College and Research Libraries VERNER W. CLAPP and ROBERT T. JORDAN Quantitative Criteria for Adequacy of Academic Library Collections The authors challenge accepted doctrine which asserts that the ade- quacy of an academic library cannot be measured by .the number , of books which it contains. Out of their feeling that the Standards for College Libraries and the Standards for Junior College Libraries are inadequate_ for estimating the _sizes (in volumes) required for minimum . aaequacy by libraries of institutions of higher education of widely , differing characteristics, they developed new formulas for this purpose. Tfiese formulas attempt to identify the principal factors affecting academic needs for books and to ascribe suitable weights to each factor. The authors then illustrate the application of the formulas to specific institutions, and conclude that · while the results are useful, further research is needed. They end by suggesting specific topics for such research. c AN THE ADEQUACY of the collection of an academic library be measured by the number of books which it contains? Re- ~~ctable authorities say "No!" "The adequacy of the college library's collections cannot be measured in quan- titative terms," asserts a well-known text- book in the field of college library ad- ministration. "To judge a collection su- --perior or inferior on the basis of the vol- ume holdings," it maintains, "is as ab- surd as rating a college on the basis of its enrollment."1 · Regional accrediting agencies agree. "The actual number of books whic1i: a library contains is not a stable measure of the adequacy of the library."2 "More important than the 'total number of books in the stacks is the extent to which the selection of volumes accurately re- fleets the needs of the institution as de- fined by its educational task."3 "It will be noted that no mention is made here of required minima for ... library hold- ings .... The adequacy of each institu- • tion's resources must be judged in terms of its program."4 "Every [academic] li- brary must . . . be evaluated in its own setting rather than by comparison with general patterns or norms, because each library must support a particular edu- cational program."5 And similarly the Northwest Association, 1957, and the Western Association, 1963, while con- cerned for the "adequacy" of the aca- demic library, provide no yardstick for 2 North Central Association of Colleges and Second- ary Schools Committee on Standards, College and Uni- versity Accreditation Standards-19 57 (Chicago : ACRL, 1958), p. 11. 3 North Central Association of Colleges and Secon- ary Schools. Commission on Colleges and Univer- sities, Guide for the Evaluation of Institutions of 1 G. R. Lyle, The Administration of the College Higher Education, 1961, p. 16. Library. Third ed. (New York: H. W. Wilson Co., 4 Middle States A.ssociation of Colleges and Second- 1961), p. 399. ary Schools. Commission on Institutions of Higher Education, Characteristics of Excellence in Higher Education and Standards for Middle States Accred- Mr. Clapp is President and Mr. Jordan is itation, 1957, p. 3. on the staff of the Council on Library Re- ~;:o;:~h~· ~~~~:ds~~~ch:::c~t~!n~~b~~~:;v:~d sources, Inc., Washington, D.C. R esearch L i braries, XIX (July 1958), 305-20. /371 372 I College & Research Libraries • September 1965 the measurement of that quality.6 • 7 The only regional association which makes an obeisance in the direction of a quan- titative measure (but in a manner which approximates mockery) is the Southern Association: "The book and periodical collection should, by quality, size, and nature, support and stimulate the entire educational program . . . the following should be used as a reference: Library Statistics of Collegf!s and Universities. Annual Analytic Report .... In using this reference, institutional authorities should consider it a serious danger signal if the library regularly falls in the lowest quar- ter of any of the categories analyzed."8 _ When, as in these cases, standardizing authorities omit or refuse to set stan- dards in quantitative terms, the budget- ing and appropriating authorities, who cannot avoid quantitative bases for their decisions, are compelled to adopt mea- sures which, though perhaps having the virtue of simplicity, may be essentially irrelevant. 9 It is not surprising, in consequence, that the Standards for College Libraries adopted in 1959 by the Association of College and Research Libraries of the American Library Association, while properly placing primary emphasis upon quality and the means for achieving it, should also include sufficient numerical criteria to meet to a degree the need for quantitative standards. Specifically, these Standards provide that fifty thousand "carefully chosen" volumes may serve as the minimum for the library of a college of up to six .hun- 8 Northwest Association of Secondary and Higher Schools, Guide for Self-Evaluation · and Accreditation of Higher Schools, 1957, p. 9. 1 Western Association of Schools and Colleges. Ac- crediting Commission for Senior Colleges and Uni- versities, Statement of 'Standards, 1963, p. 2. 8 Southern Association of Colleges and Schools. Col- lege Delegate Assembly, Standards, 1962, p. 31. 9 For example, in California a formula for the an- nual book fund of the state colleges provided four books per student for the first one thousand students, two for the next four thousand, etc. A recommenda- tion to change this formula proposed the provision of forty books per student by a certain date. But neither formula is directly related to the quality of the library. Program for the Development of Cali- fornia State CoUege Libraries (n.p., August 1962). p. 2-3. dred students (full-time equivalent); that "steady growth" is essential but may slacken when the collection reaches ap- proximately three hundred thousand vol- umes; and that for each two hundred stu- dents above the initial six hundred there should be an additional ten thousand volumes. It is e~phasized that these are minimal figures. 1o The Standards for Junior College Li- braries, likewise promulgated by the Association of College and Research Li- braries, are similarly insistent upon qual- ity, but similarly offer some quantitative assistance. They require that an institu- tion of up to one thousand students (full- time equivalent) shQuld have a mini- mum of twenty thousand volumes ex- clusive of duplicates and textbooks and suggest that this figure should be in- creased by five thousan? for each ad- ditional five hundred students beyond one thousand. Again, it is emphasized that these are minimal figures. 11 In neither case, however, are the sug- gested quantitative criteria convincing in the sense that they rest on demonstra- tions of actual numbers of books re- quired for specific educational purposes. Instead, the §uggested figures admittedly r~eGt th_e accidentals of college library statistics (without indication of how t IS reB.ection is effected) Or agreement among librarians consulted. The require- ments for additional books are based in one case upon an apparent "correlation between the growth of the student body and the growth of the collection," and in the other simply upon "consultation with many junior college librarians." Finally, the Standards for College Li- braries are by definition inapplicable to institutions stressing advanced research or granting degrees beyond the Master's, while the Standards for Junior College 10 [American Library Association. Association of College and Research Libraries], "Standards for Col- lege Libraries.'' CoUege and Research Libraries, XX (July 1959), 274-80. u [American Library Association. Association of College and Research Libraries], "Standards for Jun- ior College Libraries.'' CoUege and Research Libraries, XXI (May 1960), 200-206. Quantitative Criteria for Adequacy I 373 Libraries, although recognizing that in- stitutions with a multiplicity of programs may need minimal collections of two or three times the basic figure of twenty thousand volumes, do not state at what point this requirement takes effect. The present authors recently needed formulas for producing estimates of the size required for minimum adequacy by the library collections of a number of academic institutions of widely differing characteristics. It was important that these estimates should carry conviction to the planning, budgeting, and ap- propriating bodies concerned. Available standards were found unsuitable for pro- ducing the desi:t;ed result. Accordingly, an attempt was made to develop formu- las in which separate account would be taken of the principal factors that affect the requirements for books in connec~ tion with academic programs, and in which each factor woulc.t be weighted in a manner capable of being relat_~_9. to and justified by practice. The results of this attempt, though admittedly but a beginning and needing much improvement, were found useful for the purpose for which they were de- signed,12 and are consequently presented here as of possible wider interest. They invite exploration of the conditions which affect academic needs for books, of the relative weights which should be attached to the various controlling fac- tors, and of the basic hypothesis itself- namely, that it is possible to provide a meaningful quantitative measure of ade- quacy in library collections. FoRMULAS FOR EsTIM.ATING SizE OF AcADEMIC LmRA~Y CoLLECTIONs REQUIRED FOR MINIMUM ADEQUACY The minimum size required for the adequacy of "an academic li"brary differs from institution to institution depending 12 V. W. Clapp and R. T. Jordan, Th~ Li~raries of the 'State-Assisted Institutions of Higher Education in Ohio--Their Maintenance and Development- Guidelines for Policy. Prepared for Academy for Ed- ucatiqnal L?evelopment, Inc. (Washington, D. C., 1964). . . . . upon the combined effect of the vari- ables constituting the controlling factors in each case. Among the most important of these are: • The student body-size, composi- tion (graduate or undergraduate, full-time or part-time, resident or nonresident, etc.), scholastic apti- tude, socio-economic and intellec- tual background. • The faculty-size, involvement in research, "library-mindedness," etc. • The curriculum-number of depart- ments of instruction, number of courses,. proportion of laboratory to literature courses, number of under- graduate "majors," number- of fields of masters' and doctors' degrees, number of professional schools, etc. • Methods of instruction-extent and us~ of textbooks, reading a~signed and independent study, honors work, etc. • Availability of suitabie-- places for study on the campus. • Geography of the campus-prox- imity to metropolitan. areas, to · other large libr·aries, etc. • The .intellectual climate-induce- ments and distractions to study, etc. It is obvious that these factors cjiffer widely in their susceptibility to measure- ment. Only those that can be most easily and meaningfully measured were given places in the following tables which con- stitute the formulas. NoTES ON TABLE 1 The formula presumes that even lim- inal . or minimum adequacy can be achieved with its assistance only if all material is carefully chosen with a view to the purpose to be served, and ·the weeding program is as active and real- istic in relation to needs as is the pro- gram of acquisition. Averages. Because of wide disparities in the extent of the literatures of various subjects, .the figures suggested by the table must be considered as averages of the. liter?tures of subjects of academic 374 I College & Research Libraries • September 1965 TABLE 1 FoRMULA FOR EsTIMATING THE SIZE FOR LIMINAL ADEQUACY OF THE CoLLECTIONS OF SENIOR CoLLEGE AND UNIVERSITY LmRARIEs BOOKS PERIODICALS Docu- MENTS TOTAL (1) To a basic collection, viz.: 1. Undergraduate library 0 0 Add for each of the following as in- dicated: 2. Faculty member (full time 0 equivalent) . 0 0 0 0 3. Student (graduate or under- graduate in full time equiv- alents) . 0 0 0 0 0 4. Undergraduate in honors or independent study - pro- grams . 0 0 0 0 0 5. Field of undergraduate con- centration-"major" subject field 0 • • • • • • 6. Field of graduate concentra- tion-Master's work or equivalent . . 0 • • 7. Field of graduate concentra- tion-Doctoral work or equivalent Titles (2) 35,000 50 10 200 2,000 15,000 interest. It is not too difficult to estimate the size of a collection for work at a given level in a single subject; !_tis when the library is required to serve the inter- est of many users at many levels in many subjects, as in an institution of higher education, that estimates of size become difficult. -Interdependence of factors. No factor represented in the formula will be oper- ative in isolation; each is dependent on others. For example, it is not suggested that 240 monograph volumes are suf- fi~h~nt for an undergraduate field of con- centration (line 5). Obviously, there will be contributions to each field of concen- tration resulting from each of the other variables (lines 1 through 4) . Microcopy. The table presumes ·that niost of the materials estimated in lines 1-1_will be in fu - e_format. Even here, liowever, some of the less-frequent- ly:-used material (such as back files of newspapers) may be in microcopy. With respect to much of the little-used re- search material to be added in accord- Volumes (3) 42,000 60 10 12 240 2,400 18,000 Titles (4) 250 1 3 10 100 Volumes (5) 3,750 15 1 45 150 1,500 Volumes (6) 5,000 25 1 50 500 5,000 Volumes (7) 50,750J 100 12 12 335 3,050 24,500 ance with the estimates contained in lines 5-7, .. adequacy" can be achieved with almost as much efficiency through the use of microcopy as with full-scale material. The table assumes that fully cataloged material in microform 'will be measured in volumes as though it were in original form. Title-volume ratios. The title-volume ratio employed for books .(columns 2 and 3) is 1:1.2 which falls between that ( 1:1.37) found to obtain in the National Union Catalog13 and that ( 1:1.15) which is found in the Lamont library catalog.14 The ratio us.ed for periodicals (columns 4 and 5) has been set at 1:15 ( cf. the n<:>te on line 1, column 4). For documents · (column 6) a title-volume ratio does not seem to be meaningful. In consequence, the total sizes of collections obtained by using the table are expressed only in volumes. 13 E. E. Williams, "Magnitude of the Paper-De- terioration Problem as Measured by a National Union Catalog Sample," CoUege and Research Libraries, XXIII (November 1962), 499, 543. u Catalogue of the Lamont Library, Harvard Col- lege (Harvard University Press, 1953). Quantitative Criteria for Adequacy I 375 Line 1, Column 2. The figure of 50,750( more, the figure of · two hundred .and volumes suggested as capable of provid- 1 fifty is 50 per cent of th~ number of titles ing threshold adequacy for an under-' j covered by the foll~wmg standard pe- graduate collection derives authoritY riodicaf indexes pubhs~ed by the. H. W. from experience in the actual construc-' 1 Wilson Compa~y, w~thout whiCh no tion of lists for this purpose. The mosf ( general) Amencan lib~ary can expect important of these lists have been: to rende_r adequate service: TITLE LIST DATE LISTED Shaw1 5 1931 14,000 Lamont14 1953 39,000 Michigan16 1964 56,550 California17 1965 55,000 The Shaw list was a pioneering e!fort which set the pattern and the stan~ard of excellence. The Lamont list wa~ the first to be related to an actual under- graduate library, but it had many faults. The Michigan list learned from these. The California list (under construction at the library of the University of. Cali- fornia at San Diego) has not only bene- fited from previous experience but has been executed under auspicious circum- stances. The Library Council of the Uni- versity of California recommended that the three new campuses currently being planned each have seventy-five thousand- volume libraries at opening day, since the experience of the growing campuses, Irvine in particular, suggests that it is difficult to give ·adequate service with a smaller collection. The California list, in consequence, provides for about sixty/'- thousand volumes of monographs and fifteen thousand volumes of serials. Line 1, Column 4. The figure of 250 periodical titles is supported by the Michigan list which includes 245 such titles and the California list which pro- vides for fifteen thousand serial volumes representing nine hundred titles, of which the three hundred most useful are in runs of twenty or more years. Further-. 15 C. B. Shaw, A Liat of Books for College Libraries (American Library Association, 1931) . . 16 University of Michigan. Undergraduate Library, Shelf Liat. Rev. ed. (Ann Arbor, Michigan: Univer- sity Microfilms, Inc., 1964) . 11 [University of California at San Diego. Library, List of books selected for the libraries of three new campuses of the University of California.] In prepa- ration for the press. Readers' Guide to Periodical Litera- ture (selected general and nontech- nical periodicals) . . 130 International Index (social sciences and humanities) . 170 Applied Science and Technology In- dex . 200 Total 500 Line 1, Column 6. The figure of five thousand documents would admit the most important publications of the U.S. Congress, the Bureau of the Census and other federal executive agencies, the United Nations and its specialized agen- cies states of the United States, etc. Lfne 2. If the library which provides merely threshold adequacy for under- graduates is to permit the me~bers ~f the teaching staff to keep up In therr subjects even liminally, the collection must be enriched for their benefit. An enrichment . amounting to fifty titles (e.g., three per year for sixteen years), one periodical subscription and twenty- five documents per faculty member would seem to be a minimum.18 resented b · e 1 takes no account what Line 3. The under raduate libr p-J- soever o e size of e student body. As this increases, the number of copies (not titles) will have to be increased. At the su gested rate of twelve volumes per student, every book in the undergrad- uataJibran could be duplicated by the time that the student body had risen to 4,230. In other words, there could then be, if desired, two identical undergrad- 18 The observed tendency for stable and continuing academic libraries to double in size every sixteen years that is associated with the name of Frem~nt Rider suggests that sixteen years represents a period at which the collections of such libr.aries r~uire ~ substantial degree of renewal. Accordmgly, this peri- od is here adopted for the cycle of renewal for the additional materials purchased for faculty, etc. r 3761 College & Research Libraries • September 1965 uate coliections, each serving 2,115 stu- dents: It is more likely, of course, that all 4,230 would use the same library but that the books more in demand would be supplied in multiple copies. Line 4. The typical student in an hon- ors or independent study program may read or use hundreds of books each year. However, since the criterion sought here is merely threshold adequacy, a very low figure is used. . Line 5. The undergraduate collection (line 1) will rarely have as many as several hundred titles in each field in which an undergraduate "major" is of- fered. By contrast, "basic lists" for such subjects typically .include two thousand and more titles (see note on line 6, be- low). Accordingly, the reinforcement suggested here, amounting to only 17 per cent of this quantity, is very modest. Line 6. At the point at which graduate work is offered leading to the master's degree or its equivalent, the . collection must assume some of the characteristics of • a research collection, albeit at the lowest level. The quantity of material for addition here is suggested by the numerous "basic lists" which typipally include two thousand and rnore titles, - e.g.: Anthropology19 Area studies (Asia, Africa, Eastern Europe, Latin America) 2 0 Art reference books21 China modem-economic and social develbpment22 • • Communism-books in English .only23 . 2,000 7,000 2,850 2,000 2,500 19 D.~ G. Mandelbaum and others, eds., Resources for the Teaching of Anthro12ology; Including a Basic List of Books and Periodicals for CoUege Libraries Compiled by Rexford S. Beckham with the Assistance of Marie P. Beckham (University of . California, 1963). 2,000 titles. 20 American Universities Field Staff, A Select Bib- liography: Asia, Africa, Eastern Europe, Latin Amer- ioa. (AUFS, 1960) ; Supplements, 1961, 1963. 6,000 titles in basic list, 500 in each of the supplements. :n M. W. Chamberlain, Guide to. Art Reference Books (Chicago: American Library Association, 1959). 2,500 titles, 250 journals, 100 series. . . 22 T.-L. Yuan: Economic and Social :Development of Modern. China: a bibliographical guide, (New Haven: Hu:rv.an Relations Area Files, 1956). Over 2,000 titles. . ~ W. Kolarz, Books on Communism;, a . .Bibl~g:aph'J!, 2d ed. London: Allen & Unwin, 1964.,. Approximate- ly 2,500 titles, restricted to English. · · · ' · Electronics24 · Physics25 . . . . . · · United States of America-life and thought26 . 2,000 1,883 6,500 Line 7. These 24,500 volumes repre- sent but a fraction of the literature of any but the most recently-developed subject, and can ordinarily be expected to present a subject only in its most re- cent aspects, neglecting historical de- velopment. Yet as recently as 1955 one of the most literature-based of the learned professions adopted twenty thousand volumes as a passing grade for its training centers in the United States,27 and even in 1964 sixteen of these centers still had fewer than thirty thousand vol- umes. It is also true that the literatures of several disciplines support each other, as chemistry, biochemistry, physiology, anatomy, neurology, psychology, and other related sciences contribute to make a medical library. NOTES ON TABLE 2 As with Table 1 it is presumed that all material will be carefully selected- and weeded-with reference to the pur- pose to be served. As with Table 1, also, the formula pro- vides only for a minimum. When it is seen, e.g. in the notes on lines 2 and 4, out of what this minimum is constructed, few institutions should be willing to stay there. Averages. Similarly as for Table 1, the figures sugested here must be construed as averages. Obviously, courses in court stenography or in conversational Spanish do not require the same library support as courses in theatre or decorative arts. Government publications. No special 24 C. K. Moore, Electronics; a Bibliographic Guide (Macmillan, 1961). Over 2,000 titles in 68 subject areas. 25 American Institute of Physics, Check List of Books for an Undergraduate Physics Library (New York: AlP, 1962). 1,883 titles. · 26 U. S. Library of Congress, General Reference and Bibliography :Pivision : A Guide to the Study of the United States · of America (.Washington: U. S. Gov- ernm(mt Printing Office, 1960). 6,500 titles . 21 Association of American Law Schools, Proceed- ings, 1955, p. 325. ' ·-. ,. Quantitative Criteria for Adequacy I 377 TABLE 2 FoRMULA FOR EsTIMATING THE SIZE FOR LIMINAL ADEQUACY OF JUNIOR OR CoMMUNITY CoLLEGE LIBRARIEs (1) To a basic collection, viz.: 1. A collection to support a two-year gen- eral education or liberal arts (transfer or university parallel) program . . . Add for each of the following as indicated: 2. Faculty member (full time equivalent) 3. Student (full time equivalent) . 4. Subject field of study, either transfer or terminal, in which courses are offered beyond the standard general education or liberal arts transfer pmgram . prov1s10n has been made for these; to the extent included, they would be con- sidered as books or periodicals. Title-volume ratios. Same as for Table 1. Line 1, Column 2. Similarly as for the senior colleges, there have been attempts to prepare basic selections of books for junior college libraries, of which the more important are as follows: LIST Mohrhardt28 Bertalan29 Trinkner30 . DATE 1937 1954 1963 TITLES LISTED 5,300 4,000 20,000 The earlier of these are out of date, and none is now authoritative. It is con- sequently not possible to give to the ini- tial step in the formula of Table 2 even the degree of empirical support which is available for Table 1. The development of such support would be an important step toward the improvement of the standards for junior college libraries. Line 1, Column 4. The number of periodicals is arbitrarily set at one half the number for the four-year colleges. Line 2. This provision amounts to few- 28 F. E. Mohrhardt, A List of Books for Junior Col- lege Libraries (Chicago: American Library Associa- tion, 1937) . 211 F. J . Bertalan, Books for Junior CoUeges (Chi- cago: American Library Association, 1954). 3° C. L. Trinkner, Basic Books for Junior CoU'eges ( Northport, Alabama: Colonial Press, 1963). BOOKS PERIODICALS TOTAL Titles Volumes Titles Volumes Volumes (2) . (3) (4) (5) (6) 12,500 15,000 125 1,875 16,875 30 36 1 15 51 . . .. 4 . . . 1 5 100 120 3 45 165 er than two books per faculty member per year (if spread over sixteen years) plus one periodical.1 8 Line 3. This item provides for addi- tional copies (not titles) required by the size of the student body. At the rate suggested the basic collection could be duplicated by the time there were 3,375 students. This figure obviously needs testing in practice. Line 4. This item provides for each ad- ditional . subject at the rate of six titles per annum with replacement over a six- teen-year period.18 In this connection, it may be noted that for the diversified pro- gram of the community college as con- trasted with the narrower one of the junior college, the recent Rutgers Guide has the following to say: The community college library should probably be larger than that of a compa- rable-sized four-year liberal arts college ... because a greater amount of materials is needed to maintain the diversified programs offered by a comprehensive community col- lege.31 ExAMPLES oF APPLICATION oF THE FoRMULAs In Tables 3-5 the formulas of Tables 1-2 have been applied, by way of il- 81 F. P. Merlo and W. D. Walling, Guide for Plan- ning C