local authority and flexibility in resource allocation. Sources: Brewer and Smith (2008); Authors 10 L Trend 1:0ngoing Statewide Fiscal Instability “Could there be a worse five year period?” “We’re playing defense now and it’s hard to generate momentum.” Analysis of school governance in California must be set within the context of the chaos of the state budget crisis. The overriding budgetary concerns have, in essence, halted any governance reform. The result is few major policy changes in the last five years. Interviewees contend that merely getting the budget done is the primary goal of policy-makers. A USC Dornsife/Los Angeles Times poll (2011) confirmed that parents perceive bigger class sizes, increased out-of—pocket expenses for student supplies, fewer arts and after-school programs and increased local bonds or taxes as some of the consequences of the consistent decrease in state spending on public schools due to the state budget deficit. In addition to declining funding, local districts have had to work within an environment of overall uncertainty, including multiple budget revisions. Planning has been extremely difficult. The last five years have also seen an increase in the legislative manipulations of Proposition 98, a complex set of formulas that guarantees a minimum funding level for K—12 education. These machinations have included funding deferrals, the redefinition of General Fund revenues, and programs moved into and out of Prop 98. Tiered budget “triggers,” which provide for automatic budget reductions if expected revenues fall below oflficial projections, were enacted in the 2011-12 California state budget. However, the state legislature responded to the budget crisis by passing AB 114 in July 2011, stipulating that school officials must ignore the prospect of the triggers and maintain staffing and program levels at the same funding level as the previous year. In essence, AB 114 limited the options of local entities to prepare for cuts in funding. In December 2011, triggers were activated prematurely, cutting the school budget in the middle of the school year; these automatic spending cuts took effect in February 2012. Fortunately, reductions to K-12 schools totaled only $330 million, far less than some projected. Governor Brown has announced his intention to place an important choice before voters in November 2012: approve an additional $7 Billion in taxes to support schools or face cuts of an equal amount. Regardless of the outcome of Governor’s Brown revenue proposal, schools will likely cope with additional uncertainty in 2012-13. Trend 2: Some Policy Simplification in Sacramento “Categoricals are a real life experiment.” “The SBE is going to be a stronger policymaking entity.” An unintended consequence of the California budget crisis has been some simplification in policies related to state level funding and governance. In 2009, California changed its school finance system by identifying a set of ‘flex items’ to be reclassified from categorical to general purpose programs. The Budget Act relaxed spending restrictions through 2014-15 on over 40 categorical programs that constituted 30 percent of all categorical revenues in 2009-10. Essentially, local districts and schools received greater flexibility in exchange for fewer funds. Flexibility provisions allowed districts to increase class sizes up to 25 and still retain 80 percent of K—3 Class Size Reduction (CSR) funds; districts could also reduce the length of the school year. Such changes in categorical funding represent a real life experiment in priorities, allowing local districts to choose which programs to maintain and which to eliminate. The legislature will need to decide whether to return to a restricted categorical funding system or to pursue further restructuring efforts for a simpler, flexible finance system such as the weighted student funding approach proposed by Governor Brown. Reactions and responses to the changes in categorical funding have been mixed. Although district leaders’ perceptions of flexibility varied widely, most welcomed the flexible dollars and praised the change (Fuller et al., 2011). Some local authorities have taken advantage of the flex, while others have increased their distrust of the state, wondering whether funding reductions will continue. Though districts have increased flexibility, there are still variations and disparities in funding across school districts that raise equity concerns. Additionally, policy—makers and district oflficials have expressed concern about funds being used to increase teacher salaries and benefits. There are unintended consequences of the flexible use of categorical spending, such as reductions in adult education funding. The flexibility in categorical dollars may have drastic effects on dropout recovery programs and other efforts to increase the capacity of the labor market. Overall, flexibility has challenged districts to seek efficiency and reassess spending priorities while remaining committed to programs (Fuller et al., 2011). One of the first steps Governor Brown took upon entering office was to choose not to appoint a Secretary of Education. Whether this change is permanent is unknowable, but it reflects a simplification of the state level landscape — although some interviewees also noted the unintended consequence of the Governor not necessarily having enough staff capacity. Governor Brown also signaled his intent to make the State Board of Education (SBE) a stronger policymaking entity. He appointed as president of the SBE his major education policy advisor, Michael Kirst, who served on the state board during Brown’s earlier time as governor (1975- 1983). Due to the legislature’s failure to confirm some of Governor Schwarzenegger’s SBE appointees, Governor Brown took advantage of an unusual opportunity to appoint seven members in Ianuary 2011. In one of his first oflicial acts, the governor replaced several vocal proponents of charter schools, parent empowerment and teacher accountability (Mehta, 2011). His new appointees have transformed the state board to one more closely aligned with traditional educational interests. As of early 2012, there are still three state board openings that Governor Brown has yet to fill on the 11-member board, and the term of the current student board member is set to expire in July 2012. The two- thirds vote in the Senate required to approve a SBE nominee, coupled with the upcoming ballot initiative, may partly explain the Governor’s cautious approach. Although the new appointments may make it easier to build consensus, the board still faces several challenges with an increasingly overburdened staff. An Expanded Federal Role “The Honeymoon is over with the Obama administration.” Interviewees agreed that there has been an increase in federal influence in educational governance over the past five years, but they disagreed on whether it will have lasting impact. The federal role in education policy has been increasing since the enactment of No Child Left Behind (NCLB). The American Recovery and Reinvestment Act, which provided approximately $1000 per pupil from 2008-09 through 2010-11, increased federal support to California schools by nearly 45 percent in 2009—10 compared to 2007- 08. Additionally, the aggressive reform agenda pursued by the Obama administration through the competitive Race to the Top (RTTT) initiative has had a significant impact on federal-state relations (see, for example, Hess and Kelly, 2012). RTTT is a federal system of incentives used by the Obama Administration to induce states to make particular educational reforms. California’s RTTT application resulted in a flurry of legislative activity as the state sought to improve its prospects of winning RTTT funds. In January 2010, Governor Schwarzenegger signed an RTTT legislative package that included standards and assessments that embrace the national common core standards; the use of data in instruction and teacher evaluation; and the “parent trigger,” a school—turnaround law that allows parents to petition for the conversion of failing schools into charter schools (Office of Assembly Speaker Karen Bass, 2010). Although California lost its bid for RTTT funding, the legislative package committed the state to linking teacher evaluation to student test scores; student achievement would now account for at least 30 percent of a teacher’s evaluation. However, California’s RTTT application was not signed by any of the big districts, such as Los Angeles Unified or Long Beach Unified, or by the major state unions (Blume, 2010). Though RTTT did result 11 12 in new laws regarding education reform, there is little money to implement many of the proposed changes, and it is not clear whether there is much appetite to do so. In early 2012, California was mulling whether to apply for an NCLB waiver for greater flexibility and different conditions than those stipulated by the Obama administration (Cavanagh, 2012). One of the major points of contention in NCLB is a federal provision requiring states to include standardized test scores in the performance evaluations of teachers, a policy option repeatedly opposed by teachers unions in California. The state also wants the federal government to relax several of NCLB’s biggest sanctions, as well as the waiver eligibility requirement of establishing a new accountability system. The active federal— state dynamic is a significant new development in educational governance and is likely to persist for some time. Indeed, federal initiatives combined with parent activism may be the external catalysts that initiate educational reforms at the local level. The Local Dynamic, Parental Activism and the Role of Special Interests One interesting, if subtle, change in perceptions of governance over the 2007- 2011 period has been the growing visibility of local activism around some important policy developments. Several of our interviewees noted this phenomenon, which was absent from the data collected during the original GDTF interviews. Interviewees generally believed that local activism was having an impact; some believed that it might even be a “game changer.” In particular, upheavals in the Los Angeles education landscape have reverberated in Sacramento. In 2009, Los Angeles Mayor Antonio Villaraigosa supported parents, reform groups and school board vice president Yolie Flores in an attempt to open more of the LAUSD’s new and reopened schools to competition from charter groups. In response, United Teachers Los Angeles (UTLA) mounted a campaign to put many of those schools in the hands of teacher groups backed by the union. Since then, several changes have been made to the “Public School Choice” initiative. It will be interesting to see how the reform efforts in Los Angeles unfold, and how they impact educational governance in California. Additionally, in 2010 the Los Angeles Times published performance measures for existing teachers based on value-added estimates that highlighted performance at the classroom level. This public airing sparked union outrage and added fuel to the debate over teacher accountability. Interviewees also mentioned the shifting nature of special interest politics in the state’s educational governance. There is at once more division within the education coalition, more players and active advocacy, and a greater effort to collaborate and align interests. As one interviewee put it, “the CTA (California Teachers Association) is the gorilla,” but there are other special interests that are coalescing around particular issues and shaping the formulation and implementation of education reforms. Some stakeholders believe that special interests “feel pretty furious,” as if they are fighting over a smaller pie. In the current funding climate, everything seems like a zero—sum game as the winners win what the losers lose and interest groups battle over shrinking resources. Parent Revolution, a reform- minded community organization launched in January 2009 to help implement the parent trigger policy in California, played a prominent role in organizing the parent petition to takeover McKinley Elementary School in Compton. In December 2010, the parent trigger was invoked for the first time at McKinley. The parents were eventually defeated on a petition technicality in a contentious battle that saw the intended charter school opening a few blocks away but being filled with few students from McKinley. In early 2012, Desert Trails Elementary School in Adelanto was the site of a second attempt at a parent takeover. A group of parents backed by Parent Revolution is hoping to enact the parent trigger on low— performing schools, but it faces opposition from another group of parents supported by state and local teachers unions. Although the staying power and future nature of parent activism is unclear, the interactions between local districts, parents and the legislature will play a substantial role in the state’s educational reforms. Prospects for the Future Most stakeholders participating in this study believe relatively little has changed in California’s educational governance; one described the past five years as “being in a holding pattern;” another asserted that “the budget situation hijacked any real conversation about policy reform.” The extent and uncertainty of revenue cuts presents numerous challenges at all levels. In general the mood among many in Sacramento has been that until there is more money, no real reform (e.g., more local flexibility, changes to accountability systems) can happen. Amid the chaos that pervades California’s school finance and governance systems, there is some hope. Education remains a top priority on the state policy agenda and local concerns have a growing impact in Sacramento. Moreover, the budget crisis has provided some opportunity for progress — notably changes in categorical funding that have led to more local flexibility. There will likely be one or more ballot propositions in November 2012 to increase funding for public schools by raising taxes. The USC Dornsife/Los Angeles Times poll suggests that the majority of Californians are in favor of increased funding for public schools even if it means a tax increase. There is an ongoing attempt to craft a single measure, but competing propositions are quite possible. Competing initiatives would each raise extra revenue but would likely differ on important issues such as who bears the burden of the tax increase, whether all of the money goes into the state’s general fund, the degree of structural changes required as well as the level of funding for schools and whether it is temporary or permanent. The governor has called for $6.9 billion in temporary new taxes, but the Legislative Analyst’s Office (LAO) has questioned some of the assumption in the governor’s proposal (Yamamura, 2012). More important, $6 billion of the revenues in the governor’s initiative will go to balancing the budget, with only about $1 billion devoted to increasing funding for schools. Alternative proposals may result in a more permanent and greater increase in funding for schools. A measure proposed by Molly Munger (Our Children, Our Future) suggests a 12-year initiative splitting $10 billion in the first four years between state bond repayment ($3 billion) and K-12 schools ($7 billion); in the final eight years the proposal would add $10 billion a year toward K—12 and pre-kindergarten programs. Overall, a crowded and dynamic ballot initiative landscape is emerging; there may be more consolidation of coalitions as the election date draws nearer. Until recently the governor has given relatively few clear signals of an education policy agenda. This has begun to change. In his 2012 State of the State speech, Governor Brown declared, “we should set broad goals and have a good accountability system, leaving the real work to those closest to the students,” and “I embrace both reform and tradition—mot complacency.” The governor proposes to replace categorical programs with a new weighted student formula with additional funding for disadvantaged students and struggling English learners that simplifies funding streams and reduces bureaucracy. (See the chapter on school finance.) Additionally, the governor stressed the importance of timely data and called for a reduction in the number of tests, with results returned more quickly to teachers, principals and superintendents. He also signaled his intention to work with the SBE to develop a proposal for a qualitative system of teacher assessments that includes school and classroom visitations. Changes to accountability, more local authority and increased transparency are in line with the GDTF recommendations, but it is not clear what the path to implementation may be. Finally, the governor has indicated a desire to attack a looming public employee pension shortfall. The California State Teachers’ Retirement System (CalSTRS) and the CTA want the legislature to approve $4.1 billion a year that CalSTRS needs to level up its pension plan. Resolving this shortfall may offer an opportunity for other state level reforms. We reiterate the two policy changes that may help facilitate improved educational governance in California: the development of a comprehensive statewide data system and changes to the system of state financing. Action on both recommendations is necessary to build local capacity. CALPADS, the student element of state data system, is operational but restoring funding for CALTIDES will enhance California’s long term data capacity. Some interviewees noted that schools need local 13 sources of revenue and believed that long term solutions for state financing lie in removing local revenue generation constraints. These changes can help solidify the foundation for continuous improvement in the performance of California’s public schools and ensure that all California students have the opportunity to succeed. Timeline of Major Events in California Governance and Finance, 2007-201 1 January 2010 December 201 0 October 2011 Special RTTT inspired First use of parent State $654M short legislative package trigger at McKinley in revenue including parent Elementary School trigger provision in Compton Governor Brown embracing common vetoed Steinberg’s Jul 2009 core standards and June 2011 Accountability Bill using student test with new Paymentdstowards scores and teacher AB 114 education index 03" a JUSted to evaluation is passed h save $450M passed t at limits districts January 2012 option to respond to Governor Brown budget cuts State of the State speech called for local authority and increased transparency in public education August 2010 California lost bid for RTTT V Budget Deficit $93 Budget Deficit $163 Budget Deficit $21 B Budget Deficit $193 Budget Deficit $253 K-12 PPF = $8,235 2008 School Year 2007-08 K-12 PPF = $8,415 2009 2010 School Year 2008-09 School Year 2009-10 A A K—1 2 PPF = $7,933 K-12 PPF = $7,830 2011 School Year 2010—1 1 K-12 PPF = $7,713 2012 School Year 2011-12 14 February 2009 Budget Act relaxed spending restrictions on over 40 categorical programs through 2014—201 5 Federal stimulus W Jerry Brown elected CA Governor Voters passed ballot measure that w Triggers pulled at lower than expected amount $13 in additional cuts funding offset budget now allows June 2011 reductions by $63 October 2009 budgets to be ($3.23 from SFSF, $2.83 passed with the Budget trigger from ARRA) Governor simple majority as provision enacted Schwarzenegger Opposed to with automatic calls special session tWO'thlrdS spending CUES tied of legislature to address obstacles to competitive RTTT application Sources: Legislative Analyst’s Office, Sacramento Bee, LATImes, EdSource. to revenue projections References Blume, H. (2010, August 24). California loses bid for federal Race to the Top education grant. Los Angeles Times. Retrieved on December 15, 2011 from http://www.latimes.com/ Brewer, DJ, 8: Smith, J. (2006). Evaluating the “Crazy Quilt”: Educational Governance in California. Los Angeles, CA: University of Southern California. Brewer, D.J., 8: Smith, J. (2008). A framework for understanding educational governance. Education Finance and Policy, 3(1), 20-40. Brewer, D.J., Pelayo, I., & Ahn, I. (2008). Revisiting the “Crazy Quilt”: Recent Developments in California’s Educational Governance. PACE. Cavanagh, S. (2012, February 29). California asks for NCLB waiver, but without strings. Education Week. Retrieved from httpzl/blogs. edweekorg EdSource. (2011, May). California’s Fiscal Crisis: What does it mean for schools? Retrieved from http://edsource.org Fensterwald, J. (2012, March 1). Cutting it close to a quorum: Brown has three State Board openings. Thoughts on Public Education. Retrieved on March 10, 2012 from http://toped.svefoundation.org Fuller, B., Marsh, J., Stecher, B., & Timar,T. (2011). Deregulating School Aid in California: How 10 Districts Responded to Fiscal Flexibility, 2009-10. RAND and PACE Research Network. Governor Brown State of the State speech 2012: “California on the Mend.” Hess, F., 8: Kelly, A. (2012). Carrots, Sticks and the Bully Pulpit. Harvard Education Press. Institute for Research on Education Policy and Practice. (2007). Getting Down to Facts: A research project examining California’s school governance and finance systems. Retrieved on June 12, 2008 from http://irepp.stanford.edu/ projects/cafinance-studies.htm I Mehta, S. (2011, January 8). Many see influence of teachers’ union in Gov. Jerry Brown’s shakeup of California board of education. Los Angeles Times. Retrieved on January 15, 2012 from http://www.1atimes.com/ Office of Assembly Speaker Karen Bass. (2010, January 6). Assembly passes Race to the Top Legislation. University of Southern California Dornsife College of Letters, Arts and Sciences and the Los Angeles Times Poll. (2011, November). Yamamura, K. (2012, January 11). LAO: Brown’s budget restores balance, but estimates questionable. The Sacramento Bee. Retrieved on January 12, 2012 from http://Www.sacbee. com Yamamura, K. (2011, December 23). Molly Munger changes tax initiative to address budget deficit. The Sacramento Bee. Retrieved on January 1, 2012 from http://www.sacbee. com 15 1.6 alifornia’s school finance system is in disrepair to the point that it no longer meets the needs of the state or its students. Starting about 35 years ago, funding schools became a state responsibility, guided largely by court decisions and state ballot initiatives.‘ These external forces rather than a strategic vision linking the state’s finance system to its educational goals have driven state policy for decades. Not surprisingly, the original Getting Down to Facts (GDTF) research concluded that the current system is overly complex, irrational, and fails to link resource allocations with student need or district costs.2 Under the current system, the State provides the bulk of funding for schools, with only minimal local and federal support. In 2006—07, the year GDTF was released, California schools received about 86 percent of their daily operating revenue from the state.3 This statewide portion includes the district’s share of local property tax revenue, because those taxes are essentially set statewide and allocated by the legislature, not local districts.4 On average, another 6 percent of district revenue came from a district’s local discretionary sources, such as revenue from leases, rentals, interest and parcel taxes. Lastly, the federal government provided about 8 percent of school district revenue for specific educational programs. 1Sonstelie (2008) and Sonstelie, Brunner, and Ardon (1999) describe how school finance responsibilities shifted from local districts to the state. 2 Loeb, Bryk, and Hanushek (2007). 3The data in this section come from the state’s standardized accounting data (SACS) as reported by Ed—Data. ‘ Sonstelie (2008). The revenue school districts receive from the state falls into two main categories. In most years, about three-quarters of state revenue is unrestricted and can be used for any legitimate expense for day to day operations. These funds are typically referred to as “revenue limit” funds and were originally intended to equalize per-pupil revenue across school districts. By design, however, disparities in revenue limit funding persist because the limits are different for unified, elementary and high school districts, despite the lack of empirical justification of any difference in needs.5 The other quarter of state funding is restricted and must be spent on particular programs.6 Although the state has created about 60 of these programs, typically about half of this categorical funding is earmarked for just four programs: Special Education, K-3 Class Size Reduction, Economic Impact Aid, and Targeted Instructional Improvement Block Grants.7 These and the other programs are highly prescriptive, require applications and compliance reports, and generally impede both the efficient use of funds and local allocation of resources in innovative ways that best improve student outcomes in the local environment. This system has been criticized not only for its complexity, but also for its overall funding level. Although California has set some of the highest academic content standards in the nation, it provides its schools with relatively modest 5 Weston (2010). 6Some of these restrictions have temporarily been relaxed as will be described shortly. 7 The actual percentage depends on the year. This percentage excludes child care programs. resource levels. Consistently over the last decade, California has had fewer staff per pupil than other states — only three—quarters as many teachers per pupil and only half as many counselors per pupil compared with other states.8 This resource deficiency motivated a strand of the GDTF research that attempted to estimate how much revenue schools would need for students to achieve the state’s academic goals. These studies suggest that the relationship between resources and outcomes is unclear — there is no magic dollar amount that would ensure that students succeed academically. And, although more funding may help students, without systematic reforms to several aspects of education in California additional revenue alone is unlikely to improve student achievement. Not only the research community has been critical of California’s current school finance system. The system is under attack through legislative, public initiative, and legal avenues. Two current lawsuits, Robles- Wong 1/. State of California (2010) and the Campaignfor Quality Education v. California (2010), claim that the current finance system is unconstitutional and demand that the state develop a new system that is better aligned with the State’s academic goals. In November 2011, the Advancement Project filed a ballot initiative for 2012. The measure, titled Our Children, Our Future: Local Schools and Early Education Investment Act, would increase the state income tax to create the California Education Trust Fund, raising additional funding sooner rather than later for California’s education system. Finally, the legislature and governor are developing alternatives to the current system. The GDTF research and subsequent work by Rose, Sonstelie, and Weston (2010) have identified five key principles that California’s school finance system should embody: I Adequacy: Provide students with enough resources to the meet the state standards. I Cost differentiation: Recognize that costs vary depending on student need and other specific " For example, see Rose et al. (2003), Loeb, Grissom, and Strunk (2007), and EdSource (2010). district factors. A district, however, should not be able to manipulate those cost factors. I Transparency: Design funding formulas that are straightforward and transparent. Formulas should adjust for key cost differences without trying to adjust for every slight difference between districts. l Equity: Allocate the same per-pupil revenue to all districts with the same cost factors. I Local flexibility: Provide more local authority in how revenue is used, so districts can address their local needs given their mix of available personnel and other resources. The current system fails to some extent in each of these areas, and any reform proposals should be judged against these criteria. This paper provides an overview of the steps California has made in the last five years that adhere to these elements of school finance reform. Most of the concrete reform discussions have focused on the structure of the system, rather than the level of funding. Nonetheless, the economic conditions of the state certainly act as an important backdrop to these discussions. A New Approach:Weighted Student Formulas Soon after the release of the GDTF studies, two groups proposed alternative systems embodying many of the five principles and crystallizing how a new system could look. Bersin, Kirst, and Liu (2007) and the Governor’s Committee on Education Excellence (2007) both proposed consolidating the current array of funding programs into just three simple programs and using a formula to distribute those funds equitably.9 The base program would constitute the majority of funding, which would be disbursed on an equal per—pupil basis and would be unrestricted in its use. Both the Bersin, Kirst, and Liu and GCEE proposals allow for different base funding rates by grade level, although the " Rose, Sengupta, Sonstelie, and Reinhard (2008) and Reinhard, Rose, Sengupta, and Sonstelie (2008) analyze the effect ofthese two proposals on the distribution of resources across schools. 17 18 L Governor’s Committee makes it a more explicit part of its proposal. An additional program would provide funding for Special Education. This revenue would also be allocated on an equal per- pupil basis, but would be restricted for the use of special education programs. The last program would target additional funds to economically disadvantaged students and English learners. Both proposals incorporated a targeted program because economically disadvantaged students disproportionally fail to achieve proficiency on the state standards, and many experts believe they may need additional resources to meet the state’s goals (Figure 1).10 The Legislative Analyst’s Oflfice (2008) proposed a similar restructuring approach, with one additional program that distributes funds on an equal per-pupil basis but restricts that money to be spent on staff professional development. FIGURE 1: English-Language Arts Proficiency on the California Standard’s Test and the Percent Economically Disadvantaged, Elementary Schools, 2007. 100 .. Percent Proficient Percent Economically Disadvantaged Source: Public Policy Institute of California (Rose et al., 2008). Note: Every dot represents an elementary school.The line represents the average relationship between the two factors, proficiency and economic disadvantage. These funding formulas fall into the class called weighted student formulas, because they essentially provide all students with the same base level of per-pupil revenue while weighting certain student groups more heavily with additional revenue. The ‘0 Chambers, Levin, and DeLancey (2006), lmazeki (2006), Rose, Sonstelie, and Richardson (2004), and Sonstelie (2007). formulas are simple and transparent, adjusting for only a few key differences in cost. They supply similar students with similar funding amounts, and they provide much more local discretion in how funds are spent by removing the categorical program restrictions. Through their targeted programs, these proposals take a big step toward linking revenue to student need. The Bersin, Kirst, and Liu proposal also adjusts funding in the three programs for the differential costs of labor around the state. About 85 percent of school district expenditures go toward personnel compensation, and the geographic variation in that compensation is highly correlated with the geographic variation in the salaries of college- educated non-teachers, in large part because school districts must compete with other employers for their staffing needs. Bersin, Kirst, and Liu suggest augmenting revenue in regions with high non— teacher wages so that schools can hire the same number of staff per pupil as schools in areas with lower non-teacher wages.“ Although these proposals all take important steps toward meeting the principle of cost differentiation, their formulas violate one aspect of that principle. They provide additional funding for students who are classified as English Learners. Because schools districts use their own criteria to classify students as English Learners, basing funding on this classification creates fiscal incentives for districts to prevent students from being reclassified as proficient in English. Providing a more standardized measure of English Learner status for the purpose of funding could still direct resources to these students who have additional needs while keeping this cost adjustment more in line with the principles of reform. All three proposals focus extensively on the structure of the school finance system, providing an improved framework for the distribution of additional school funding as it becomes available. The Governor’s Committee on Education Excellence takes their proposal one step further, “ Rose and Sengupta (2007) develop the regional wage index used by Bersin, Kirst, and Liu (2007) and Reinhard, Rose. Sengupta, and Sonstelie (2008) show how districts in regions with higher non—teacher wages end up with lower staffing ratios. suggesting that that the state would need to spend an additional $5 billion to achieve its education goals. Moving Toward Reform Amidst California’s Budget Crisis The California Budget Crisis The Getting Down to Facts research, subsequent policy proposals, and analysis from the research community created potential momentum for school finance reform. Unfortunately, these ideals collided with a severe budget crisis in California. In 2008-09, the California government cut state general fund spending a dramatic 12 percent. Spending continued to decline the next year, ending the decade 15 percent lower than its 2007—08 peak. Given that California public schools receive most of their funding from the state’s general funds, it is not surprising that total state contributions to school funding also fell.12 This state revenue to schools fell 12 percent during the last three years of the decade. Despite the reduced state funding, an infusion of federal funds helped temporarily insulate total school district revenue. The American Recovery and Reinvestment Act of 2009 (ARRA) provided nearly $1,000 per pupil, spread out over the three years from 2008—09 through 2010-11.13 The bars in Figure 2 show total school district general fund revenue by source. Federal funding, the top segment of each bar, was about 50 percent higher in 2008-09 and 2009-10 than it was in 2007-08. The federal support meant that total school district general fund revenue fell only 7 percent, rather than 12 percent, between 2007—08 and 2009-10. The school district revenue in Figure 2 comes from the SACS accounting data that districts report to the state. Although these data nicely track total school district revenue from state, federal, and local sources, they are not available after 2009— 10. The Legislative Analyst’s Office (2012) reports that the portion of revenue from state school funding rebounded slightly in 2010-11 but was still 8 percent below the 2007-08 peak, where it FIGURE 2: State General Fund Expenditures and K—1 2 School District General Fund Revenue 10,000 — 8,000 6,000 4,000 Revenue Per Pupil (5) 2,000 — 1 00,000 I State General Fund Expenditures (Right Axis) 80,000 I Federal Categorical I Local - State Categorical 60,000 5 Millions 40’000 I Revenue Limit 0 Total School District Revenue from State Sources 20,000 Sources: The state’s general fund expenditures come from the California Department of Finance’s Chart J, Historical Data, Growth in Revenues, Transfers and Expenditures, General Fund. The data on school district revenue come from the state’s standardized accounting data (SACS) as reported in Ed-Data Statewide Financial Reports of school district general fund revenue. Enrollment comes from the department of education and is used to compute per—pupil funding levels. Revenue to county offices is excluded. ’3 Proposition 98 earmarks about 40% of the state’s general fund for K—14 education, with about 90% of that going to K- 12 education and about 10% going to community colleges. ‘3 In addition, the Federal Education Jobs program provided about $200 per pupil spread over the years 2010—11 and 201 1— 12. 19 20 is expected to hover this year (the dashed gray line reflects these data).14 Although this slight recovery of state funds is welcome, it is somewhat counterbalanced by a reduction in federal support as ARRA and the Education Jobs program wind down. The revenue data from the SACS and LAO in Figure 2 obscure another important issue facing schools. They include revenue deferrals, essentially IOUs from the state to school districts. In 201 1- 12, the state issued $350 per pupil in new deferrals.15 In the future, the state will need either to increase the funding it provides schools to cover this annual deficit or acknowledge this deferral as a cut which would translate into another 5 percent decline in state funding of schools. The short-term path California takes will surely depend on the state’s economy, but also on whether Governor Jerry Brown’s proposed tax hikes are approved at the ballot box in November 2012. If voters approve Brown’s tax increases, state support of schools will remain relatively flat in 2012-13. If the ballot measure is not approved, the LAO estimates state funding per pupil will drop 4.4 percent that year. The state’s funding crisis initially caused many in the education community to focus immediate efforts on maintaining current educational programs and funding levels. Although this focus on funding preservation may naturally divert some attention away from long-run school finance reform, the crisis has actually caused the legislature to adopt changes in line with some key principles for structural reform. The Flex Item: Reform in the Absence of Revenue In February 2009, the legislature eliminated the restrictions on about 40 state categorical funding H The LAO school revenue data include the portion of state funding required by Proposition 98, which is about 88% of general state aid, property taxes, and state categorical programs (CDE, 2011). Examples of non—Proposition 98 funding include excess taxes above the revenue limit, state lottery funds, and in some years, pupil transportation funding (Weston, 2011). The annual changes in LAO’s state funding are very similar to those in the SACS data. Figure 2 applies the LAO changes to the SACS levels to extrapolate state revenue. '5 LAO (2012). These accounting maneuvers highlight yet one more aspect of the state’s complex finance system that makes it difficult to track money from the state to school districts. programs, essentially consolidating them into one funding stream and allowing the revenue to be used for any legitimate education expense.16 The state granted this funding flexibility to help districts deal with the severe budget cuts. This new aggregate set of programs, commonly referred to as the flex item, represented about 40 percent of the categorical funding (LAO, 2011). Although the flex item simplifies the finance system and grants districts more local control, it violates other important reform principles. It is neither equally distributed nor highly related to student need.17 On average, elementary districts tend to have the lowest per-pupil flex-item funding rates while high school districts tend to have the highest, although this discrepancy would change if K-3 Class Size Reduction funds were included in the flex item. Even within each type of district, per-pupil funding varies substantially. Although flex item funding increases slightly as the level of student need increases, the relationship is loose. Finally, the share of flex item funds that a district receives is primarily based on the share it received near the time the funds were originally flexed; a district’s allocation is not explicitly related to the number of pupils it serves. This allocation method does not account for enrollment growth or declines, and so has the potential to further exacerbate per- pupil funding differences in the long run. The flex item was a reactive rather than proactive policy. Although this temporary program is set to expire in 2014-15, it has already been extended once; the LAO suggests there is “no clear exit strategy,”18 and interviews with California policy-makers suggest it may be diflicult to backtrack from this flexibility.19 With a more strategic design of programs included (e.g., including K-3 class size reduction and excluding adult education) and a mechanism to equalize per- pupil funding rates over time, the flex item could represent a much bigger step toward school finance reform.20 ‘6 SBX3 4 and ABX4 2 established these new rules and extended them to 2014-15. l7Weston (2011), Figure 5. 1" LAO, 2011, pp. 19. l"Fuller, Marsh, Stecher, and Timar (2011), pp. 27 3" LAO (2009, 2010, 2011) and Weston (2011) propose alternative program groupings and analyze these. Education Finance Reform Act (Assembly Bill 18) A more proactive legislative approach to school finance reform is Assembly Bill 18, the Education Finance Reform Act, which would consolidate about 36 categorical programs into just three streams of funding. This proposal meets several of the principles of good finance reform. It simplifies the current system and allows for more local flexibility in how funds are used. The bill also focuses a stream of resources on disadvantaged students, recognizing that their educational cost may be higher. However, the bill fails to address all of the principles of finance reform. Its Targeted Pupil Equity program to help low income and English Learners is susceptible to financial manipulation by school districts because districts define which students are English Learners. Although this bill represents a more thoughtful approach to reform, the funds to be aggregated represent only about $1,000 per pupil in 2010—1 1. Furthermore, the bill does not currently provide a mechanism to equalize the per-pupil amount districts get in each stream. The current disparity in funding across districts in the categorical programs to be aggregated would still exist in the aggregate program. This bill is a work in progress, and future iterations certainly could include an equalizing mechanism. From Program Consolidation to Equal Per- Pupil Funding Rates All these efforts have laid the groundwork for Governor Brown’s recent education finance proposal, which puts many of these pieces together to create a weighted student formula.21 In the Governor’s plan, each student would generate a base funding amount — for example $5,000.22 Disadvantaged students (defined as students either eligible for the free or reduced price lunch program or classified as English Learners) would each generate an additional 37 percent of that base, Revenue per Pupil ($) 3} See the Weighted Student Formula [422, 424]Trailer Bill Language for the Education area of the 2012-13 Governor’s proposed budget, accessed 2/15/12 from http://www.dof.ca.g0v/ budgeting/trailer_bill_language/education/documents/ 31 The bill does not include an actual base amount. The base would be set so that the program’s cost would not exceed what is available through Proposition 98 funding levels. or $1,850 in the example. Once a district’s share of disadvantaged students reaches 50 percent or higher, disadvantaged students generate even more revenue. The additional amount is proportional to the percentage of disadvantaged students in the district. Figure 3 shows the relationship between the percentage of disadvantaged students and the district’s total revenue per pupil, when all revenue is spread out across all students. FIGURE 3: Revenue per Pupil under Governor Brown’s School Finance Proposal 9,000 8,000 - 7,000 — 6,000 '— 5,000 l l l | 0.2 0.4 0.6 0.8 1.0 % Disadvantaged Students 4,000 O Source: Author's calculation based on hypothetical base funding of $5,000 per pupil. This sweeping reform attempts to adhere to the five key principles. If enacted, it would substantially simplify California’s school finance system. Aside from a handful of categorical programs (such as special education and programs beyond the control of the legislature), all state funding would be distributed through this formula with very few strings attached. Schools would be monitored based on what they achieve, not how they spend their money. The plan improves transparency and local flexibility, and the weights target substantial additional resources to districts with needy students. Furthermore, the formula equalizes per— pupil funding across districts with the same cost factors within a six—year time span, so that similar students are treated similarly sooner rather than later. This reform proposal is a dramatic opening bid from the Brown Administration to move California toward a weighted student formula. Representatives from the administration have indicated they would consider adjusting some technical aspects of this 21 22 formula. This bold new proposal has already attracted close scrutiny of its technical details as well as its broader policy changes. During a recent California Senate Budget and Fiscal Review Committee hearing, Senators generally favored the concept of a weighted student formula but questioned some elements of the plan.23At the forefront of their concerns was the lack of cost adjustments for regional differences in labor cost, for student grade level, and for the potentially higher transportation costs facing some districts. The LAO (2012) has pointed out that although districts receive additional funds based on their population of disadvantaged students, districts are not required to spend those supplementary funds on those students. The California State Senate Republican Caucus is concerned about classifying English Learners as disadvantaged students and would like to see cost adjustments based solely on factors outside the control of schools districts.24 (The Governor’s proposal does not double count English Learners if they are also poor, so this concern pertains to the 15 - 25 percent of English Learners who do not participate in the free and reduced price lunch program.) There are related issues about how exactly to measure whether a student is classified as low income and whether the weights for disadvantaged students are appropriate.25 These technical concerns can all be addressed while maintaining the integrity of the proposal. A more fundamental issue, however, is that the Governor’s proposal would drastically change how revenue is distributed, producing winners and losers.26 This issue is particularly salient in the short run. The six-year timeline for phasing in the new funding system means that, if additional state funding is not forthcoming for schools, revenue will be redistributed from some districts to others. Although the Department of Finance projects 23 California Senate Standing Committee on Budget and Fiscal Review Budget Overview Hearing (February 16, 2012), available at http://www.calchannel.com/channel/viewvideo/ 3330. 24 California State Senate Republican Caucus (2012). 25 For more information on these cost factors, see Reinhard, Rose, Sengupta, and Sonstelie (2008) and Rose, Sengupta, Sonstelie, and Reinhard (2008). 26 Rose, Sonstelie, and Weston (forthcoming 2012) analyze these changes. ample revenue growth to insure that six years from now almost all districts will have higher revenue under the new formula than they currently have, California’s recent recession has certainly made some observers wary of optimistic budget forecasts and particularly attuned to the possibility of even more budget cuts. Despite these immediate financial concerns, the state economy is cyclical, and state coffers will eventually rebound and grow. Demographic projections suggest the student population will grow at a slower rate than the adult population. If taxpayers continue to spend the same share of their income on public schools, these concurrent trends suggest per-pupil spending could rise by about 30 percent in the next twenty years.” Yet, even if the new formula is phased in over a longer horizon, some districts will receive substantially less of the additional revenue than others, a feature that will likely cause some of them to challenge the proposal. These challenges, however, stem from the inequities and irrationality inherent in California’s current school finance system — a system that has consistently been assaulted on all fronts by researchers, legislators, the judiciary and the public. The Governor’s proposal is to date the most comprehensive attempt to fix this system. This momentum toward developing a weighted student formula for California is consistent with what the original GDTF research envisioned. Eventually, any finance system will need to come to terms with uncertainty about how resources actually translate into higher academic achievement. But one message is clear: without a stronger finance system, reaching California’s academic goals will be an uphill battle. Pouring more money into the current system is akin to pouring a concrete foundation without putting the form boards in place. It consumes substantial resources, makes a mess, and doesn’t improve the stability of your house. The groundwork has been 27 Rose, Sonstelie, and Weston (2010) provide the assumptions behind this growth rate and model pathways the state could take to transition from the current finance system to a more equitable system over the next two decades. In the scenarios they model, the equalization mechanism funnels new growth in school funding to districts that are currently below their targeted funding rates under the new formula. laid for reform, and it is time to continue working in that direction. A strong finance foundation is the first step in a stronger education system. But even with a rational mechanism for allocating money to school districts, those districts need to use their revenue wisely to improve academic results. Only then can California schools resume their place among the best in the nation. References I Bersin, A., Kirst, M., & Liu, G. (2007). “Getting Beyond the Facts: Reforming California School Finance,” Chief Justice Earl Warren Institute on Race, Ethnicity, and Diversity, University of California, Berkeley. I California Department of Education. (2011). “2010—11 Budget Tables.” Retrieved on February 29, 2012 from http://www.cde.ca.gov/fg/fr/eb/ budtablesZOlO.asp I California State Senate Republican Caucus. (2012, February 15). “Briefing Report: Weighted Student Formula.” Retrieved on February 29, 2012 from http://www.cssrc.us/publications. aspx?id=11860 I Chambers, J., Levin, J., & DeLancey, D. (2006). “Efficiency and Adequacy in California School Finance: A Professional Judgment Approach.” Getting Down to Facts Project: Stanford University. I EdSource. (2010, September). “How California Ranks.” Retrieved on February 14, 2012 from http://www.edsource.org/pub10—how-ca—ranks. html I Fensterwald, J. (2012, January 9). “How weighted funding would work.” Educated Guess. Retrieved on January 10, 2012 from http://toped.svefoundation.org/ 2012/ 01/ 09/ how—weighted-funding—would—work/ I Fuller, B., Marsh, J., Stecher, 8., 8r Timar, T. (2011). “Deregulating School Aid in California: How 10 Districts Responded to Fiscal Flexibility, 2009—2010.” RAND and PACE Research Network. I Governor’s Committee on Education Excellence. (2007). “Students First: Renewing Hope for California’s Future.” I Imazeki, J. (2006). “Assessing the Costs of K-12 Education in California Public Schools.” Getting Down to Facts Project: Stanford University. Legislative Analyst’s Office. (2008). “Analysis of the 2008-09 Budget Bill.” Legislative Analyst’s Office. (2009). “Analysis of the 2008-09 Budget Bill: Proposition 98 Education Programs.” Legislative Analyst’s Office. (2010). “Year—One Survey: Update on School District Finance and Flexibility.” Legislative Analyst’s Office. (2011, February). “The 2011—12 Budget: Year—Two Survey. Update on School District Finance in California.” Legislative Analyst’s Office. (2012, February). “The 2012-13 Budget: Proposition 98 Education Analysis.” Loeb, S., Bryk, A., & Hanushek, E. (2007). “Getting Down to Facts: “Getting Down to Facts: School Finance and Governance in California.” Getting Down to Facts Project: Stanford University. Loeb, S., Grissom, J., 8t Strunk, K. (2007). “District Dollars: Painting a Picture of Revenues and Expenditures in California’s School Districts.” Getting Down to Facts Project: Stanford University. Reinhard, R., Rose, H., Sengupta, R., 81 Sonstelie, J. (2008). “Funding Formulas for California Schools 11: An Analysis of a Proposal by the Governor’s Committee on Education Excellence.” Public Policy Institute of California. Rose, H., Sonstelie, J., Reinhard, R., & Heng, S. (2003). “High Expectations, Modest Means: The Challenge Facing California’s Public Schools.” Public Policy Institute of California. Rose, H., Sonstelie, J., & Weston, M. (2010). “Pathways for School Finance in California.” Public Policy Institute of California. Rose, H., Sonstelie, J., & Weston, M. (forthcoming 2012). “Funding Formulas for California Schools III: An Analysis of Governor Brown’s Weighted Pupil Funding Formula.” Public Policy Institute of California. Rose, I-I., Sonstelie, J., 81 Richardson, P. (2004). “School Budgets and Student Achievement in California: The Principal’s Perspective.” Public Policy Institute of California. 23 24 I Rose, H., & Sengupta, R. (2007). “Teacher Compensation and Local Labor Market Conditions in California: Implications for School Funding.” Public Policy Institute of California. I Rose, H., Sengupta, R., Sonstelie, 1., 81 Reinhard, R. (2008). “Funding Formulas for California Schools: Simulations and Supporting Data.” Public Policy Institute of California. I Sonstelie, I. (2008). “Financing California’s Public Schools.” Conditions of Education in California. PACE, Stanford, CA. I Sonstelie, 1., Brunner, E., & Ardon, K. (1999). “For Better or For Worse? School Finance Reform in California.” Public Policy Institute of California. I Weston, M. (2011). “Funding California Schools: The Revenue Limit System.” Public Policy Institute of California. I Weston, M. (2011). “California’s New Funding Flexibility.” Public Policy Institute of California. ' ive years ago, the Getting Down to Facts (GDTF) Project identified several problems that prevent districts in California from using their human resources more effectively. Since then, little actual progress has been made to solve these problems. At the state level, there have been no significant changes in teacher or leadership policies. If anything, things have gotten worse. The state’s budget woes have led to cuts in training and professional development programs and to layoffs that have reduced and demoralized the educational labor force. However, progress may be near. Pressure from federal initiatives has sparked promising discussions around key human resource issues such as teacher evaluation and incentives. Several individual districts across the state are currently experimenting with new ways to assess, compensate and support their teachers. At a minimum, state policy should focus on removing regulatory barriers to these local efforts and encourage further experimentation. Several studies in the Getting Down to Facts project focused on the personnel in California’s education system and concluded not only that California has fewer professionals per pupil than most other states, but that those adults working in the schools are not always well prepared or supported. Furthermore, state policies often hinder district administrators from identifying and retaining the best personnel for their local schools. In a nutshell, state policies “do not let state and local administrators make the best use of the pool of potential teachers nor adequately support current teachers” (Loeb, Bryk and Hanushek, 2007, page 4). The specific problems identified in the GDTF studies included: ' A relatively low number of adults per student. In 2002—03, California had fewer teachers, administrators and support staff per student, and fewer district-level administrators per school administrator, than most other states. Administrators with relatively poor training. California superintendents and principals received less training and engaged in less professional development than administrators in other states. Difficulties identifying and dismissing weak teachers. When asked what changes were needed to help raise student outcomes, California principals were more likely to identify additional freedom to dismiss teachers than any other factor, and most felt it was almost impossible to remove low—quality teachers with tenure (Fuller et al, 2007). At the same time, few districts had effective processes for evaluating either prospective or current teachers. Most relied upon using easily observable characteristics (such as credentials or experience) that are not necessarily correlated with teacher quality. Systemic flaws in teacher compensation and distribution. Although research has established that generic requirements like 26 educational credits and experience are not well-correlated with teacher quality, almost all districts used these measures as the primary determinant of salary. A closely related problem was inequity in the distribution of effective teachers across schools, with high-need schools facing greater challenges attracting and retaining high-quality teachers. I Reforms inhibited by state policies. Several of the concerns pertaining to the evaluation, assignment and retention of teachers were strongly influenced by district-specific collective bargaining agreements, but these contracts were constrained by state policies that determined what districts can and cannot do. I Lack of program evaluation. Finally, the GDTF investigators noted that even when innovative programs were in place, there was little attempt to assess or evaluate them. Thus, although many of the problems were well— documented, and prior research could suggest avenues for reform, little direct evidence as to whether a particular policy actually works was collected or analyzed. What has changed since 2007 — and what hasn’t Many of the problems identified five years ago continue to pose challenges for California. Schools still have low levels of adults per student, and there has been little progress at the state level in policies related to administrator training or to the hiring, evaluation, compensation or retention of teachers. However, recent initiatives at the federal level have changed the national conversation about measuring and rewarding teacher effectiveness. Within California, a number of individual districts may provide important examples of new ways to evaluate and compensate teachers. A shrinking state budget has forced layoffs and professional development cuts California still ranks close to last among all states in the number of teachers, staff and administrators per pupil. (See Figure 1.) As FIGURE 1: Adults in California Schools, 2004-2011 a 5.0 — c 48 — D 4.6 — \ 271' o 4.4 "— ‘\‘ I o 42 — o n "f 40 .\O h . ~— 0 cu 3 8 /O/‘ \ . Q. ' _ . . ‘ _ 36 __ /A \ OE) ' ‘ I Teachers (1 Os) . o 3'4 _ OAdministrators i’.’ 3.2 r- . . cu A PupII Servrces 0- 3.0 — l l l l | l l 2004-05 2006—07 2008-09 2010—1 1 2005-06 2007-08 2009-10 Source: California Department of Education highlighted in the 2007 GDTF studies, that lower adult-student ratio means larger classes and fewer services, such as instructional coaches or counseling for students. That low ratio also means school principals have to spend more time on issues that in other states assistant principals or district administrators might handle. As a result, principals are less engaged in working on curriculum and teacher development. For example, in one survey California principals were much less likely than principals in other states to report that they “work with teachers to change teaching methods where students were not succeeding” or “work with faculty to develop goals for their practice and professional development” (Darling- Hammond, 2007). In another survey (Fuller et a1, 2007), California principals reported spending as much time on compliance paperwork as they spent on teacher evaluation and support. Furthermore, California principals were less likely to have had an internship as part of their training, to have received mentoring or coaching from more experienced administrators, or to engage in professional development with teachers (Darling—Hammond, 2007). In 2007, California had few programs for leadership development. Five years later, little has changed. Some of the lack of progress can be blamed on the budget crisis. Certainly the budget situation has contributed to layoffs across the state, and many districts have cut back on non-core services such as professional development and training in order to protect core classroom instruction (Fuller et a1, 2010). In this fiscal environment, statewide reforms to improve hiring practices or provide training for new personnel are moot. Several targeted categorical programs related to teacher distribution and professional development were included in the group of programs granted budgetary flexibility in the 2008 budget package,‘ but funding for these programs was reduced by 30 percent. A survey of district financial officers (Fuller et a1, 2012) suggests that in many districts these professional development funds were swept into district general funds to help offset overall cuts and maintain staffing levels and instructional activities. Teacher education gets a closer look Although there has been little progress in professional development for teachers and administrators once they are on the job, there is some reason for optimism that teachers may soon get better training before they enter the classroom. In 2010, a scathing report from the state auditor prompted a shake—up of the California Commission on Teacher Credentialing (CTC). Since then, the CTC has focused on improving oversight of the state’s teacher education programs, particularly through better assessments of the specific skills graduates are supposed to acquire. At the same time, the US. Department of Education has been pushing for reforms in teacher education, including stronger institutional reporting and state accountability. However, all of these efforts were hindered by the Governor’s decision not to fund the California Longitudinal Teacher Integrated Data Education System (CALTIDES). The statewide longitudinal teacher database would have been invaluable for determining which teacher education programs produce effective teachers, as well as where those effective teachers are placed. Teacher effectiveness moves into the national spotlight At the time of the GDTF studies, few districts ‘ For example, among the programs that were flexed were the Professional Development Block Grant, the Teacher Credentialing Block Grant and AB 430 Administrator Training, the only dedicated state training for principals still left. had effective processes for evaluating either prospective or current teachers. Most relied upon easily observable characteristics that are not necessarily correlated with teacher effectiveness. Furthermore, although research has established that generic requirements like educational credits and experience are not well—correlated with teacher quality, almost all districts used these measures as the primary determinant of salary. 'Ihere was therefore no financial incentive for teachers to engage in activities that have been found to improve teacher quality, such as targeted professional development (Loeb and Miller, 2007). ARRA and RTTT. In the last few years, teacher evaluation and compensation have become key issues in education policy across the country, largely driven by initiatives of the Obama administration. One of the four focus areas for the American Reinvestment and Recovery Act (ARRA) and the accompanying Race to the Top (RTTT) competition was “recruiting, developing, rewarding, and retaining effective teachers and principals, especially where they are needed most.” More specifically, state applications needed to address several related reform efforts, including “attracting and keeping great teachers and leaders in America’s classrooms, by expanding effective support to teachers and principals; reforming and improving teacher preparation; revising teacher evaluation, compensation, and retention policies to encourage and reward effectiveness; and working to ensure that our most talented teachers are placed in the schools and subjects where they are needed the most” (White House RTTT Fact Sheet). A related focus area was the development of statewide longitudinal data systems, which the Obama administration has emphasized are necessary for providing the type of data that will allow standards— based teacher evaluation. To qualify for ARRA funds, states had to provide “assurances that they are advancing the four reforms described in the statute and complying with maintenance of effort requirements.” States also needed to provide baseline data on their current status in each of these areas and basic information on how the funds will be used. In addition, to compete for RTTT funds, states needed to adopt systems that tie educator evaluations to 27 28 personnel decisions such as hiring, compensation, tenure and dismissal. Value-added measures of teacher effectiveness. A key part of the discussion about evaluation is how best to measure teacher effectiveness, an issue that education researchers have long debated. Loeb and Miller (2007) point out that there is little evidence that observable teacher characteristics like education or experience are well correlated with a teacher’s ability to improve student outcomes. Thus, education researchers have increasingly turned to measures based directly on student outcomes on standardized tests. These value-added measures attempt to capture the year-to-year improvement in a student’s test scores that can be attributed to that student’s teacher, controlling for other variables that may also impact test scores but are outside the control of the teacher. Value-added measures have rarely been used by schools for high-stakes teacher evaluation, in part because the methodology is complex but even more because of debates about the appropriateness of basing teacher evaluations on student test scores. This debate became front page news in 2008 when the Los Angeles Times published value-added scores for thousands of teachers in the Los Angeles Unified School District (LAUSD). Aside from issues of privacy, critics argued that even when done well, value-added scores can be volatile. For example, it is possible that a teacher who is at the top of the distribution in one year may be at the bottom the next year. Critics also maintain that value-added measures cannot fully account for the myriad factors that influence test scores but are outside a teacher’s control. Another common criticism is that test scores do not fully represent all that good teachers do: Is an effective teacher simply one who raises test scores? The LAUSD controversy sparked extensive discussion across the state about the benefits and drawbacks of these test—based measures. That discussion has been particularly salient in the face of the federal push to include such measures in any system of teacher evaluation. Even strong supporters of value-added measures are unlikely to argue that they should be the sole criterion for determining teacher effectiveness. However, to do well in the RTTT competition, a state needed to show that a ‘significant part’ of teacher evaluations will be based on test scores or other measures of student growth. It is unclear what percentage would be considered ‘significant,’ and evaluations must also include multiple other measures. The national conversation on teacher evaluation hasn’t reached Sacramento Although most changes to teacher evaluation, compensation, retention and assignment would need to be negotiated at the district level, one of the key findings of Getting Down to Facts was that state policies could do much more to give districts flexibility at least to try new approaches. For example, the Education Code dictates minimum requirements for frequency and scope of teacher evaluations, as well as the process and criteria for dismissal and layoffs. Although the federal push for teacher evaluation has increased the attention paid to these issues, relatively little of substance has been done at the state level in California. A special 2009 legislative session on Race to the Top did result in bills that would allow longitudinal data to be used for teacher evaluation, something that had been explicitly prohibited in earlier statutory language.Z There have been a few other attempts to change state laws about teacher evaluation, but nothing of any significance has passed. For example, in the 2009—10 session, SB955 (Huff) would have made major changes to teacher evaluation, assignment, layoffs and dismissal by allowing districts to base layoffs on performance evaluations rather than seniority. That bill was re—introduced in the 2011- 12 session as SB355, but it died in committee. AB 5 (Fuentes) and SB 257 (Liu) would also require changes to teacher evaluation, including using multiple measures and giving some weight to measures based on student test scores.3 Both were introduced in the 2011-12 session; as of this writing, their futures are uncertain. 2 The Special Session also produced a bill that established the Technology, Engineering, Math, and Career Technical Education Educator Credentialing Program, which would provide alternative routes to licensure for STEM teachers, another provision to qualify for RTTT. 3 AB 48 (Perez) died in committee. Action at the state level may become even more important in the wake of a lawsuit aimed at holding districts accountable for existing requirements for teacher evaluation. The Stull Act, originally passed in 1971, is the primary state law governing teacher evaluation. It specifically requires that schools evaluate teachers based, at least in part, on student performance on state tests. According to EdVoice, an education advocacy group, few districts are in full compliance and a lawsuit (Doe v. Deasy) is now pending against LAUSD to force compliance. The success of such a suit would have ramifications for other districts as well. It will be important for the state legislature to provide clarification of exactly what is, and is not, required of districts. One concern that has been raised in California about value—added measures is that the tests being used to calculate them were never intended to be used in this way. The California Standards Tests were not designed to compare student performance from one grade to the next, a very real problem in using them for value—added models. However, California is now part of the SMARTER Balanced Assessment Consortium, a group of states working on assessment systems aligned with the Common Core. These efforts may eventually lead to more accurate measures of the value added by teachers. Teacher distribution is still inequitable Critics of the traditional step-and-column teacher salary schedule point out that not only does the system base compensation on factors that have no proven correlation with teacher quality, it also contributes to inequities in the distribution of effective teachers across schools within districts. When teachers with identical education and experience earn the same pay across a district, there are likely to be shortages in some fields and at schools with more challenging working conditions. Although this is often blamed on the transfer and seniority rules in collectively-bargained contracts, Koski and Horng (2007) found no evidence that those rules affected the within—district distribution of teachers. Their findings suggest that policy— makers need to look elsewhere for ways to address the difficulties that some schools face in hiring and retaining effective teachers. Although layoffs due to budget cutbacks have prompted renewed concern about the role of seniority in teacher retention and transfer decisions, California has done little in this area either. In 2010, parents at three Los Angeles schools brought a suit against the state and LAUSD, arguing that their schools suffered disproportionately from layoffs (Reed v. State of California and LAUSD). The plaintiffs’ schools served large numbers of minority and low-income children and had many novice teachers who were first in line for layoffs. Ultimately, an agreement was reached in which the district agreed to exempt teachers at 45 schools from the ‘last hired, first fired’ layoff rules. SB1285 (Steinberg) attempted to address this problem statewide by requiring districts to balance distribution of new teachers so that the percentage of teachers laid off at a district’s lowest-performing schools would be no higher than the average for all schools in the district. However, after an intense battle the bill was defeated. The one state-level policy change aimed at improving the distribution of teachers within districts took place in 2006. SB 1655 (Scott) allows principals of schools in the lowest 30 percent to refuse voluntary transfers and does not allow seniority to be the dominant priority over other qualified applicants for transfers after April 15. Pockets of Progress: Districts are taking the lead in innovation Although Sacramento has done relatively little on issues of training, evaluation and distribution of personnel, several individual districts have been developing programs on their own. For example, Los Angeles Unified Superintendent Iohn Deasy has made teacher and administrator evaluation a top priority; LAUSD is currently piloting an evaluation system that incorporates multiple measures. One component of the evaluation is academic growth over time, a form of a value- added test score measure. The evaluations also include direct observations, stakeholder feedback from parents and students, and contributions to community, although the district is still working out how exactly to capture each of those factors and how much weight to give to each. The district is starting with a small voluntary pilot and plans to scale up over time. Any change to the evaluation 29 30 process district-wide will need to be negotiated with the local teachers union, and such a radical change will likely be contentious. San Francisco Unified School District (SFUSD) is taking a very different approach. In 2008, San Francisco voters passed Proposition A, the Quality Teacher and Education Act. Proposition A is a parcel tax largely dedicated to increasing teacher salaries, including bonuses for hard-to-staff schools and subjects. It also supports other reforms to teacher training and evaluation. Under the Act, teachers are evaluated more often, receive more professional development and have an opportunity to serve as Master Teachers. Changes were also made to the Peer Assistance and Review process. Less extreme than the changes proposed in Los Angeles, these changes to evaluation are part of a larger district Human Capital Initiative to improve “the way SFUSD selects, trains, evaluates, and supports staffing with particular emphasis on teachers” (SFUSD, 2011). There has been considerable effort to build support from all the relevant stakeholders, including the union and local community. It is particularly worth noting that both LAUSD and SFUSD have teamed up with outside evaluators to provide objective analysis of their initiatives. LAUSD has contracted with a team of researchers from the University of Southern California, and SFUSD teamed with PACE to evaluate the implementation of San Francisco’s policies (Hough et al, 2011). As these programs evolve, policy-makers will have solid evidence about which aspects are most effective. Both Los Angeles and San Francisco are also part of the California Office to Reform Education, a consortium of reform-minded districts that applied for the second round of Race to the Top funding. The other five were Long Beach, Sacramento, Fresno, Clovis and Sanger, and Oakland has since joined as well. These eight districts are now working 4The College-Ready Promise (TCRP) is a coalition of five Charter Management Organizations: Alliance College-Ready Public Schools, Aspire Public Schools, Green Dot Public Schools, Inner City Education Foundation Public Schools, and Partnerships to Uplift Communities. The REACH Consortium is led by ARISE High School, together with the Bay Area School of Enterprise, Lighthouse Community Charter School, and Lighthouse Community Charter High School. together to implement innovative strategies in several areas, including teacher recruitment, preparation and evaluation. Similarly, two different consortia of charter schools“ recently applied for and received Teacher Incentive Fund grants from the US. Department of Education. These grants support the development and implementation of new systems of evaluation and compensation for teachers and administrators in high-need schools. Northern Humboldt Union High School District and Lucia Mar Unified School District also received TIF grants. Policy recommendations The original GDTF project made several recommendations to improve the training, evaluation and compensation of personnel. Most are still relevant today. For example, the state could explore developing support networks for principals, or career ladders for teachers that would move excellent teachers into roles as instructional leaders. Alternative training programs and stronger assessment of teacher education programs could also promote a larger and more effective teaching force. At the very least, the state could make it easier for districts to experiment with different systems of evaluation and compensation, such as those already happening in San Francisco and Los Angeles. Although budget constraints are often invoked as a barrier to reform, many of the needed reforms at the state level require regulatory changes more than funding increases. For example, changing the Education Code so it is easier for districts to experiment with performance-based transfer and dismissal policies and clarifying the teacher evaluation requirements of the Stull Act are issues of policy, not appropriations. Encourage district experimentation — and be sure to evaluate it. Perhaps the simplest action Sacramento policy—makers could take immediately would be to track the many district— led initiatives already taking place around the state, and to push for more evaluation of those programs. For example, a clearinghouse that tracks district initiatives would help principals and superintendents learn of ongoing efforts and connect with other administrators who are developing similar programs. The State Board of Education took a step in this direction last year when it voted to create an online database to share information about local, state and national efforts to measure educator effectiveness (Watanabe, 2010). Compiling information about initiatives around the state is important. Assessing and evaluating these initiatives is just as important. As individual districts experiment with innovative new systems, we need to know if and how well these programs are working. Many districts do not have the internal resources or expertise to conduct these evaluations themselves, but the Department of Education could either assist districts with evaluation directly or help match districts with independent organizations, such as PACE, the RAND Corporation, WestEd, or any of the many other research organizations in California qualified to conduct this sort of analysis. In addition, longitudinal data along the lines of the proposed CALTIDES database are essential for assessment of these policies. In many ways, the situation in California is not much different today than it was five years ago. California still needs to put in place better policies to attract and retain high—quality teachers and administrators, and to learn from the effects of policies that are implemented. Still, there is clearly some momentum in the state on issues of teacher training, evaluation and compensation. There is also a new governor, a new state superintendent, and new leadership on both the State Board of Education and the Commission on Teacher Credentialing, all of whom have expressed willingness to consider policies that had little traction previously. The conversations about measuring teacher effectiveness and tying those evaluations to compensation, sparked by the federal initiatives, are a good beginning. They must continue if concrete progress is to be made. References I Darling-Hammond, L., 8: Orphanos, S. (2007). “Leadership Development in California.” Getting Down to Facts Project: Stanford University. I Fuller, B., Loeb, S., Arshan, N., Chen, A., & Yi, S. (2007). “California Principals’ Resources: Acquisition, Deployment and Barriers.” Getting Down to Facts Project: Stanford University. I Hough, H., Loeb, S., & Plank, D. (2011). “The Quality Teacher and Education Act: First Year Report.” PACE. Retrieved from http://www. stanford.edu/group/pace/PUBLICATIONS/ TECHNICALREPORTS/ 201 1_MAY_QTEA. pdf I Koski, W., 8r Horng, E. (2007). “Curbing or Facilitating Inequality? Law, Collective Bargaining and Teacher Assignment Among Schools in California.” Getting Down to Facts Project: Stanford University. I Loeb, S., Bryk, A., St Hanushek, E. (2007). “Getting Down to Facts: School Finance and Governance in California.” Getting Down to Facts Project: Stanford University. I Loeb, S., & Miller, L. (2007). “A Review of State Teacher Policies: What are they, what are their effects and what are their implications for school finance?” Getting Down to Facts Project: Stanford University. I San Francisco Unified School District (2011). Human Capital Initiative project status report. Retrieved from http://wwwsfusd. edu/en/assets/sfusd-staff/about-SFUSD/files/ initiatives%20and%20plans/files/Human%20 Capital%20Initiative%20Update_Final_04021 1. pdf I Watanabe, T. (2010, September 16). “California Board of Education addresses teacher evaluation issue.” Los Angeles Times. Retrieved from http://articles.latimes.com/2010/sep/ 16/ local/la-me—ed-board—ZO100916 I White House Race to the Top Fact Sheet. Retrieved from http://www.whitehouse.gov/ the—press-oflice/fact—sheet-race-top f there was one thing that the Getting Down to Facts (GDTF) studies made clear, it was that California was in urgent need of a comprehensive educational data system. We The authors of the GDTF studies were themselves handicapped by the lack of accessible, reliable data on key features of the state’s education system. Beyond this, though, the authors agreed that the creation of a robust data system was a necessary condition for sustained improvement in the performance of California’s schools and students. More and better data were needed both to measure performance and to guide improvement. The GDTF findings were endorsed and amplified by the Governor’s Committee on Educational Excellence (http://www. everychildprepared.org ), which characterized clear, accurate, and reliable data as the “cornerstone” of a continuously improving education system. In the view of the GCEE, the development of a comprehensive educational data system “must go forward if any real reforms are to be possible” (bold in the original). When it came to implementing the GCEE recommendations, building California’s data system was supposed to be the easy part. Even as California’s economy deteriorated and political support for the wide—ranging reforms proposed by GCEE weakened—notably in Governor Schwarzenegger’s Office—a comprehensive educational data system remained an apparently attainable goal. Improving data collection, access and use would at least provide the basis for diagnosing problems, and ultimately guide the effort to make California’s education system a learning system capable of sustained, continuous improvement. Five years later California has made some progress toward a comprehensive educational data system, but far less than GDTF or GCEE hoped for or expected. On the one hand, California’s longitudinal student data system (CALPADS) has been up and running for two years, and will soon be fully operational. The completion of CALPADS greatly increases the quantity of reliable information on the performance of California schools and students, including the state’s first accurate measures of drop-out and graduation rates. Student achievement data were added to the system in 2012, which makes it possible to ask and answer additional questions. On the other hand, though, CALPADS represents a far more limited data system than anything foreseen by GDTF or GCEE, and further progress toward a comprehensive system seems unlikely. As the drive to build a comprehensive educational data system has stalled, however, a very different policy conversation has begun. Governor Brown has sought to shift the focus from the statewide system foreseen by GDTF and GCEE towards a system that gives priority to the data needs of local educators and parents. Whether the Governor’s new approach will bear more fruit than past efforts to build a comprehensive system remains to be seen. A comprehensive data system is a necessary condition for improving California’s schools, for two main reasons. First, citizens and public officials need reliable data to hold educators accountable for their performance. California has set ambitious goals for its education system, but without a comprehensive data system it is difficult to ascertain whether schools and students are achieving these goals. Without reliable data on enrollment and attendance, teacher licensure and assignment, and local resource use, the state cannot fairly identify schools and classrooms where students are falling short of performance goals and hold them to account. In addition, California needs a comprehensive education data system to help guide long—term improvement in educational performance. At present very little is known about which policies and programs are effective in moving students toward the goals set for them, and little more can be learned unless policies are designed in ways that support policy learning and the systematic evaluation of policy alternatives. A robust and reliable data system can help to target scarce educational resources to effective programs and away from those that are less successful. Designing a Comprehensive Data System Following the publication of the GCEE recommendations an improved data system rose to the top of the education policy agenda, with strong support from the Governor, the California Department of Education (CDE), major foundations and non—profits, and key legislators. Strengthening data collection and expanding data access and use emerged as key priorities. A pair of PACE reports, “Building an Information System to Support Continuous Improvement in California Public Schools” and “Continuous Improvement in California Education: Data Systems and Policy Learning” (both available at http://www.edpolicyinca.org) identified the creation of a comprehensive education data system as a necessary condition for the continuous improvement of education in California. The two reports described what such a system might look like and how it might be used. A report from McKinsey 8: Company, commissioned by Governor Schwarzenegger and Superintendent Jack O’Connell and funded by the Hewlett and Gates Foundations, was published at the end of 2008. With extensive input from a stakeholder task force of educators, policy-makers and researchers the McKinsey report laid out a far more ambitious and detailed plan for California’s educational data system than either GDTF and GCEE had done (http://foundationcenter.org/ educationexcellence/report.jhtml?id=fdc75300004). McKinsey foresaw the development of a comprehensive data system unfolding in three phases following the completion of basic data infrastructure, which the authors characterized as “Step Zero:” 0) Complete the longitudinal student and teacher data systems (CALPADS and CALTIDES). 1)Enhance the quality, accessibility, completeness, and basic use of current K-12 data systems, including the development of user-friendly interfaces and reports. 2)Expand the use of K-12 education data by building more advanced systems that encourage collaboration and best—practice sharing for instruction, administration, and other district functions; provide standard ways to evaluate local, state, and federally funded programs; and improve educator and administrator recruiting, effectiveness, professional development, and retention. 3)Create interagency linkages to improve decisions using data beyond K-12, including higher education and workforce data, and data from other social service agencies including pre—K, foster care, health, and criminal justice. The vision of a comprehensive educational data system was given legislative life in SB 1298 (Simitian and Steinberg, sponsored by Children Now) in 2008. SB 1298 established a statutory framework for the continued development, governance, and use of California’s data system. Key provisions included the creation of a working group to develop a governance structure for the system, the development of a plan to guide implementation, and the introduction of a common student identifier to facilitate data linkages and information sharing across the multiple segments of the state’s fragmented education system. With 33 34 the approval of SB 1298, the foundation appeared to be in place for rapid progress on data issues in California. Getting to Zero In the years since 2008 California has made significant progress towards the development of a comprehensive educational data system. One set of benchmarks was established by the Data Quality Campaign (DQC), which has worked for a decade to encourage all states to build robust longitudinal data systems. DQC identified ten essential elements of a comprehensive educational data system, and conducted annual surveys to monitor how states were doing in implementing these elements. In 2006, when the GDTF studies were conducted, California had implemented only four of the DQC’s essential elements. Five years later, like virtually all other states, California had implemented all ten. Since 2009—10 CALPADS has provided comprehensive information on enrollments, including statewide graduation and drop-out rates. In 2010—11, data on teacher assignments and the performance of English Language learners were collected for the first time, and three years of student level achievement data (from CAHSEE, CELDT, and STAR) were recently included as well. The final data elements, including student grades and credits, suspension and expulsion records, and data on high school program placement, will be incorporated in 2011—12, at which point CALPADS will be “complete.” The completion of CALPADS allows the state to track students over time, through their educational careers. It makes it possible to track students who move from one school district to another, which is necessary for the accurate calculation ofgraduation and drop-out rates, and also to link teachers to the students they teach. The additional data elements that are now being incorporated will greatly enrich our understanding of students’ experiences in high school, and how these affect their access to postsecondary education. These are real advances, but the completion of CALPADS gets California only halfway to the starting gate on the path toward a comprehensive educational data system. The other half of McKinsey’s Step Zero was the completion of the state’s teacher data system, CALTIDES. In parallel to CALPADS, CALTIDES was expected to collect detailed longitudinal information on teachers and other certificated employees, including data on their training, licensure, salary, and assignment to schools and classes. These data would support efforts to answer questions about the efficacy of different approaches to teacher preparation, for example, or about the prevalence of out-of- field assignments for teachers in schools serving low-income students. Full information about California’s teachers is a necessary complement to information about students in any effort to evaluate the effects of different policies and programs, which is in turn essential to long—term improvement in the performance of schools and students. In August 2011, however, Governor Brown vetoed state funding for CALTIDES and declined to apply for federal funding to link K- 12, higher education and workforce data, thus effectively halting the further development of California’s educational data system. This does not undo the progress that has been made with the completion of CALPADS and the implementation of statewide educator identifiers (SEIDs), but it means that California’s effort to build a comprehensive education data system is for now stalled mid-way through Step Zero. Key provisions of SB 1298 have yet to be put into full effect, and policy action continues to focus almost entirely on compliance with federal directives and requirements rather than strengthened accountability or system improvement. Local Initiatives As the state falters in its efforts to build a comprehensive educational data system, some local school districts are moving to take up the slack. In 2008 PACE worked with the research offices in a consortium of eight urban school districts to calculate drop—out rates that were more accurate and reliable at the time than those produced by CDE. Districts including Long Beach, Fresno and San Francisco have deepened these efforts in the years since, committing themselves to use data to support organizational learning and to guide local decision—making. The seven California school districts that have come together under the auspices of the California Office to Reform Education (CORE) are working to create a “federated” data system that could serve as an example to districts across the state. These efforts are impressive and have contributed to significant improvements in the performance ofschools and students in participating districts. Nevertheless, they cannot take the place of a comprehensive educational data system at the state level, for several reasons. First, and most obviously, a state system is necessary to keep track of the very large number of students who move between districts, or from districts to private or charter schools. School districts only have data on the students enrolled in their own schools. Without a statewide data system it is impossible to calculate accurate graduation and drop—out rates because local districts cannot know whether students who leave have dropped out or enrolled in another part of the education system. When students graduate, most local districts lack the ability to track them into further education or employment. Second, most California school districts do not have the resources or the personnel to build or maintain truly comprehensive local data systems. Even when they do, they cannot match a statewide system when it comes to tasks including, among many other things, federal reporting, student records transfer, and the immediate identification of special needs and elimination of redundant testing for transfer students. The CORE districts and a few others have committed themselves to rely on data to guide local decision—making and inform educational practice, but even their efforts have been hobbled by the state’s ongoing budget crisis. In most districts the collection and reporting of data beyond those required for compliance with state and federal mandates is a luxury they can no longer afford. Third, even if all 1000 California school districts were to make data systems a priority, it is both foolish and costly to rely on each district to separately acquire or upgrade their systems and develop their own tools and procedures for data collection, use, and reporting. Many elements of a comprehensive educational data system would clearly be common to all school districts, and it makes sense for the state to develop these elements and share them with local authorities rather than depending on variable local commitment and resources. Finally, beyond the efficiencies of a fully developed system, and the demands of state and federal reporting and accountability, state officials have questions about educational policy and finance that are quite different from those posed at the local level. State officials, researchers and the broader public need access to common information across all schools. They need to identify statewide trends and monitor the effectiveness of state programs and investments. They may want to know about the relative effectiveness of charter schools when compared to traditional public schools, or about the effectiveness of different teacher training strategies in preparing excellent teachers for California’s classrooms. Answers to these questions are essential to long-term improvement in the performance of schools and students in California, and the answers can only come from a comprehensive state-level data system. California Partnerships for Achieving Student Success (CAL—PASS) illustrates both the possibilities and the limitations of what can be accomplished at the local level. CAL-PASS maintains records on the educational experiences and academic performance of California students as they move from K-12 to postsecondary education. Schools, colleges and universities provide data to CAL-PASS on a voluntary basis. These data are shared across institutions under the auspices of inter—segmental learning councils, which use the data to inform their discussions about curriculum, instructional practices and student performance. The work of these regional learning councils has led to closer alignment between K— 12 and postsecondary educators, to improvements in curriculum and instructional practice, and to increased student success in many parts of California. At the same time, however, reliance on voluntary agreements and regional councils imposes strict limits on what CAL-PASS can accomplish, and on the number of students who benefit from its work. CAL-PASS provides an example of how a comprehensive educational 35 36 data system can support improvement in the performance of schools and students, but it also illuminates why the state must ultimately take the leading role in building such a system. Getting Past Zero Building upon its 2006 set of essential elements, a recent report from the Data Quality Campaign (DQC) identifies ten additional policy actions that states should take as they move to build comprehensive data systems. (See Table 1.) TABLE 1. DQC Essential State Actions California reports progress on four of these, but clearly the hard work remains to be done as the state works to build a data system that can support teachers, schools, and school districts and improve the quality of schooling. Despite recent setbacks, there are three critical steps that California can take now to move the state farther along the path laid down by the McKinsey report toward the comprehensive educational data system foreseen by GDTF and GCEE. Link Data Systems Create Stable, Sustained Support Develop Governance Structures Build State Data Repositories System—wide Improvement Efforts Develop P-20/Workforce Research Agenda Implement Systems to Provide Timely Access to Information Create Progress Reports Using Individual Student Data to Improve Student Performance Create Reports Using Longitudinal Statistics to Guide Promote Educator Professional Development and Credentialing Promote Strategies to Raise Awareness of Available Data X 1. Build a data warehouse The first critical step is to create a data warehouse that links and integrates data collected by the multiple segments of California’s fragmented education system. One key goal of a comprehensive educational data system is to track students through and beyond their educational careers, with data that provide rich information about their experiences in school and after. This requires the creation of a data warehouse to compile and link data from multiple sources including the pre—K, K—12, and postsecondary education systems. California collects vast quantities of educational data, but at present far too little of it is of any use to parents, educators, or policy—makers. In 2005, when the GDTF studies were underway, the CDE alone was conducting 125 annual data collections in K—12 schools, and the number has barely declined in the years since. For the most part, though, these data remain “siloed,” in that information collected by an office or agency for one particular purpose is difficult or impossible to link to data collected by other agencies for other purposes. In CALPADS, for example, the only data elements on students that are included are those required by the provisions of the federal Elementary and Secondary Education Act. CDE collects additional data on students to comply with other federal and state requirements, but these are not included in CALPADS and can be linked to CALPADS data only with difficulty, if at all. With the demise ofCALTIDES, California has no centralized repository ofinformation on teachers. The Commission on Teacher Credentialing (CTC) collects information on licensing and authorizations, but linking teacher authorizations to assignments remains difficult and data on teacher pay are not collected by the state. It is possible to link teachers to their students through CALPADS, but without richer information on the teachers themselves these links are of little interest or value. The three public segments of California’s postsecondary education system collect their own data, as do the providers of early childhood and preschool programs. The California Postsecondary Education Commission (CPEC) previously maintained a warehouse of data from postsecondary institutions, but Governor Brown eliminated CPEC in 2012. The California Community College Chancellor’s Office (CCCCO) has for now agreed to assume responsibility for managing postsecondary education data, but this is a temporary arrangement for which CCCCO receives no funding. Discussions with CDE about the integration of postsecondary education data with data from CALPADS in a K—20 data warehouse remain at a very preliminary stage. A key technical issue in the creation of an effective data warehouse is the assignment of a unique student identifier that remains with the student as s/he moves from school to school and into postsecondary education or employment. To facilitate linkages between data collected in the education system and data from other sectors (e.g., employment, social services, corrections) the identifier would ideally be based on the student’s social security number, or a robust alternative, which would remain with the student throughout his or her life. An encrypted version of students’ social security numbers is already in use in California’s postsecondary education systems and by CAL-PASS, but not by CDE. Reliance on students’ social security numbers to generate identifiers raises privacy concerns in California, particularly for recent immigrants and other young people who lack documentary proof of citizenship. As an alternative CDE is therefore generating unique student identifiers for students that in principle can support data linkages between K— 12 and postsecondary education. This is better than nothing, but the CDE identifiers do not move easily from school to school, or from K-12 to postsecondary education, and connections to workforce, corrections, and social service data are for now beyond reach. 2. Facilitate access to data A second critical step for the state would be to provide a more effective set of tools to make data accessible and useful to local educators, parents, and other constituencies in the education system. Bringing data from multiple sources together in a common warehouse would be a big advance toward the comprehensive educational data system foreseen by GDTF and GCEE. Compiling data in common formats in a single place would make them more readily available to policy audiences, providing necessary support for the rigorous evaluation of alternative policies and practices that is essential to continuous improvement in the performance of California schools and students. In the absence of user-friendly tools for making sense of these data, however, they will be of little value for those with the greatest immediate interest in learning how schools and students are performing. For K—12 schools, the CDE has taken some important steps to make data more useful and accessible. Many important data elements are currently made available to parents and others in the School Accountability Report Cards (SARCs) that schools are required to produce each year. The SARC includes data on enrollment, student achievement, graduation and drop-out rates, and teacher qualifications, among many other things. Additional information can be obtained by using the CDE’s “DataQuest” tool, which allows users to ask further questions about individual school districts and schools. Some of these capabilities are being further developed by private- sector providers, and by California Business for Educational Excellence (CBEE), which make use of state data to produce customizable reports for parents, schools and districts. The state can do a great deal more, however, by expanding the array of information that is provided and enhancing the tools that it now makes available to access and analyze school and district data. One key step that could be accomplished at little cost would be to standardize data presentation 37 38 on the SARC, and to bring the school-level data included on the SARC into a common archive that would allow parents and others to compare schools’ performance on key indicators. Developing user- friendly data dashboards for schools and districts would help to make resource allocation and use transparent to parents, educators, communities, and policy-makers by providing school-level information on revenues and spending (as already happens in Oakland). It is not impossible for parents and others to find answers to such questions now, but additional work by the state would make it easier and would also make the answers more reliable. 3. Design policies to support policy learning Building a data warehouse and developing tools for easier data access and use would go some distance toward the goal of supporting continuous improvement in the performance of schools and students. But to answer the questions that are most important for achieving this goal, California needs to leverage policy change to learn which policies and practices work and which don’t, and under which circumstances. Legislators and local administrators need to design and implement educational policies in ways that support policy learning. This is not what California does now. On the one hand, the state often makes dramatic changes in education policy that affect all schools simultaneously, as with class-size reduction, which precludes any opportunity to learn whether the new policies have any independent positive impact on the performance of schools and students. On the other hand, many policy changes are little more than random acts of innovation, developed and implemented in a single district or school, where they are seldom if ever subjected to rigorous evaluation. The consequence is that policy learning rarely occurs, and when it does there are no effective mechanisms to ensure that what is learned in one school or school district is shared with others. Designing policies to support policy learning would involve far greater reliance on pilot projects and policy experiments, in which innovative practices or programs are implemented in a limited number of schools or classrooms and evaluated to measure their effectiveness. Those that produce desired effects can then be extended to additional schools and classrooms, while those that do not can be abandoned. Policy learning was one of the key goals of the Quality Education Investment Act (QEIA), but flaws in the design and implementation of QEIA imposed strict limits on how much was learned. Conclusion Both GDTF and GCEE aflirmed that a comprehensive educational data system was a necessary condition for long-term improvement in the performance of California schools and students. The state has made some progress toward this goal since 2007, but far less than expected, and the obstacles to further progress are daunting. In the meantime Governor Brown has argued on behalf of an approach to education data that focuses on information of immediate use to parents, teachers, and local administrators in their work to expand opportunities and improve outcomes for their students, rather than the concerns of researchers and policy—makers. California is thus in a transitional moment, halfway through Step Zero on the path toward the GDTF vision of a comprehensive system and simultaneously approaching the starting gate on the path toward a system that would favor timeliness and local relevance over comprehensiveness and standardized measurement. It is too early to say which of these visions is more likely to prevail, but it is unarguable that for now the persistent lack of useful educational data continues to handicap all efforts to improve the performance of California schools and students. EEEEEEEEEEEEEEEEE IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII Cll'filz'l '476'4 Policy a Analysis for California PACE Education http://www.edpolicyinca.org