College and Research Libraries The Serial/Monograph Ratio in Research Libraries: Budgeting in Light of Citation Studies Robin B. Devin and .Martha Kellogg Librarians are concerned with the ever-increasing portion of the budget being devoted to seri- als. No formula or objective guideline presently exists to determine the correct allocation ratio for serials vs. monographs. Over the past 40 years, over 50 studies have analyzed the use of the research literature in various subject areas. These studies use citation analysis to determine the actual percentage of serials vs. monographs used by researchers in each field. The present study uses these figures to provide a guideline for serial vs. monograph budgeting. II esearch library collections have faced a complex budget squeeze for more than ten years. Escalating prices and burgeoning numbers of journals com- bined with limited budgets have placed tremendous pressure on research libraries to maintain a "balanced" collection of se- rials and monographs to meet the educa- tional and research requirements of stu- dents and faculty. The ratio of serials to monographs has been a concern in aca- demic libraries since the 1970s as serial subscription costs have outstripped other materials costs. Librarians view with alarm the trend toward devoting a larger proportion of the materials budget to seri- als. University libraries have undertaken major serials deselection projects in reac- tion to the perceived imbalance in the se- rial/monograph budget ratio. This paper will review causes of the serial/mono- graph budget squeeze, reactions in the lit- erature, and responses of libraries to the problem. It will present a method for de- termining reasonable serial/monograph budget ratios for research libraries. A serial is now identified by most au- thorities in terms of the AACR2 definition: II A publication in any medium issued in successive parts bearing numerical or chronological designations and intended to be continued indefinitely.' ' 1 Although serials take on many different forms (peri- odicals, journals, newspapers, maga- zines, annuals, proceedings, etc.), li- braries acquire the majority of them by subscription (payment in advance of pub- lication), a characteristic which often con- stitutes a fixed cost in the library's budget. COPING WITH THE SERIALS EXPLOSION Library materials budgets have been un- able to keep up with the escalating num- bers and costs of serials. White reported Robin B. Devin is Head, Acquisitions Department and Martha Kellogg is Assistant Acquisitions Librarian at the University of Rhode Island, Kingston, Rhode Island 02881-0803. 46 on studies of the Indiana University Re- search Center for Library and Information Science on the interaction of libraries and scholarly publishers for ten years starting in 1969. Strategies devised by libraries to cope with the budget crisis included the following: 1. Libraries tended to freeze periodical budgets at their current level and cut down drastically on new subscriptions. 2. Libraries cancelled duplicate sub- scriptions. 3. Libraries cancelled foreign titles ( es- pecially foreign language titles). 4. Libraries did not make cancellation decisions based on availability from other sources. 5. Libraries did not cancel because of • 2 pnce. When these strategies were insufficient to maintain serials collections within the budget, libraries coped, White concludes, by transferring funds from the mono- graph budget to the serial budget. He notes: in 1969, academic libraries spent $2.00 on books for every $1.00 on serials. In 1973, they spent $1.16 on books for every $1.00 on serials. In 1976, they spent $1.23 on serials for every $1.00 on books. 3 The serial/monograph budget ratio had shifted in favor of serials for the first time. Other authors have pointed with grow- ing anxiety to the same phenomenon. Taylor reiterates: "Since it is difficult to forecast the exact cost of serial renewals, some [libraries] have merely treated serial renewals as a fixed cost and used one fund for serials and book purchases .... After the serials have been paid for, what is left is available for books. The book fund has had to bail out the serial fund when the se- rial fund is inadequate for the purpose. " 4 Statistics show that, indeed, a greater percentage of academic library budgets is now devoted to serials than to mono- graphs. But available data from the Asso- ciation of College and Research Libraries (ACRL) and the Association of Research Libraries (ARL) indicate that this change is not as dramatic as one might assume, con- sidering the tremendous growth in the numbers and costs of serial publications. ''Serials Expenditures as Percent of Mate- rials Expenditures,'' published in the ARL Serial/Monograph Ratio 47 and ACRL table, Analysis of Selected Varia- bles of University Libraries, reveals that the median has actually changed very little in the last ten years. As can be seen in table 1, the ratio of serial to monographic expendi- tures has shifted slightly in favor of seri- als, but the median in academic libraries has fluctuated between 45 to 57%. Al- though some academic libraries spend two-thirds or more of their materials budget on serials, others spend one-third or less, and the median has stayed below 60%. Additional funds to cope with serial costs have not come from other areas of the university budget. "Materials Ex- penditures as Percent of Total Operating Expenditures,'' from the same ACRL and ARL tables, shows that the materials budget as a percentage of library expendi- tures has remained virtually constant over the same period. Table 2 presents avail- able statistics on the relationship between the materials budget and the total operat- ing budget of academic libraries reporting to ACRL and ARL. With academic library materials budgets remaining relatively constant (some grow- ing at the rate of inflation, others remain- ing constant, some even decreasing), the TABLE 1 SERIALS EXPENDITURES AS PERCENT OF MATERIALS EXPENDITURES Date ACRL * 1978-79 1983-84 1985-86 ARU 1980-81 1982-83 1983-84 1984-85 1986-87 High 82 92 97 83 71 77 77 75 Median 49 45 51 57 54 55 53 56 Low 15 25 27 36 30 33 34 32 * ACRL University Library Statistics 1978-1979 (Chicago : Associ- ation of College and Research Libraries, 1980); ACRL Univer- sity Library Statistics, 1983-84 (Chicago: Association of Col- lege and Research Libraries, 1985); ACRL University Library Statistics, 1985-86 & 1986 "100 Libraries" Statistical Survey (Chicago: Association of College and Research Libraries, 1987). t ARLStatistics, 1980-81 (Washington, D.C.: Association of Re- search Libraries, 1981); ARL Statistics, 1982-83 (Washington, D.C. : Association of Research Libraries, 1984); ARL Statistics, 1983-84 (Washington, D.C. : Association of Research Li- braries, 1985); ARLStatistics, 1984-85(Washington, D.C.: As- sociation of Research Libraries, 1986); ARL Statistics, 1986-87 (Washington, D. C.: Association of Research Libraries, 1988). 48 College & Research Libraries TABLE2 MATERIALS EXPENDITURES AS PERCENT OF TOTAL OPERATING EXPENDITURES Date ACRL* 1978-79 1983-84 1985-86 ARU 1980-81 1982-83 1983-84 1984-85 1986-87 High 55 54 47 44 47 50 48 49 Median 36 37 35 32 32 32 32 34 Low 19 19 17 20 20 17 22 20 * ACRL University Library Statistics 1978- 1979 (Chicago : Associ- ation of College and Research Libraries, 1980); ACRL Univer- sity Library Statistics, 1983-84 (Chicago : Association of Col- lege and Research Libraries, 1985); ACRL University Library Statistics, 1985-86 & 1986 "100 Libraries" Statistical Survey (Chicago: Association of College and Research Libraries, 1987). t ARL Statistics, 1980-81 (Washington, D. C.: Association of Re- search Libraries, 1981); ARL Statistics, 1982-83 (Washington, D.C.: Association of Research Libraries, 1984); ARL Statistics, 1983-84 (Washington, D.C.: Association of Research Li- braries, 1985); ARL Statistics, 1984-85 (Washington, D.C.: As- sociation of Research Libraries, 1986); ARL Statistics, 1986-87 (Washington, D. C. : Association of Research Libraries, 1988). serials budget has been caught in a squeeze. The number, cost, and impor- tance of serial publications have grown dramatically during the latter half of the twentieth century, but library resources have not kept pace. Competition for scarce resources has affected the budget allocation between serials and mono- graphs in academic libraries. ALLOCATION OF THE MATERIALS BUDGET Allocation of the materials budget among competing interests is at the heart of academic librarianship. It concerns ac- quisitions and collection development li- brarians, subject specialists, administra- tors, and the teaching faculty. The concept of allocating book budget funds by for- mula among the various university de- partments has long been advocated in the literature. 5 Allocation formulas may be quite complex, assigning numerical weights to many variables considered im- portant in distributing funds among sub- jects or departments. As summarized by Kohut and Walker, four major factors have generally been considered in alloca- tion formulas: (1) subjective judgments based on collection evaluations and his- torical inequities; (2) size of academic de- partments (number of students, credit January 1990 "Most allocation formulas deal with the book budget, leaving serials as a fixed cost not allocated by subject.'' hours, faculty, etc.); (3) level of program (graduate/undergraduate) and library us- age; and ( 4) size of the published litera- ture. 6 Each library assigns numerical val- ues to these factors (and sometimes dozens of other variables7) based on its own programs and emphases to arrive at an allocation for the academic depart- ments at that institution. Most allocation formulas deal with the book budget, leaving serials as a fixed cost not allocated by subject. (Kohut provided a model which takes into account the dif- ferential rate of cost increase, especially among serials. 8 He assumed, however, that that allocation among departments and the serial/monograph ratio within de- partments had already been determined.) Department allocations based on most for- mulas provide an equitable balance for book acquisitions. When serial subscrip- tions are treated as a fixed cost, however, the explosive growth in number and price of serial publications has created serious problems with the allocation of the materi- als budget. It has led to the recent concern for the serial/monograph budget ratio in academic libraries. THE SERIAL/ MONOGRAPH RATIO The importance of determining a ra- tional ratio between serials and mono- graphs within a limited materials budget has assumed greater importance under the circumstances detailed above. A re- view of the literature, however, yields lit- tle guidance for determining what the bal- ance should be. Stueart and Miller maintain: The question as to what proportion of a li- brary's acquisitions funds should be allocated for serial commitments has been asked ever more frequently as costs have risen and bud- gets have lagged. There can be no definitive an- swer, even for a given type and size of library. No single recommendation can be made as to the proportions of a book budget that should be spent on the two major types of publication [se- rials, monographs] .... There is no general agreement on the desirable ratio, but collection development officers seem to feel uneasy when periodical expenditures rise far above 60%. 9 According to Magrill and Hickey: ''With- out question, one of the knottiest prob- lems in many large research libraries is the distribution of limited acquisitions money between monographs and serials. " 10 Standards and guidelines for academic libraries developed by the American Li- brary Association (ALA) recognize gener- ally that budget allocation is necessary, but remain silent on the issue of a budget ratio between serials and monographs. ALA's Standards for University Libraries states only that ''a university library's col- lections shall contain all of the varied forms of recorded information.' ' 11 ALA's ''Guidelines for the Allocation of Library Materials Budgets'' suggests allocation by form, by subject, or a combination of both, but presents no ratio between the two ma- jor forms of publication. 12 Since monographs consumed the lion's share of the materials budget (often re- ferred to as the "book budget") until the mid-1970s, the idea of a reasonable se- rial/monograph ratio is a comparatively recent concept. To a certain extent, the concept implies a fear that the monograph collection will be overwhelmed by a del- uge of serials. As Kohut noted: "Implicit in the concern about serials costs is the as- sumption that too much of the book budget is devoted to serials. Most libraries could balance the acquisition of mono- graphs with serials so as to maximize their potential to serve the university's infor- mation needs." 13 Kohut also pointed out that a reasonable ratio between mono- graphs and serials in a university library varies depending on the subject disci- pline: "Every discipline has its own opti- mal ratio between serials and mono- graphs., ,14 White, while detailing libraries' ineffec- tual attempts to deal with the budget crunch, recognized that there is a relation- ship between the use of different types of literature and academic discipline. Of course, this shift from the monographic to the serials budget impacts some disciplines more than others. The physical sciences, in Serial/Monograph Ratio 49 general, are strongly dependent on the periodi- cal literature. The humanities, on the other hand, are far more oriented to monographs. Where shifts have taken place across depart- mental disciplines, there has been a shift in em- phasis. There is nothing necessarily wrong with this, as long as it is done consciously .15 Swindler develops the theme that librar- ians, as traditional "book people," have had little understanding of library re- sources beyond the monograph, and that, until the recent financial crisis, "serials were quite neglected in comparison to books.' ' 16 Budget allocations, he suggests, should take into account both types of lit- erature according to their importance within subject disciplines. Depending on the programs, serials can be vi- tally important-often more essential than books in many subject areas and for satisfying certain needs. The tendency of libraries to budget primarily by format not only does not make sense in terms of supporting programs with the most appropriate configuration of re- sources, but a budgetary model based on for- mat impairs the institution's ability to meet needs equitably, given the traditional bias to- ward maximizing the number of books pur- chased and the relatively new trend of arbitrar- ily setting a limit to how much the library will spend on serials subscriptions. Rather, expend- itures should be considered as part of a consoli- dated budget allocation for the support of a par- ticular program. 17 Although Swindler suggests that the ra- tio between serials and monographs will vary depending on the discipline, he does not specify how the consolidated budget allocation for particular programs can be determined. DETERMINING THE RATIO It is our contention that a reasonable budget ratio between serials and mono- graphs for individual subject disciplines can be determined. A serial vs. mono- graph budget allocation can be made us- ing statistics provided in studies on the characteristics of literature usage in each subject area. Over the past 60 years, nu- merous articles have been published that characterize the use of the research litera- ture in various fields. These studies have been conducted by selecting journal arti- cles or other research done in specific sub- ject areas and analyzing the references TABLE 3 SERIALS USE BY SUBJECT LCOass Subject Gtation %Serials Use A American Studies Bolles1 42.6 BL-BX Theology Whalen2 23.3 BL-BX Theology Heussman3 24.8 0-F His to~ Baughman 4 23.3 DA Englis History Jones5 27.1 M Music Vaughan6 28.2 M Music Bakel 23.5 N Fine Arts Simonton8 28.6 p Philology Tucker9 38.4 PA Classics Tucker10 28.5 PN Speech Broadus(1953)11 45.7 p Literature Stem12 15.0 PR-PS British and American Literature Cullars(1985) 13 13.3 p English Literature Chambers14 28.1 p English Literature Heinzkill15 19.9 p ForeTs Literature Cullars(1988)16 10.9 Soci Sciences Guttsman17 36.0 Social Sciences Earle18 29.0 Social Sciences Garfield(1976)19 38.0 BF Psycholofo Xhignesse 20 35.0 GN Anthropo ogy Baughman(1977)21 42.9 HB Economics Fletche~ 47.3 HB Economics Baughman(1977)23 59.0 HD Agricultural Economics Littleton24 31.4 HD-HJ Business Administration Sarle25 42.3 HD-HJ Business Popovich26 58.6 HM-HV Sociology Broadus(1952)27 46.3 HM-HV Sociology Broadus(1967)28 38.5 HM-HV Sociology Lin29 38.8 HM-HV Sociology Baughman(1974)30 38.5 HM-HV Sociol~ Baughman(1977) 31 44.4 J Politic cience Baum 32 31.5 J Political Science Baughman(1977) 33 34.6 JF Public Administration Intrama 34 26.1 L Education Broadus(1953)35 42.6 L Education Chambers36 40.5 L Education Mochida37 41.7 z Library Science Barnard38 52.3 z Library Science Penner39 50.7 z Library Science Brace40 33.0 z Library Science LaBorie41 28.2 Science Garfield(1976)42 80.0 Science Earle43 82.0 QA Mathematics Brown44 76.8 QB Astronomy Lan0 45 85.4 QC Optics Lin 76.5 QC Physics Fussler47 89.7 QC Physics Brown48 88.8 QD Chemistry Fussler49 92.9 QD Chemistry Barke~ 86.1 QD Chemistry Brown 51 93.6 QE Geology Gross52 85.4 QE Geology Laosunthara53 83.0 QE Geology Craig54 77.4 QE Geology Brown 55 87.2 QK Botany Hintz56 86.3 QK Botany Brown 57 82.7 QL Zoology Brown 58 80.8 QL Entomology Brown 59 81.2 QP Physiolo~ Brown 60 90.8 QR Microbio ogy Kanasy61 93.1 R Medicine Sherwood62 85.2 T Technology Earle63 70.0 T Engineering Waldhart64 72.4 TK Electrical Engineering Coile65 61.9 TP Chemical Engineering Patterson66 75.8 Serial/Monograph Ratio 51 TABLE 3 Cont. REFERENCES 1. Charles A. Bolles, "Characteristics of the Literature of American Studies as Indicated by Bibliographic Citations " (Ph .D. diss ., Univ. of Minnesota, 1975). 2. Sr . Mary L. Whalen, "The Literature Used in Catholic and Protestant Doctoral Research in Theology" (Ph .D. diss ., Columbia Univ ., 1965). 3. John W. Heussman, "The Literature Cited in Theological Journals and Its Relation to Seminary Library Circulation" (Ph .D. diss. , Univ . of Illinois, 1970) . 4. James C. Baughman, "Toward a Structural Approach to Collection Development," College & Research Libraries 38:241-48 (May 1977). 5. Clyve Jones, Michael Chapman and Pamela C. Woods, "The Characteristics of the Literature Used by Historians," Journal ofLibrari- anship 4:137-56 (July 1972) . 6. David L. Vaughan, "Characteristics of the Literature Cited by Authors of Articles in The Musical Quarterly, 1955-58 and The American Musicological Society Journal, 1953-56" (Master's thesis, Univ . of North Carolina, 1959). 7. David Baker, "Characteristics of the Literature Used by English Musicologists," Journal of Librarianship 10:182-200 (July 1978) . 8. Wesley C. Simonton, "Characteristics of the Research Literature of the Fine Arts During the Period 1948-1957" (Ph.D. diss., Univ. of illinois, 1960). 9. B. R. Tucker, "Characteristics of the Literature Cited by Authors of the Transactions of th e American Philological Association, 1956 and 1957" (Master' s thesis, Univ . of North Carolina, 1959). 10. Ibid . 11 . Robert N . Broadus, The Research Literature of the Field of Speech. A.C.R .L. Monograph no .7 (Chicago : Association of College and Reference Libraries, 1953). 12. Madeleine Stern, " Characteristics of the Literature of Literary Scholarship," College & Research Libraries 44:199-209 (July 1983). 13. John Cullars, "Characteristics of the Monographic Literature of British and American Literary Studies," College & Research Libraries 46 :511-22 (Nov . 1985). 14. George R. Chambers and James S. Healey, "Journal Citations in Master's Theses: One Measurement of a Journal Collection, " Jour- nal of American Society for Information Science 24:397-401 (Sept.-Oct. 1973). 15. Richard Heinzkill, "Characteristics of References in Selected Scholarly English Literary Journals," Library Quarterly 50:352-65 (July 1980). 16. John Cullars, "Characteristics of the Monographic Scholarship of Foreign Literary Studies by Native Speakers of English, " College & Research Libraries 49:157-70 (March 1988) . 17. W. L. Guttsman, "The Literature of the Social Sciences and Provision for Research in Them," Journal of Documentation 22:186-94 (Sept. 1966). · 18. Penelope Earle and Brian Vickery, "Social Science Literature Use in the UK as Indicated by Citations, " Journal of Documentation 25:123-41 (June 1969). 19. Eugene Garfield, "Anonymity in Refereeing? Maybe-But Anonymity in Authorship? No!," Current Contents: Engineering, Technol- ogy, & Applied Science 7 no.11 :5-7 (March 15, 1976) . 20. Louis V. Xhignesse and Charles E. Osgood, "Bibliographical Citation Characteristics of the Psychological Journal Network in 1950 and in 1960," American Psychologist 22:778-91 (1967) . 21. Baughman, " Toward a Structural Approach to Collection Development ." 22 . John Fletcher, "A View of the Literature of Economics," Journal of Documentation 28 :283-95 (Dec . 1972). 23. Baughman, "Toward a Structural Approach to Collection Development ." 24. Isaac T. Littleton, "The Bibliographic Organization and Use of the Literature of Agricultural Economics," (Ph .D. diss. , Univ . of Illinois, 1968). 25 . Rodney G. Sarle, "Characteristics of the Literature Used by Authors of Journal Articles in Business Administration," (Master's thesis, Univ. of North Carolina, 1958). 26. Charles J. Popovich, "The Characteristics of a Collection for Research in Business/Management," College & Research Libraries 39:110-17 (March 1978) . 27. Robert N. Broadus, "An Analysis of Literature Cited in the American Sociological Review,'' American Sociological Review 17:355-57 (June 1952). 28 . Robert N. Broadus, "A Citation Study for Sociology," American Sociologist 2:19-20 (Feb. 1967). 29. Nan Lin and Carnot E. Nelson, "Bibliographic Reference Patterns in Core Sociological Journals, 1965-1966, " American Sociologist 4:47-50 (Feb. 1969). 30. James C. Baughman, "A Structural Analysis of the Literature of Sociology," Library Quarterly 44:293-308 (Oct. 1974) . 31. Baughman, "Toward a Structural Approach to Collection Development." 32. William C. Baum, "American Political Science Before the Mirror: What Our Journals Reveal About the Profession," Journal of Politics 38:895-917 (1976). 33. Baughman, " Toward a Structural Approach to Collection Development." 34. Navanitaya lntrama, "Some Characteristics of the Literature of Public Administration," (Ph .D. diss., Indiana Univ ., 1968) . 35. Robert N. Broadus, "The Literature of Educational Research," School and Society 77 :8-10 (Jan. 1953). 36. Chambers and Healey, "Journal Citations in Master's Theses ." 37. Paula Mochida, "Citation Survey of Education Literature," Hawaii Library Association Journal33:29-42 (1976) . 38. Walter M. Barnard, "Characteristics of the Literature Used by American Authors of Journal Articles in Library Science," (Master's thesis, Univ. of North Carolina, 1957). 39. Rudolf J. Penner, "Measuring a Library's Capability ... ,"Journal of Education for Librarianship 13:17-30 (Summer 1972). 40. William Brace, "Frequently Cited Authors and Periodicals in Library and Information Science Dissertations; 1961-1970, " Journal of Library and Information Science 2:16-34 (April1976) . 41. Tim LaBorie and Michael Halperin, "Citation Patterns in Library Science Dissertations," Journal of Education for Librarianship 16:271-83 (Spring 1976) . 42. Garfield, " Anonymity in Refereeing." 43. Earle and Vickery, "Social Science Literature Use in the UK." 44. Charles H. Brown, Scientific Serials: Characteristics and Lists of Most Cited Publications in Mathematics, Physics, Chemistry, Geology, Physi- ology, Botany, Zoology, and Entomology. ACRL Monograph no.16 (Chicago: Association of College and Reference Libraries, 1956). 45. NormaL. F. Lang, "Characteristics of the Serial Literature of Astronomy, " (Master's thesis, Southern Connecticut State College, .1966) . 52 College & Research Libraries January 1990 TABLE 3 Cont. 46. Lin and Nelson, "Bibliographic Reference Patterns." 47. Herman H. Fussier, "Characteristics of the Research Literature Used by Chemists and Physicists in the United States, Part II," Lzbrary Quarterly 19:119-43 (April1949). 48. Brown, Scientific Serials . 49. Fussier, "Characteristics of the Research Literature." 50. Dale Barker, "Characteristics of the Scientific Literature Cited by Chemists of the Soviet Union," (Ph.D. diss., Univ. of Illinois, 1966) . 51. Brown, Scientific Serials. 52. P . L. K. Gross and A. 0. Woodford, " Serial Literature Used by American Geologists," Science 73 :660-64 Oune 1931). 53. Maria Laosunthara, "Some Bibliographical Characteristics of Serial Literature in the Field of Geology," (Ph .D. diss ., Indiana Univ., 1956) . 54. ]. E. G. Craig, "Characteristics of Use of Geology Literature," College & Research Libraries 30:230-36 (May 1969). 55. Brown, Scientific Serials. . 56. Carl W. E. Hintz, "Internationalism and Scholarship: A Comparative Study of the Research Literature Used by American, British, French, and German Botanists," (Ph.D . diss., Univ . of Chic~go, 1952.) 57. Brown, Scientific Serials. 58. Ibid. 59. Ibid . 60. Ibid . 61. James E. Kanasy, "Citation Characteristics and Bibliographic Control of the Literature of Microbiology," (Ph .D . diss ., Univ. of Pitts- burgh, 1971). 62. K. K. Sherwood, "Relative Value of Medical Magazines, " Northwest Medicine 31:273-76 Oune 1932). 63. Earle and Vickery, " Social Science Literature Use in the UK." 64. Thomas]. Waldhart, "The Relationship Between the Citation of Scientific Literature and the Institutional Affiliation of Engineers " (Ph .D. diss ., Indiana Univ. 1973). ' 65 . Russell C. Coile, "Information Sources for Electrical and Electronic Engineers," I.E . E. E. Transactions on Engineering Writing and Speech 12:71-78 (October 1969). 66. Austin M. Patterson, "Literature References in Industrial and Engineering Chemistry for 1939," journal of Chemical Education 22:514-15 (Oct . 1945). cited. The references are then categorized by form and the percentage of use of the various categories is given . To date more than 50 such articles have appeared. They cover almost the entire spectrum of subjects from the fine arts to engineering. These studies provide statis- tics on the use researchers make of serial versus monograph literature in their fields. Not every study categorizes the ref- erences cited in quite the same way, but it is possible in most cases to extract the total percentage of serial usage. One of the most striking results of these various studies is the consistency of the data they provide. In certain subject areas this type of citation analysis has been con- ducted by numerous researchers over a substantial time span with similar results. In the field of sociology, for example, five studies of journal citations conducted over a 23-year-period gave serial use percent- ages ranging from 28-46% (see table 3). In fact, three of these studies by different au- thors found the same figure of 38% serial use. Even when the materials analyzed are different, the percentage of serial use is very similar within a discipline. Three studies of sources in the field of education cited in table 3 illustrate this point. In 1953 Broadus did a study of the form of litera- ture cited in the Encyclopedia of Educational Research and found that 42.6% of the cita- tions were to serials. Chambers, in 1973, studied citations in master's theses and found a serial citation percentage of 40.5%. Finally, in 1976 Mochida studied citations in journal articles and found that serial citations accounted for 41.7% of the total. Table 3, "Serial Use by Subject," pro- vides a summary of the data on serial ver- sus monograph use given in these studies. Most librarians have long been aware that researchers in the sciences rely much more on serials than do researchers in the humanities. Yet this information has not been translated into the budget allocation process. We have found that the data pro- vided in these studies can be used to de- termine how much should be allocated for serials versus monographs in each subject area. The percentage of serial usage can be directly applied to the budgeting process. Many libraries use some variation of an allocation formula to divide their mono- graph budget into subject categories. These monograph allocation figures can then be used as a basis for determining the "Using this formula, one could find, for example, that a library allocating $10,000 for the purchase of mono- graphs in the fine arts should be allo- cating approximately $4,000 for seri- als in that subject." guidelines for serials expenditures in those areas . Using the serial use percent- age given in table 3, the serial allocation is determined by multiplying the mono- graph allocation figure (M) by the serial percentage (%) given in table 3 for that subject and dividing by (100 - that same percentage) as in the following formula: (M) (%) s = (100-%) Using this formula, one could find, for ex- ample, that a library allocating ($10,000) for the purchase of monographs in the fine arts should be allocating approximately $4,000 for serials in that subject. (10,000) (28.6%) ----- = $4,005 (100-28.6%) A library may also use its serial allocation breakdown to determine a guideline for monograph expenditures. The formula would then be s M = --S. % The use of either of these formulas as- sumes that funds for each subject should be allocated between serials and mono- graphs in the same proportion as each form of literature is used by the re- searcher. Thus, if fine arts research has a serial citation rate of 28.6%, then 28.6% of the library materials budget for the fine arts should be allocated to serials. At first glance it appears that such a sim- plistic formula fails to take into account a number of factors generally believed to be important when establishing allocations. However, it must be remembered that the formula is actually only establishing a ra- tio between serial and monograph bud- Serial/Monograph Ratio 53 gets within each subject area. Most of the factors such as collection intensity, num- ber of students and faculty in the area, cir- culation statistics, and average cost per volume should have already been taken into consideration when the original monograph (or serial) allocation was made. The application of the formula just translates that dollar figure into a corres- ponding serial (or monograph) dollar fig- ure using the citation ratio. Once the percentages from table 3 have been used to establish the guidelines for serial versus monograph budgeting of each subject area, the individual figures can be totaled to determine an overall guideline for serial versus monograph spending. This method therefore links the library's budget allocation to the actual use of library material by researchers in each subject area . CONCLUSION The information explosion of the late twentieth century has placed a severe strain on the materials budgets of research libraries . University libraries that must support teaching and research in tradi- tional disciplines as well as in emerging programs have been hard pressed to meet the needs of students and faculty. Massive increases in the number and price of serial publications have exacerbated the prob- lem, and serial purchasing has come un- der closer scrutiny than ever before. Statistics reveal that in the mid-1970s re- search libraries, for the first time, spent a larger share of their materials budgets on serials than on monographs. This unprec- edented development led to the concept of a ratio between serials and monographs for budgeting purposes. But a search of the literature discloses no objective guide- line for determining the optimum se- rial/monograph ratio for any type of li- brary. Some writers suggest an overall percentage based on little more than a hunch; others suggest that the percentage of serials and monographs purchased will vary depending on the discipline. The authors believe that it is possible to determine guidelines for developing a se- rial/monograph ratio for research library budgeting based on objective criteria. We 54 College & Research Libraries propose that, for a research library, the percentage of serials and monographs purchased should correlate with the char- acteristics of literature usage in each sub- ject area. Citation studies conducted over the past 60 years covering most subject ar- eas have been analyzed to determine the serial/monograph usage for each subject. Using the data presented in table 3, are- search library with a current book alloca- tion formula can determine its own appro- January 1990 priate serial/monograph ratio for each subject using a simple formula. In conclusion, we find that there is no one optimum serial/monograph ratio for all research libraries. Rather, the ratio for each library will vary depending on its own priorities and emphases . For each subject, however, the serial/monograph ratio should be based on the use of the lit- erature by researchers in that subject area as determined by citation studies. REFERENCES AND NOTES 1. See, for example, Andrew D. Osborn, Serial Publications: Their Place and Treatment in Libraries, 3rd ed. (Chicago: American Library Assn., 1980), p.10; Marcia Tuttle, Introduction to Serials Manage- ment (Greenwich, Conn.: }AI Press, 1983), p.6. 2. Herbert S. White, ''Strategies and Alternatives in Dealing With the Serials Management Budget,'' in Serials Collection Development: Choices and Strategies (Ann Arbor, Mich.: The Pierian Press, 1981), p.31-32. 3. Ibid., p.32. 4. David C. Taylor, Managing the Serials Explosion: The Issues for Publishers and Libraries (White Plains, N.Y.: Knowledge Industry Publications, 1982), p.19-21. 5. See, for example, William E. McGrath, "A Pragmatic Book Allocation Formula for Academic and Public Libraries with a Test for Its Effectiveness," Library Resources and Technical Services 19:356-69 (Fall1975); Bette Dillehay, "Book Budget Allocation: Subjective or Objective Approach," Special Libraries 62:509-14 (Dec. 1971); Joseph}. Kohut, II Allocating the Book Budget: A Model," College & Research Libraries 35:192-99 (May 1974); Steven D. Gold, II Allocating the Book Budget: An Eco- nomic Model," College & Research Libraries 36:397-410 (Sept. 1975). 6. Joseph}. Kohut and John F. Walker, "Allocating the Book Budget: Equity and Economic Effi- ciency," College & Research Libraries 36:403-10 (Sept. 1975). 7. See, for example, "Factors in Designing an Allocation System" [at Cornell University], in G. Edward Evans, Developing Library and Information Center Collections, 2nd ed. (Littleton, Colo.: Li- braries Unlimited, 1987}, p.267-69. 8. Kohut, "Allocating the Book Budget: A Model," p.192-99. 9. Robert D . Stueart and George B. Miller, eds . Collection Development in Libraries: A Treatise (Greenwich, Conn.: }AI Press, 1980), v.2, p.508; v.1, p.59. 10. Rose Mary Magrill and Doralyn }. Hickey, Acquisitions Management and Collection Development in Libraries (Chicago: American Library Assn., 1984), p .160. 11. "Standards for University Libraries," College & Research Libraries News 40:101-10 (Apr. 1979). 12. David L. Perkins, ed., Guidelines for Collection Development (Chicago: American Library Assn., 1979), p.35-36 . 13. Kohut," Allocating the Book Budget: A Model," p.192. 14. Ibid., p.193. 15. White, "Strategies and Alternatives," p.33-34. 16. Luke Swindler, "Developing the Serial Collection," in Tuttle, Introduction to Serials Management, p.20-23. 17. Ibid., p .23.