College and Research Libraries MELVIN J. VOIGT Acquisition Rates in University Libraries While the allocation of book funds in university libraries and the size of the ultimate collection have received considerable attention in the literature, suggestions for determining the annual acquisition rate have related primarily to numbers of students or .the size of the col- lection, neither of which is satisfactory. A model for determining acquisition rates for currently published material for universities with extensive doctoral programs is developed. A basic rate is estab- lished for the library of a university with a specifically defined pro- gram. The rate is then modulated, based on variations in programs from the university as defined. DuRING THE PAST FIFTY YEARS, the li- brary literature has been replete with ar- ticles on the allocation of book funds in university libr~ries, but very little has been written on annual funding for book acquisitions or how to calculate and justify the acquisition rate neces- sary to provide adequately for academic and research requirements. The lack of discussion on determination of annual funding requirements is curious, for, as Rogers and Weber point out, "Of all policy issues confronting a university library, none is more central to the na- ture of the institution or has greater single impact on costs than the program for collecting books and other informa- tional materials."1 The size of collec- tions · and their exponential growth have been given attention both in textbooks and in the periodical literature but have been approached more often in terms of holding them in check rather than ~!elvin ]. Voigt is university librarian, University of California, San Diego. in meeting academic research require- ments. The preponderance of literature on university libraries today relates to the quite necessary objectives of coopera- tion and networking to improve access, while slowing down accelerating costs. There is danger that enthusiasm in this direction may result in a weakening of individual research libraries to a degree that advanced graduate programs and research are no longer possible. Dix, in a paper presented to the National Com- mission on the Financing of Postsecond- ary Education on behalf of the Associa- tion of Research Libraries, after an ad- mirable discussion of the importance of research libraries, devotes the sug- gested action part of his paper almost entirely to methods of achieving econo- mies and does not discuss the need for continued growth to maintain quality collections at every major university. 2 It is most important that attention be paid to quantifying the accession rates essen- tial to maintaining collections that will support advanced graduate education I 263 264 I College & Research Libraries • July 1975 and research. This paper describes a model developed for this purpose. SIZE AND GRowTH oF CoLLECTIONS Many writers have made generalized statements on the importance of contin- uing development of the university li- brary's collections. As Downs states, "A book collection that has stopped grow- ing is a dead collection and soon loses most of its interest and value for the scholar."3 . Major emphasis, however, has been on how large a collection is need- ed, giving the impression to many a naive university administrator that once the goal of x million volumes is reached, acquisitions can stop. The most often quoted attempt to quantify college and university collec- tion requirements, the Clapp-Jordan formula, 4 deals exclusively with collec- tion size and does not consider acquisi- tion rates. In the study of Ohio institu- tions from which the article was devel- oped,5 Clapp and Jordan did consider the "amount of annual book fund" and recommended an acquisition rate of 6 percent, thus following the lead of oth- er writers in an expectation of continu- ing exponential growth. The 6 percent was calculated from Rider's finding that research libraries doubled in size approximately every sixteen years. Clapp and Jordan quote Rider's statement that "this may be as- serted as almost axiomatic: unless a col- lege or university is willing to be stag- nant, unless it is willing not to maintain its place in the steady flow of education- al development, it has to double its li- brary in size every sixteen years, or thereabouts."6 In correspondence Ver- ner Clapp stated that the annual book fund discussion "was omitted from the published article as likely to add unnec- essarily to sufficient trou~le."7 In an ear- lier letter Clapp stated that "when pub- lished in C&RL (with numerous er- rors), the formula claimed nothing be- yond an attempt to demonstrate a meth- od. . . ."8 He considered the numbers used as experimental, expected that they should vary when applied to different educational institutions or groups of institutions, and was unhappy that they were generally accepted as gospel with- out further testing. The most thorough analysis of the Clapp-Jordan formula is that of Mclnnis,9 who found it diffi- cult to determine rational collection size through empirical analysis, indicating that collection size is not the major con- sideration. It has always seemed to this writer that in the university situation, when a reasonable start has been made in build- ing an adequate research collection, first consideration must be given to main- taining a satisfactory acquisition rate- a continuous flow of currently pub- lished material-and that ultimate size must become a secondary concern. Over an extensive time period, size has always been a meaningless term, for as long as scholarship continues and results in printed material or other physical sub- stitutes for print, such material will be added to research libraries. The amount of these current acquisitions to be per- manently housed and retained on cam- pus will depend on an institution's abil- ity to add library space, availability or ability to construct easily accessible stor- age, and on ease of access to other re- search collections in the geographical area for inspection of large quantities of material. When the acquisition rate is restricted to materials needed in situ, the possibilities of using interlibrary loan have been taken into account in determining that rate. Interlibrary loan availability, for practical purposes, then becomes a minor factor in determining permanent retention of those materials. Extensive use of interlibrary loan by a scholar is limited to materials that can be identified as relevant without inspec- tion. As noted above, most writers have ex- pressed requirements for acquisition rates in terms of a percentage of the existing collection. Various studies, in- cluding those of Rider10 and the Pur- due study, 11 have concluded that the sixteen-year, or faster, doubling rate, established through examination of li- brary growth statistics over considerable p eriods of time, would and should con- tinue indefinitely. The result has been a belief that libraries should have ever- increasing acquisition rates, the theory of "exponential growth." Various per- centages have been adopted in long- range growth plans of university li- braries and multicampus systems-the p ercentages usually varying from 4 to 6. It has become obvious that this ap- proach is no longer viable. Factors, such as increased costs, decreased funding , b etter bibliographic access and network- ing, and the fact that the quantity of scholarly publishing is no longer accel- erating, have resulted in few major li- braries showing exponential growth pat- terns in the last few years. CALCULATION OF GROWTH RATES If major emphasis is to be given to maintaining consistent acquisition rates, how can they be calculated and justi- fi ed? Comparison with similar institu- tions has always been used and will con- tinue to be of some value as a measure of adequacy. But there is a multiple mirror effect that could result in inade- q uacies everywhere, if comparisons alone are used as justification. Almost all other formulas that have b een used or proposed relate acquisition rates to numbers of students, a totally unsatisfactory solution. W. M. Randall, in 1931, pointed out: "It appears obvi- ous, for example, that the number of students enrolled in courses in econom- ics in College A can have no possible ef- fect upon the number of authoritative and necessary books published during a year in the subject .. . . it is easy to see that the size of the student enrollment is not a factor in the budget of these Acquisition Rates I 265 purchases."12 Massman and Patterson al- so make this point in their article on acquisition budgets: "Naturally the col- lege with a larger number of students needs more duplicate copies, and it may also have a greater variety of programs. However, course for course and major for major there is no difference in the number of separate titles needed by any institution .... ten students majoring in a subject area will need access to the same quantity of sources as one hun- dred students."13 This is even more true for advanced graduate programs and scholarly research generally. It would seem that a model or series of models should be developed that would permit a university or group of universities to establish minimum an- nual acquisition rates. The model should provide for continuing acquisi- tions that would change in quantity over a period of years only as variables in the model changed. The purpose of this article is to suggest and describe such a model. The calculation and the numbers used here would need to be tested in a variety of institutions before being generally adopted. Individual in- stitutions would probably need to pro- vide variations in some of the factors to meet specialized situations. However, the philosophy of the model is that a basic acquisition rate can be created that is relatively easy to understand and ap- ply, and that every university will have additional factors that could be taken into account in developing the acquisi- tion rate, but that the base rate should be high enough to allow for these addi- tional factors, which will tend to aver- age out and, therefore, need not compli- cate the model. The model, as described, has been applied to a number of univer- sities and appears to yield results within the range of adequacy. DEVELOPMENT OF THE MODEL Before the model can be developed, a fundamental division of the acquisi- 266 I College & Research Libraries • July 1975 tion progran1 must be made. Acquisition of currently published books and jour- nals ·must be separated from acquisition of retrospective materials. It is the ac- quisition of currently published books and journals that must be maintained at a relatively constant rate. While no library can reduce retrospective pur- chases to zero, there is flexibility in how rapidly retrospective collections are brought to any determined level of ade- quacy. Acquisition of current material is not as directly related to a library's existing collection as retrospective pur- chases are. A number of writers have recognized the importance of making this distinction. Meyer and Demos state that, among other things, an acquisi- tion policy for current materials must recognize that "the pattern of spending between current and retrospective ma- terial will tend to shift as less and less of the budget need be allocated to fill- ing lacunae .... funds for current im- prints will remain major (and probably increase, as the price of books and the quantity of books continue to rise) \vhile funds set aside for retrospective buying will be less."14 The model, as developed here, relates to that portion of acquisitions devoted to currently published materials. It seems clear that any attempt to break down acquisitions of currently pub- lished materials to specific numbers of volumes for every undergraduate major, every master's program, and every doc- toral program leads to a morass of Ineaningless numbers, due primarily to the. great amount of overlap between fields, the extensive use of materials in several fields by students and scholars in other fields, and the variations in breadth and depth which programs with the same. or comparable titles will have in various universities. Thus, while the model results in a specific number of justified volumes, ultimate use or distri- bution of acquisition funds is not de- termined by the model, because ( 1) all of the factors are interrelated and do not in themselves necessarily take into account all of the needs of a specific subject or field, and ( 2) the factors are set at levels necessary to include general and related fields not specifically covered in the model. The model is intended for general universities with extensive advanced graduate (Ph.D.) programs in a defina- ble and broad array of fields. The model then provides for modulation of a basic acquisition rate for the "univer- sity" as defined-the modulation based on additional professional programs, additions or subtractions of advanced graduate programs to or from those in- cluded in the definition, increasing num- bers of undergraduate students, and, where applicable, extensive, sponsored research programs and ease of access to other major libraries. Basic Principles The basic principles of the model are as follows: 1. Rates for acquisition of current books and journals are related to the totality of academic needs, which remain relatively constant and which require continuing funding largely unrelated to lev- els of existing retrospective collec- tions. 2. The level of current book and journal acquisition rates provided through use of the model supplies those materials without which quality master's and advanced graduate programs could not ex- ist; but with dependence on other institutions, particularly those close enough to be easily used for material not in demand frequent- ly enough to warrant duplication. Special ·requirements at every uni- versity result in unique materials not requiring duplication at other !learby institutions, materials that can and should be made generally . available. 3. ·A university, for purposes of the model, is defined as offering ad- vanced graduate (Ph.D.) work in American/ English and at least two other major . European litera- tures; in at least three social sci- ences; in the major sciences (chemistry, physics, biology, and mathematics); in history, with concentrations in a variety of geographic areas and time peri- ods; in psychology; and in philos- ophy. A base acquisition rate for current books is then established for a university as defined ( M.l). 4. The established rate of acquisi- tions of current materials ( M.l) is expected to provide for mini- mal needs of undergraduate stu- dents in the fields beyond those included in the definition. 5. The rate for current materials is set at a level to provide the mate- rials needed regularly by master's- level graduate · students, whether or not there are advanced gradu- ate programs in all fields in which there are master's programs. Every university will have some master's programs with special library re- quirements. The basic current ac- quisition r.ate ( M.l) is set at a level to provide for these pro- grams without further additions. 6. The basic acquisition rate ( M.l) for current materials for the uni- versity as defined is a composite, which, because of subject interre- lationships and the breadth of coverage inherent in the de£ni- tion of a university campus, can- not be determined by the numeri- cal add~tion of specific subject book requirements in individual academic fields aJ}d at various levels, but must be . based on cumulative experience. 7. If advanced _g-raduate , programs Acquisition Rates I 261 do not exist in the university as defined in 3, some deductions in current acquisitions can be made, but the deductions should not eliminate materials needed in these fields for undergraduate and master's programs ( M.2). 8. Current acquisition allocations .for additional advanced graduate programs, professional schools, and for other specialized doctoral programs with a high degree of independence in their lite,.-atures are added to the basic rate at specified rates ( M.3 and M.4). 9. Additions are made for addition- al' advanced graduate programs in foreign literatures, social sci- ences, and certain physical sci- ences (earth' sciences or geology and astronomy) ( M.3). Fields, such as area programs or com- bined language-history programs, are considered as additions in ei- ther the language or social science area, but not in both, even when separate graduate degree pro- grams are established. Thus, if doctoral programs exist in both Asian studies and Oriental lan- guages, or in both Islamic studies and Near Eastern · languages, an addition is made for an addition- al foreign literature or for an ad- ditional social science, but not for both. Area programs, such as African or Far Eastern, are con- sidered as single ·additions in ei- ther foreign literature 'or social science. Foreign: languages and literatures, other than those of the major literature-producing countries, are grouped and con- sidered as single additions, e.g., Scandinavian language and, litera- ture or Slavic language and liter- ature. 10. The base acquisition rate, plus ad- ditions as provided in M .3 and M.4, is set at levels that will pro- 268 I College & Research Libraries • July 1975 vide for additional advanced graduate programs without fur- ther increases. Such programs are either contained in broader areas specifically provided for, e.g., medicine, agriculture, engineer- ing, biology, or have relatively small literatures of their own, with heavy dependence on the lit- erature of fields covered in the base, or in ones added under M.3 and M.4. Examples of advanced graduate programs for which no additional volumes are added include agricultural economics, anatomy, archaeology, atmospher- ic science, biochemistry, biomedi- cal engineering, biophysics, bot- any, classical archaeology, com- parative literature, computer sci- ence, demography, ecology, endo- crinology, engineering physics, entomology, genetics, geochemis- try, information science, logic, meteorology, microbiology, neuro- science, pathology, pharmacology, physiology, plant physiology, pub- lic policy, romance languages or philology, soil science, speech, sta- tistics, wildlife resources, and zoology. 11. Where large undergraduate en- rollments exist, provision is made for additional material, primarily duplicate copies ( M.5). No addi- tions are made for numbers of graduate students, in that dupli- cate copies are usually not re- quired for graduate work and be- cause the number and breadth of fields covered, rather than the number of students, are the gov- erning factors. 12. In universities with large amounts of sponsored research, and where the informational requirements of such research are supplied by the university library, additional books and journals (largely dupli- cates) are required (M.6). 13. To make an allowance for isola- tion of universities not within easy access to other universities, an addition factor based on dis- tance is included ( M.6). While interlibrary loans can be made re- gardless of distance, efficient re- search, especially in advanced graduate work, depends on exami- nation of large quantities of ma- terial. Thus, the absence in the area of other major libraries that can be easily consulted means that an isolated university library must regularly obtain more material than would otherwise be neces- sary. The value of additional li- braries that can easily be used by scholars and advanced students cannot be overemphasized. Even when neighboring universities in- clude the same subject fields in their academic programs, much additional material will be avail- able for consultation. As stated by Arms, "For a homogenous group of libraries the number of new titles obtained by adding the holdings of another library is pro- portional to the number of books added and the size of the library is immaterial."l5 Quantitative Factors The suggested quantitative factors to be used in the model are as follows: No. of Volumes M.1 Acquisition rate of currently published materials for a university as defined ............. 40,000 (For purposes of this model, currently published materials are defined as books or journals published in the year received or in the previous year. Thus, in 1975, currently published materi- als bear either 1974 or 1975 pub- lication dates.) M.2 Subtraction rate per field for fewer than two European literatures, or three social sciences, or if psychology or philoso- phy are not included M.3 Addition rate per field for additional advanced graduate programs in foreign literatures, social sciences, earth sciences ( geology), and astrono- my, subject to the limita- tions noted in 9 and 10 above ................ . M.4 Addition rates for ad- vanced graduate profes- sional schools or subjects: Agriculture ........... . Architecture .......... . Art .................. . Business Administration. City and Regional Plan- ning ................. . Drama ............... . Education ............ . Engineering 1,000 2,000 5,000 1,000 3,000 2,000 2,000 2,000 3,000 1,000 per major area 4,000 maximum Law . . . . . . . . . . . . . . . . . . 8,000 Library Science . . . . . . . . 1,000 Medicine . . . . . . . . . . . . . 8,000 Medicine-Related Professions . . . . . . . . . . . . 1,000 per major area 4,000 maximum Nlusic . . . . . . . . . . . . . . . . . 3,000 Oceanography . . . . . . . . . 3,000 Religious Studies . . . . . . 2, 000 Social Welfare . . . . . . . . 1,000 Veterinary Science . . . . . 2,000 (It is assumed that related pure science materials exist in the 40,000 base, thus reducing there- quirements in medicine, veteri- nary science, engineering, ocean- ography, and agriculture to the levels indicated; that social sci- ence materials help support busi- ness administration, city and re- Acquisition Rates I 269 gional planning, education, and law; that the base allocation helps support architecture, mu- sic, art, drama, and library sci- ence; that philosophy helps sup- port religious studies; and that requirements for all other areas are included within the totals for the campus.) M.5 Addition for undergraduate students. For each 2,000 undergrad- uate students or fraction thereof over 5,000 under- graduate students . . . . . . 1,000 M.6 Addition for sponsored research. For each $15 million in sponsored re- search contracts or grants . . . . . . . . . . . . . . . . . 1,000 M.7 Addition for lack of ac- cess to other research li- braries: Travel time, less than one hour each way, to a research library of 2 mil- lion or more volumes . 0 Travel time, oRe to two hours each way, to a ma- jor research library .... 10,000 Travel time, more than two hours each way, to a major research library . 20,000 Table 1 illustrates the use of the for- mula for the University of California, San Diego ( UCSD), and a number of hypothetical institutions. USE OF THE MoDEL As stated earlier, this type of model is not useful if one attempts to include retrospective acquisitions. Retrospective books and journals, and microforms, which are largely retrospective, must be added at rates related to the level of the existing collection, the funding avail- able, and the staff available to do retro- spective selection. The appropriate lev- el might be from 5,000 to 25,000 vol- 270 I College & Research Libraries • July 1975 TABLE 1 USE OF MODEL TO DETERMINE ACQUISITION RATES FOR CURRENTLY PUBLISHED MATERIALS UCSD Institution A Institution B Institution C M.1 Base 40,000 40,000 40,000 . 40,000 M.2 Subtractions -4,000a M.3 Additions: Foreign literatures 2,000b 14,oooc 8,000d Social sciences 2,0008 14,000f 6,000g Earth science, astronomy 4,000h 4,000h 2,000 1 2,000 1 M4. Advanced graduate and professional Agriculture 5,000 Architecture 1,000 Art 3,000 3,000 Business administration 2,000 City and regional planning 2,000 2,000 Drama 2,000 Education 3,000 3,000 Engineering 3,00()k 4,000 1 Law 8,000 Library science 1,000 1,000 Medicine . 8~0.00 8,000 8,000 Medicine-related professions 4,ooom 2,ooon Oceanography 3,000 3,000 Religious studies 2,000 Social welfare 1,000 Veterinary science . 2,000 M.5 Undergraduate suppl. 1,000 7,000 3,000 M.6 Sponsored research 4,000 4,000 .1,000 M. 7 Access factor 20,000 10,000 20,000 --~ --- TOTAL 87,000 112,000 56,000 103,000 a No Ph.D. in philosophy, psychology, and only one in foreign literatures and two in social sciences. b Ph.D.s in French, German, Spanish. --~ ·· · .. I . c Ph.D.s in French, German, Spanish, Classics, Italian , Scandinavian, Oriental, Slavic, Near Eastern. d Ph.D.s in French, German, Hispanic, Russian, Chinese, Italian. e Ph.D.s in anthropology, economics, linguistics, sociology. f Ph.D.s in anthropology, archaeology, criminology, economics, geography, Latin American studies, ·linguistics, paleontology, political science, sociology ( Ph.D.s also in Asian studies and Near Eastern studies, included under foreign literatures.) g Ph.D.s in anthropology, geography, linguistics, sociology, Near Eastern studies, African studies. · h Ph.D.s in earth sciences and astronomy. I Ph.D. in geology. J Ph.D. in astronomy. k Ph.D. in aerospace, applied physics, applied mechanics. 1 Ph.D. in chemical, civil, electrical, mechanical, industrial. m Doctoral degrees in dentistry, pharmacy, public health, nu,rsing. n Doctoral degrees in optometry, pharmacy. urnes per year and can, of course, vary from year to year. A factor of great importance in ap- plying the model and in determining retrospective acquisition rates, but one that appears to be impossible to insert, is that of the quality of the scholarly work at a university. It is evident that a;n institution that puts little emphasis on research-that has few scholars who are productive, in terms both of quality and quantity of research-has far less need for extensive library resources than another university with an equal number of advanced graduate students, but staffed with some of the top schol- ars of the nation and with great empha- sis on research of high quality. If a uni- versity has no aspiration for a research reputation, it can lower the base and add-on factors significantly; but if it has or expects to obtain and· maintain a position of scholarly leadership, . it will find that minimal acquisition rates along the lines of those suggested in the model will probably be necessary. Gifts and exchanges have not been considered in the construction of the model. Most gifts are of retrospective materials and therefore should . not af- fect the model. Exchanges, if they con- tribute materials which would otherwise be purchased within the construction of the model, should be deducted from the total when · acquisition funding is calcu- lated. · Finally, as has been made clear, the numbers used here in the model, or any others that might be substituted, are obviously based on judgment. Most li- brarians who have commented on the Acquisition Rates I 271 model believe them to be o.f the right order of magnitude. Work with the model indicates that the base rate of 40,000 volumes could not be reduced more than 5,000 or increased more than 10,000 without destroying its viability. Obviously, increases or decreases in the rate of scholarly publications might change the · base. If there is concern about using experience, judgment, and comparison in establishing the base rate and other factors, it should be pointed out that many other factors in educa- tion, such as student-faculty ratios, space allocations, and administrative support ratios also depend on experi- ence, judgment, and comparison with other institutions. REFERENCES 1. Rutherford D. Rogers and David C. Weber, University Library Administration (New York: Wilson, 1971), p .l13. 2. William Dix, "The Financing of the Re- search Library," College & Research Li- braries 35: 25z:-58 (July 197 4). 3. Robert B. Downs, "Development of Re- search Collections in University Libraries," in William H. Jesse, ed. , The Library in the University (Hamden, Conn.: Shoe String, 1967), p.60-75. 4. Verner W. Clapp and Robert T. Jordan, "Quantitative Criteria for Adequacy of Ac- ademic Library Collections," College & Re- search Libraries 26: 371- 80 ( Sept. 1965). See also "Corrigenda," College & Research Libraries 27:72 (Jan. 1966). 5. Verner W. Clapp and Robert T. Jordan, The Libraries of the State-Assisted Institu- tions of Higher Education in Ohio-Their Maintenance and Development-Guide- lines for Policy (Washington, D.C.: Coun- cil on Library Resources, 1964). 6. Fremont Rider, The Scholar and the Fu- ture of the Research Library: A Problem and Its Solution (New York: Hadham Pr., 1944), p.9. 7. Personal letter w ritten by Verner W. · Clapp, president, Council on Library Re- sources, to Robert M. Hayes and the au- thor on March 28, 1967. ·8. Personal letter written by Verner W. Clapp to the author on: February 16, 1967. 9. R. Marvin Mcinnis, "The Formula Ap- proach to Library Size: An Empirical Study of Its Efficacy in Evaluating Re- search Libraries," College & Research Li- braries 33:190--98 (May 1972). 10. Rider, The Scholar and the Future of the Research Library. 11. Oliver C. Dunn, Warren F. Seibert, and Janice A. Scheuneman, The Past and Like- ly Future of 58 Research Libraries, 1951- 1980: A Statistical Study of Growth and Change (Lafayette, Ind.: University Li- braries and Audio Visual Center, Purdue University, 1965 ) . 12. William M. Randall, ''The College-Library Book Budget," Library Quarterly 1:421- 35 ( 1931 ). . 13. Virgil Massman and Kelly Patterson, "A Minimum 'Budget for Acquisitions," College & Research Libraries 31:83-88 (March 1970). 14. Betty J. Meyer and John ·T. Demos, "Ac- quisitions Policy for University Libraries: Selection or Collection,"' Library Resources & Technical Services 14:395-99 (Summer 1970). 15. W. Y. Arms, "Duplication in Union Cata- logues," journal of Documentation 29:373- 79 (Dec. 1973).