ACRL News Issue (B) of College & Research Libraries 101 Standards for University L ib raries F o r e w o r d The following statem ent of university library standards has been prepared by a joint committee established by the Association of Research Librar­ ies and the Association of College and Research Libraries. A draft of the statement appeared in the April 1978 issue of C o lleg e & R e se a rc h L i­ b r a r ie s News. In August 1978, the Joint ARL-ACRL Commit­ tee on University Library Standards revised this draft. On O cto b e r 2 6 , 1978, the A RL m em ­ bership unanimously endorsed the statement as revised. At the ALA Midwinter M eeting in Janu­ ary 1979, the ACRL Board also voted to ratify the revised statement. “Standards for University L i­ braries” is being published in its final form in this issue of C & R L N ew s for the information of ACRL members. S t a n d a r d s f o r U n i v e r s i t y L i b r a r i e s P rep a r e d b y a jo in t com m ittee o f th e A ssocia­ tion o f R esea rch L ib ra r ies a n d th e A ssociation o f C o lleg e a n d R e sea rch L ib ra r ies , a division o f th e A m erican L ib ra r y A ssociation. Introdu ction These standards have been prepared to assist faculty, university administrators, librarians, ac­ crediting agencies, and others in the evaluation and improvement of university library services and resources. These statements are intended to apply only to those institutions of higher educa­ tion which have been characterized by the Car­ negie Commission on Higher Education as “doc­ toral granting institutions.”1 All of these institu­ tions em phasize graduate study, professional education, and research . D esp ite these basic similarities, university libraries are also charac­ terized by a high degree of individuality, particu­ larly with respect to policies, programs, respon­ sibilities, and traditions. Hence, these standards are not intended to establish normative prescrip­ tions for uniform application. Rather, they are m eant to provide a general framework within which informed judgment can be applied to indi­ vidual circumstances. The fundamental assumption of these standards is that the library has a central and critical impor­ tance in a university. This importance has been recognized repeatedly by analysts of higher edu­ cation. In his 1966 report to the American Coun­ cil on Education, Allan M. Cartter, for example, stated: The library is the heart of the university; no other single non-human factor is as closely related to the quality of graduate education. A few uni­ v e rs itie s with poor lib ra ry re s o u rce s have achieved co nsid erable strength in several d e­ p artm ents, in some cases because laboratory facilities may be more important in a particular field than the library, and in other cases because the universities are located close to other great library collections such as the Library of Congress and the New York Public Library. But institu­ tions that are strong in all areas invariably have major national research libraries.”2 As with all institutions, universities and their libraries have experienced considerable change over time. Further changes are taking place now, and others clearly lie ahead. Particularly notewor­ thy is the increasing sense of interdependence and commitment to coordination among universi­ ties generally. With regard to university libraries, the following developments are particularly im­ portant: the growth o f interlibrary cooperation, especially resource sharing; the strengthening and expansion of service programs, such as biblio­ graphic instruction; the increasing importance of recorded information in nonprint formats; the ap­ plication of automated systems to library opera­ tions and the growth o f machine-readable data bases; the closer interaction between librarians and faculty and the improved status of librarians within the university; increased stress on the ef­ fectiveness and efficiency of operations. A recog­ nition of such trends and their importance is fun­ damental to these standards. Recognizing the increasing interdependence of u n iv e rsitie s in d ev elo p in g and m ain tain in g scholarly resources, these standards are intended to provide guidance in identifying that level of li­ brary self-sufficiency which is essential to the health and vigor of a university and its academic programs. T he general assumption is that the primary obligation o f a university library is to m eet the instructional and research needs o f the students and faculty at that university. However, no university library can acquire all of the re­ corded information that its clientele may find use­ ful. An attempt is made, therefore, to recognize the m echanism s being developed to prom ote cooperative access to scholarly information, to identify the cu rrent lim itations o f interd ep en­ dence, and to enumerate the factors which are essential in maintaining an environment in which instruction and research can flourish. Care has been taken to limit the standards to su ccinct statem en ts focusing on the elem ents judged to be most critical in determ ining the adequacy of a university library. Amplification of the principles identified in the standards is pro­ vided in the form of commentary. 102 S e c t i o n A : S e r v i c e s Standard A .l In o r d e r to su p p o r t th e in stru ction al, r e s e a r c h , a n d p u b lic s e rv ic e p ro g ra m s o f th e university, th e s e r v i c e s o f f e r e d b y a u n iv e r s ity l i b r a r y s h a ll p r o m o te a n d f a c ilit a t e e ffe c tiv e use o f r e c o r d e d in form ation in a ll fo r m a ts b y a ll o f th e lib r a r y ’s clien tele. C o m m en tary on S ta n d a rd A .1 In developing and implementing its program of service, a university library should give priority to the needs of the students, faculty, and other academic staff of the university, who may be said to constitute the library’s “primary c lie n te le .” W hile it may also have obligations or com m it­ ments to other clienteles or constituencies, the library should recognize that these are secondary. A university library should provide the follow­ ing services: reference and information services which are available at adequately identified and d esig n ated p o in ts d u ring e s ta b lis h e d s e rv ic e hours; specialized and in-depth assistance to indi­ viduals in the use of the library’s resources; bib­ liographic instruction programs; services which w ill fa c ilita te a c c e s s to n o n p rin t m ed ia and machine-readable data bases; and services which will facilitate access to recorded information in other library collections. These services should be designed to m eet ef­ fectively the whole range o f different informa­ tional and bibliographical needs that arise in the various academic areas and in all other parts of the university. W hile universities should place great emphasis on m eeting the intensive library needs o f gradu­ ate students and faculty, they should be careful to provide adequately for the needs of undergradu­ ate students. Finally, university libraries should recognize that, to one degree or another, they share a re ­ sponsibility with all research libraries to support higher education in general and each other in particular through cooperative efforts. S ta n d a rd A .2 In o r d e r to en su re m axim um a c c ess to its c o l­ lection s a n d th e ir con ten ts, a u n iversity lib r a r y sh a ll m aintain re c o r d s o f its co llectio n s w h ich a r e co m p lete, con sisten t, a n d in co n fo rm ity with n a­ t io n a l b i b li o g r a p h i c a l s t a n d a r d s a n d r e q u i r e ­ m ents. C o m m en tary on S ta n d a rd A .2 T h e e x te n t o f b ib liograp hical co verage that must be provided in a particular library will de­ pend on many factors, such as whether or not the library has open or closed access stacks, the ex­ tent and nature of the library’s specialized collec­ tions, the history and traditions of the library and o f th e u n iv e rsity , and th e n a tu re o f sp e cific cooperative arrangem ents that the library may have entered into with other libraries and library consortia. To ensure effective access to its collections as well as to increase its operational efficiency, a university library’s bibliographic records should conform to recognized standards of cataloging and c lassificatio n , and its b ib lio g ra p h ic apparatus should be internally consistent. Its bibliographic records should be adjusted in conjunction with p erio d ic in v en tories o f th e c o lle ctio n s. E v ery multi-unit university library should have a union catalog of its cataloged holdings. S ta n d a rd A .3 W ithin th e limits o f th e university's p a r tic u la r resp o n sib ilities a n d p r io ritie s, a u n iversity lib r a r y sh a ll p r o v id e m axim um a c c e ss to its c o llec tio n s f o r a ll o f its clien tele. C o m m en tary on S ta n d a r d A .3 Various factors are involved in providing access to a lib ra r y ’s c o lle c tio n s , such as circu la tio n policies and procedures, service hours, security arrangem ents, and actual op erating efficiency. W hile practices vary significantly from library to library, certain principles should be followed in each library. Most items in the library collections should be readily available both for consultation in the library and for circulation to authorized c lie n t e le . A ccess to and c ir c u la tio n o f ra r e , fragile, and high-demand materials should be ap­ propriately controlled and restricted. To ensure maximum availability o f the collections to those authorized to use them , terms of loan should be carefully set and should generally b e similar for all user categories. Adequate precautions should be taken to con­ trol loss o f or damage to the library’s collections. The prompt return in good condition of all circu­ lated materials should be effectively enforced for all borrowers. Circulation procedures and stack maintenance operations in a university library should be effec­ tive and efficient. T here should be a regular and continuing program of shelf reading. Library ser­ vice hours should be responsive to high- and low-use periods, to the num ber of branch, de­ partmental, and other special libraries in the sys­ tem as well as to the availability of alternative study space. S e c t i o n B: C o l l e c t i o n s S ta n d a rd B .1 A u n iv ersity l i b r a r y ’s c o lle c t io n s s h a ll b e o f su fficien t siz e a n d s c o p e to su p p o r t th e u n iv er­ sity’s tota l in stru ction al n eed s a n d to fa c ilita t e th e u n iv ersity ’s r e s e a r c h p ro g ra m s. C o m m en tary on S ta n d a rd B .1 A university library should provide all of the resources that are necessary for direct support of 103 the university’s full instructional programs at both the undergraduate and the graduate levels. If these resources are not readily available in the library, the instructional programs cannot be car­ ried out successfully. These resources include re­ quired and assigned readings, reference and bib­ liographical materials, basic journals and serials, as well as any other library materials that under­ graduate or graduate students are expected to be able to consult readily in their courses of study, or in the preparation of theses and dissertations. Weak collections can hamper research. The ac­ cumulation and preservation of substantial collec­ tions and the implementation of comprehensive acquisition programs must be recognized as pro­ viding a resource whose presence within a uni­ versity is essential to the conditions under which knowledge is effectively increased and transmit­ ted. It is clear that no university library can be expected to possess in its collections all o f the recorded information which faculty or doctoral students may need to consult as they pursue their research. Nevertheless, it is essential that collec­ tions be of such size, scope, and quality that they p ro m ote ra th e r than re s tr ic t resea rch . W hile every library should take care to develop collec­ tions whose areas o f concentration reflect and support the academ ic p riorities and strengths within the university, interlibrary arrangements, which have long been established for the mutual support of exceptional research needs, must con­ tinue to be relied upon to supplement even the most comprehensive research collections. The continued rapid growth o f scholarly litera­ ture and the costs of providing access to this lit­ erature for those in the university community have necessitated formal and informal arrange­ ments among libraries to ensure maximum access to this literature. Common methods of sharing resources and improving access have been loans between libraries, provision of visiting privileges for scholars, agreem ents on the acquisitions of materials, and sharing o f bibliographic informa­ tion. W hile in terlibrary cooperation, as presently practiced, may not promise large cost savings in the immediate future, significant improved meth­ ods of supplementing local resources are in the active planning stages. University libraries must participate in the development of these new ac­ cess mechanisms to ensure that local, regional, national, and international interests are effectively served. Attempts have been made to identify precise quantitative measures of adequate collection size and growth rates for a university library. No such formula has yet been developed which can be generally applied. At present, such formulas as exist can only yield approximations which indicate a general level of need. If they are applied arbi­ trarily and m echanically, they can distort the realities of a given situation. Nevertheless, quan­ titative measures are increasingly important in guiding the qualitative judgment that must ulti­ mately be applied to university libraries and their collections. One technique is the use of regres­ sion analysis to facilitate the comparison of similar libraries to one another;3 another of some general applicability is the “index of quality” developed by the American Council on Education for relat­ ing library collection size to graduate program quality.4 S ta n d a rd B .2 A university lib r a r y ’s c o llectio n s sh a ll b e d e ­ v e lo p e d system atically a n d consistently w ithin th e term s o f explicit a n d d e ta ile d policies. C om m en tary on S ta n d a rd B .2 Given the great breadth of university library collections and the wide variations in depth of collections among subjects held, it is essential that there be a collections development policy to guide the selection and acquisition of materials. By establishing such a policy, librarians seek to ensure that the library’s collections are planned and d evelop ed in relation to the u n iv ersity ’s a ca d e m ic, re s e a rc h , and s e r v ic e goals and priorities and within the limits of resources avail­ able. Working in close consultation with faculty and adm inistration, lib rarian s, particularly su b ject specialists, should assume the responsibility for drafting and implementing this policy. Recognizing the inherent difficulties in collec­ tion development, it is imperative that the library have full and continuous access to information about all developments, actual and planned, in the academic, research, and service programs of the university and its components which affect the library. O n ce codified , th e lib ra r y ’s c o lle c tio n d e ­ velopment policy should be made known to and endorsed by the university faculty and adminis­ tration. To ensure that this policy reflects changes within the university, the policy should be regu­ larly and carefully reviewed. S ta n d a rd B .3 A university lib r a r y ’s collection s sh a ll contain a ll o f th e v a r ied fo r m s o f r e c o r d e d in form ation. C om m en tary on S ta n d a rd B .3 The university library has traditionally been recognized as the repository within the university for the printed information needed to support the university’s instructional and research programs. As recorded information becom es increasingly available in a variety of nonprint formats, such as films, sound recordings, and video tapes, it is ap­ propriate that this material, except where needed exclusively for classroom use, also be acquired, organized, and made available through the uni­ versity library.5 104 S e c t i o n C: P e r s o n n e l Standard C .1 A u n iv er sity l i b r a r y s h a ll h a v e a s u ffic ien t n u m ber an d variety o f p erso n n el to d ev elop , o r ­ ganize, a n d m aintain such collection s an d to p r o ­ v ide such re feren c e an d in form ation serv ices as will m eet th e university’s needs. C om m entary on S ta n d a rd C .1 The size of a university library’s staff is deter­ mined by many factors, including the number of physically separate library units, the number of service points requiring staff, the number of ser­ vice hours provided, the num ber and special characteristics of items processed annually, the nature and quality of the processing to which they are subjected, the size of the collections, and the rate of circulation of the collections. In­ terinstitutional cooperative arrangem ents may also affect staff size. As such factors vary widely from one institution to another, no single model or formula can be provided for developing an op­ timum staff size. A university library should have on its staff a variety of personnel: professional, clerical, and student-assistant staff. The librarians should per­ form the core academic and professional functions of the library: collection development, reference and information services, and substantive activi­ ties related to the bibliographic control of mate­ rials. All categories of personnel should have ap­ propriate education and experience, including, when necessary, graduate or professional degrees in their particular specialties. T he recognized terminal degree for librarians is the master’s de­ gree from an American Library Association ac­ cre d ite d lib rary school program , although additional graduate degrees may sometimes be desirable. The deployment of personnel within a specific university library is related to the range of opera­ tions and services provided by that library and to its total workload requirements. S ta n d a rd C .2 P ersonnel p ra ctices within a university lib ra ry shall b e b a s e d on soun d, c o n tem p orary adm inis­ trative p ra c tic e a n d sh a ll b e consistent with p e r ­ sonn el p ra ctices within th e university as w ell as the g oals an d p u rp o ses o f th e library. C om m entary on S ta n d a rd C .2 The terms and conditions of employment of the several categories of staff in a university library should be consonant with the established terms and conditions of employment of staff in related categories elsewhere within the university. Terms and conditions o f employment for librarians, for example, should parallel those of the rest of the university’s academic staff, just as terms and con­ ditions of employment for the library’s clerical and student staff should parallel those of similar employees within the university as a whole. A comprehensive university library personnel managem ent program should address re c ru it­ ment, appointment, promotion, tenure, dismissal, appeal, definition of position responsibilities, clas­ sification and pay plans, orientation and training programs, review o f employee performance, staff development, and counseling. More specific guidance on these m atters is provided in the following documents: “Statement on Faculty Status of College and University L i­ brarians”6 and “Library Education and Personnel Utilization.”7 S e c t i o n D : F a c i l i t i e s S ta n d a rd D .1 A university lib ra ry shall h av e fa c ilitie s w hich m eet th e p resen t a n d a n ticip a ted fu tu r e re q u ire ­ ments o f th e university an d its program s. C om m en tary on S ta n d a rd D .1 A university library’s buildings should be of sufficient size and quality to house the collections and to provide sufficient space for their use by stu d ents, faculty, and oth er c lie n te le . T h e re should also be adequate space for the library op­ erations necessary for the provision of its ser­ vices. Adequacy of facilities cannot be d e te r­ mined simply on the basis of present require­ ments. The size and composition of the univer­ sity’s enrollment, the nature of its instructional and research programs, the form and publication rate of library materials strongly influence library requirements, and it is necessary that these re­ quirements be subject to continuous evaluation and planning. A university library should be attractive, invit­ ing, and carefully designed to promote opera­ tional efficiency and effectiveness of use. Specific factors relevant here include general environmen­ tal features that affect clientele, staff, and collec­ tions (light, ventilation, temperature and humid­ ity control, vertical and horizontal transportation, safety features, etc.), layout of the stacks, number and variety of reader stations, relationship be­ tween stacks and reader stations, relationship among service points, effective flow of materials, and adequacy of space for staff and operations. The fundamental consideration in designing a library building should be its function. Since the nature of collections, services, operations, and the needs of a library’s clientele can change sig­ nificantly over time, present and future flexibility is an important elem ent in library design. Al­ though the architectural style and traditions of a university may dictate certain design features for a library building, such factors should not be allowed to compromise basic functional considera­ tions.8 105 Standard D .2 L ib r a r ie s sh a ll b e so lo c a ted th at th e university com m unity will h a v e con v en ien t a c c ess to them . C om m en tary on S ta n d a rd D .2 The requirem ents o f interdisciplinary studies and research, recognition of the needs of under­ graduate stu d en ts, th e urgency o f ach iev in g op eratin g eco no m ies— these and o th er factors have revived interest in centralizing physically dispersed library units in order to improve access to resources and avoid costly duplication in the d evelop m ent and m ain tenan ce o f co llection s. There are circumstances, however, such as cam­ pus geography, intensity of use, and size of col­ lections which may continue to justify the main­ tenance of multiple library units. Remote storage facilities may also be established in attempting to deal with space inadequacies although this usually inhibits convenience o f access. W here the pattern of decentralization persists in any form, it is im­ portant that libraries be located so as to minimize inconvenience to all library users.9 S e c t i o n E: A d m i n i s t r a t i o n a n d G o v e r n a n c e S ta n d a rd E .1 T he p la c e o f th e university lib ra r y within th e ad m in istrativ e a n d g o v e rn a n c e stru ctu re o f the university sh a ll b e c lea rly id en tified ‚ a n d th e re­ sp o n sibilities a n d au th ority o f th e lib ra r y ad m in ­ istration a n d its c h i e f adm in istrative o ffic e r sh a ll b e d efin ed . C om m en tary on S ta n d a rd E .1 If th ere is am biguity within the university community as to the particular place occupied by the library within the administrative and gover­ nance structure of the university, and if the au­ thority and responsibilities of the library’s chief administrative officer are not clearly identified, m isunderstanding, co nflict, and confusion can sometimes result to the detrim ent of both the university and its library. Because it is closely re­ lated to instruction and research, the university library should be formally recognized as one of the major academic units within the university, and its chief administrative officer should partici­ pate regularly and directly in university-w ide academic planning and decision making. For simi­ lar reasons, this person should report directly to the chief academic officer of the university. T he long-recognized need in institutions of higher education to involve faculty in library mat­ ters has led to the institutionalization of the advi­ sory library committee. Because of the fundamen­ tal importance of the library to instruction and research and the consequent need for close, con­ tinuing interaction between the faculty and the library, the existence of the library committee is valuable. The committee should be advisory, and its responsibilities should be clearly delineated. S ta n d a rd E .2 The university lib ra r y ’s ow n ad m in istrativ e an d g o v e r n a n c e s tr u c tu r e s h a ll b e c le a r ly s p e c ifie d a n d s h a l l b e c o n s o n a n t w ith t h e g o v e r n a n c e str u c tu re o f th e u n iv ersity as w ell a s w ith th e p a r tic u la r n eed s a n d req u irem en ts o f th e lib rary . C om m en tary on S ta n d a rd E .2 In order to facilitate effective organizational ac­ tivity and decision making, it is essential that the administrative and governance structure of the university library itself be clearly specified. This will involve identifying the roles and respon­ sibilities of all categories of library personnel in the governance of the library. It is essential that library governance reflect the principles and prac­ tice followed elsewhere within the university, al­ though they should be modified as necessary to embody those conditions and issues peculiar to an academic library. S ta n d a rd E .3 T h ere sh a ll b e a close adm in istrative relation ­ sh ip am on g a ll lib r a r ie s w ithin th e university to th e en d that lib ra r y u sers m ay m a k e fu ll a n d e f ­ f e c t iv e use o f lib ra ry reso u rces a n d services. C om m en tary on S ta n d a rd E .3 No single pattern of library administration will serve all universities equally well, but whatever pattern an institution chooses should have as its principal purpose the equitable distribution of li­ brary re s o u rc e s and s e rv ic e s. T h e ne ed s of scholars differ from discipline to discipline and often the needs of students differ from those of faculty. These competing interests cannot always be reconciled, but one important task of library administration is to achieve as much balance as possible in the provision of services to all groups. However administrative relationships among library units within a university are determined, it is e ss e n tia l th at a d e q u a te co o rd in atin g mechanisms be established and enforced to en­ sure that service policies are in reasonable har­ mony, that costs related to duplication are con­ trolled, and that access to all library collections is maximized. S ta n d a rd E .4 A university lib ra r y ’s m a jo r p o licies a n d p r o c e ­ du res sh a ll b e clea rly d efin ed a n d reg u larly r e ­ view ed. C om m en tary on S ta n d a rd E .4 In order to ensure that it is effective internally and understood externally, a university library should clearly define its major policies and pro­ cedures and record them in written form. The written statem ents of policy should be readily 106 available to all members of the library staff, and policies which have external relevance (such as the library’s collection development policy or cir­ culation policy) should be accessible to the li­ brary’s clientele and to others who may need or desire to consult them. These policies, as well as the practices that implement them , should be regularly reviewed to ensure that they continue to be appropriate. S e c t i o n F : F i n a n c e S ta n d a rd F .1 B u d g eta ry su p p o r t f o r th e university lib ra r y sh a ll b e su fficien t to en a b le it to fu lfill its o b lig a ­ tions an d responsibilities as id en tified in th e p r e ­ cedin g stan dards. C om m entary on S ta n d a rd F .1 The total budgetary needs of a university li­ brary can be determined only in relation to its responsibilities. Many attempts have been made to develop formulas or other “objective” meas­ ures for determining the budgetary requirements of a university library. These measures range from matching funding with student enrollment to defining a minimum percentage of the total university G and E budget which should be de­ voted to the library. Such “objective” approaches to budget determination do not always take cog­ nizance of the range and complexity of demands which any university library must meet, as well as the significantly different library needs of dif­ ferent universities. These conditions also make it impossible to identify a viable model that can be applied to all university libraries for allocating their budgets by major category (salaries and wages, acquisitions, binding, miscellaneous supplies, and other ex­ pense). Allocation ultimately depends on local re­ quirements and priorities. For example, if a uni­ versity library is expected to operate a substantial number of discrete units with parallel and dup­ licative activities, its expenditures for salaries and wages will be higher than if this were not the case. Under any circumstances, it is essential that a university library be provided with sufficient funding to enable it to develop appropriate col­ lections, provide appropriate services, carry out necessary operations, and satisfy identified expec­ tations and requirements. I f funding is less than is necessary to fulfill these obligations, the library will be unable to meet university needs. A university library should be expected to op­ erate on a sound financial basis. To do this, the library and its administration must be able to identify and support its fiscal request effectively and to report adequately on expenditure of funds. S ta n d a rd F.2 T he university lib ra ry b u d g et sh a ll b e a d is­ tinct p a rt o f th e university s bu dget, an d it sh a ll b e d e v e lo p ed an d m an ag ed by th e c h i e f adm in is­ trative o ffic e r o f th e university library . C om m en tary on S ta n d a rd F.2 The authority to prepare, submit, defend, and administer the university library budget should be delegated clearly and explicitly to the chief administrative officer of the university library. He or she should have full responsibility for manag­ ing this budget as well as the authority necessary to maximize the use o f the lib rary ’s total re ­ sources. He or she should have the same degree of latitude and responsibility that is exercised by oth er major administrative officers within the university. The library should be responsible for preparing adequate and regular reports on ex­ penditures throughout the year. These reports should conform to the university’s requirements and, where necessary, to its standardized proce­ dures and practices. Because of the importance of the library within the university and the need that it respond effec­ tiv ely to changing d em an ds, p rio r itie s , and academic programs, it is essential that the library budget be developed in relationship to and with full cognizance of the total university budget- planning process, and that the library’s chief ad­ ministrative officer be directly and significantly involved in this process. R e f e r e n c e s 1. Carnegie Commission on Higher Education, A C lassification o f Institutions o f H ig h er E d u ca ­ tion (Berkeley, Calif.: The Commission, 1973), p. 1 -2 , 9 -2 2 . This publication identifies 173 “doctoral granting institutions.” 2. Allan M. Cartter, An Assessment o f Q uality in G r a d u a t e E d u c a t io n (W ashin g to n, D .C .: American Council on Education, 1966), p. 114. 3. William J. Baumol and Matityahu Marcus, E co n o m ics o f A ca d em ic L ib r a r ie s (Washing­ ton, D .C .: American Council on Education, 1973). 4. Cartter, An A ssessm ent o f Q uality in G ra d u a te E d u ca tion , p. 114. 5. The best recent discussion of the importance of nonprint media for higher education is Car­ negie Commission on Higher Education, The F o u rth R evolu tion : In stru ctio n a l T ech n o lo g y in H ig h e r E d u c a tio n (New York: McGraw- Hill, 1972). 6. In Faculty Status f o r A cadem ic L ib ra ria n s: A H istory a n d Policy Statem ent (Chicago: Ameri­ can Library Assn., 1975), p .35-38. 7. “Library Education and Personnel Utilization” (Chicago: American Library Assn., 1976). 8. Considerable valuable information is available in several publications, the best of which re­ mains Keyes D. Metcalf, Planning A cad em ic a n d R e sea rc h L ib r a r y B uildin gs (New York: McGraw-Hill, 1965). 9. This issue has been the subject of considerable 107 analysis. See, particularly, Ralph E . Ellsworth, T h e E co n o m ics o f B o o k S to ra g e in A cad em ic L ib r a r ie s (M etuchen, N .J.: T he Association of Research Libraries and the Scarecrow Press, 1969). Also useful is Je ffr e y A. R affel and Robert Shishko, S y stem a tic A n aly sis o f Uni­ v e r s it y L i b r a r i e s (C a m b rid g e, M ass.: M IT Press, 1969). A p p e n d i x Q u a n t i t a t i v e A n a l y t i c a l T e c h n i q u e s f o r U n i v e r s i t y L i b r a r i e s T he university libraries1 to which quantitative measures might b e applied are so complex, so di­ verse in the programs they support, and so dif­ fe r e n t from e a ch o th e r th a t it is e x tre m e ly difficult, if not impossible, to devise a common statistical measure which could be applied to all of them. This problem is further complicated by the ch aracter and inadequacy o f the currently available data. Herman Fussier, for example, ob­ serves that “libraries, like universities, tend to have very inadequate analytical data on their own operations and performance. Such data, especially as they relate to costs and system responses to user needs, are critically important in any effort to improve a library’s efficiency and responsive­ ness. ”2 Fritz Machlup, in the course of his recent efforts to measure the holdings and acquisitions of libraries on a broad scale, has complained about the lack o f adequate data.3 O ther observers have challenged the utility o f present library data col­ le c t io n .4 T h ey focus on p e r c e iv e d failu res to measure performance or effectiveness. Neverthe­ less, academic institutions do com pete for faculty and students, and one of the elem ents in this com petition is the adequacy of library services and collections. Com parative ju d gm ents about academic libraries are made, and these compari­ sons can be aided by quantitative measures. U n fortu nately , much o f th e data which are needed to actually make interinstitutional com ­ parisons is not easily available, although some useful data can be obtained from ARL statistics. T h e L I B G I S and H E G IS surveys also supply data, but these are usually too old for cu rrent needs or in a form which is difficult to use. Con­ seq uently, the analyst is com pelled to rely on what is available: ARL statistics, authorities who have w ritten on the su b ject, and such lim ited surveys as he or sh e can m ake. All o f th e s e methods have varying degrees o f utility, but with the possible exception of the A RL data, none provide the raw data on which em pirically d e­ rived measures can be based. Certain “common” practices can be discerned, and the advice of au­ th o rities can be w eighed, but th e s e , how ever valuable, do not constitute quantitative measures in an em pirically derivable, logically justifiable sense. T o have reliab le quantitative m easures, the categories to be measured must be defined, and a mechanism for gathering the necessary data must be developed. In th e a b sence o f e ith e r of th ese necessary conditions, it is difficult to do more than perform what analyses can be performed on A RL data. Briefly, these fall into three categories: (a) in­ sights obtained by simple inspection of the data; (b) the construction of ratios which reduce the quantity of data to b e comprehended and facili­ tate comparison; and (c) regression analysis which perform s roughly the sam e function from the an alyst’s point o f view as the co n stru ctio n o f ratios but also requires an effort on the part of the analyst to group like institutions together and gives the analyst some indication of how well this has been accomplished (coefficient of determina­ tion). Simple inspection of ARL data, aided by rank­ ings, ranges, averages, and medians, does pro­ vide useful insights for the experienced library manager who can mentally discount obvious dis­ crepancies and differences betw een institutions and can restrict comparisons to a homogeneous group. However, to read, for exam ple, that in 1976-77 the number o f volumes in ARL libraries ranged betw een Harvard’s 9 ,5 4 7 ,5 7 6 and McMas­ t e r ’s 9 0 6 ,7 4 1 , th a t th e a v e ra g e lib ra r y held 2 ,1 2 7 ,0 4 7 , and the median was 1 ,6 5 3 ,0 0 0 may give the reader a sense of perspective, which is valuable, but it is of limited use in drawing com­ parisons between rather different institutions. A reduction o f data can be achieved by the use o f ratios or percentages, as is shown in the exam­ ple of ratio analysis below. Some of those which can be generated from existing data include: 1. The ratio of professional to nonprofessional staff 2. Expenditure for library materials as a per­ cent of total library operating expenditure 3. Ratio of salary expenditures to library mate­ rial expenditures This kind o f data reduction aids analysis by mak­ ing the data more comprehensible. For example, among ARL libraries in 1976-77, the ratio of pro­ fessional to nonprofessional staff ranged from 1.08 to 0 .2 4 ; the average was 0 .5 1 and the median 0 .4 9 . T h e ov erw helm ing m ajority o f lib raries tended toward a pattern of one professional to two nonprofessionals. Among A RL libraries in 1976-77, expenditures for library materials as a percent o f total library expenditures ranged from 19.14 p ercent for Toronto to 50.61 p ercent for Houston. The average was 31.46 percent and the median 3 0 .0 9 percent. The vast majority of ARL lib raries tend ed to spend 3 0 p e r c e n t o f th e ir budgets on acquisitions. The obverse of materials expenditure for libraries is salary expenditure. E x p re s se d as a ratio o f salary to m aterials it ranged from 3 .6 in the case of Toronto, to 0 .8 in the case of Houston, with the median 1.9 and the average 1.93. From ratios such as these, a deeper insight into 108 library operations can be obtained, but it would be rash to conclude that all libraries should spend 30 percent of their budgets for books and 60 per­ cent for salaries or that the ratio of professional to nonprofessional should always be 1:2. Local con­ ditions dictate differing policies. A library with many branches may require a higher ratio of pro­ fessionals to nonprofessionals. Conversely, differ­ ing o p era tin g co n d itio n s, d iffe re n t ty p es of staffing may dictate different ratios. An example of a more extended kind of ratio analysis is that of Allan Cartter’s Library Resources Index, which is described in a following section. Yet, even this kind of ratio should be viewed cautiously. At b e st, ratio analysis can serve only as a back­ ground against which local conditions may be evaluated. Regression analysis also provides a form of data reduction, but it compels the analyst to attempt to group like institutions together. Baumol and Marcus provide a guide to its use in library data analysis.5 The concluding section of this appendix gives an example of its application. But the same caveats about drawing inferences that apply to ratio analysis apply to regression analysis. In addition to these, there is a growing litera­ ture on performance evaluation o f libraries which is expressed in various ways. F . W. Lancaster summarizes some of the possible approaches: “1. The ability of the library to deliver a par­ ticular item when it is needed. “2. The ability of the catalog and the shelf ar­ rangement to disclose the holdings of particular items or of materials on particular subjects. “3. T he ability of referen ce staff to answer questions completely and accurately. “4. The speed with which a particular item can be located when needed. “5. The speed with which a reference inquiry can be answered or a literature search conducted and the results presented to the library user. “6. T he amount of effort that the user must himself expend in exploiting the services of the library (including factors of physical accessibility of the library and its collections, the size and quality of the library staff, and the way in which the collections are cataloged, indexed, shelved and signposted.”6 Perform ance measures are, however, still in the early stages of their development. They may eventually prove to be extremely important to li­ braries, but they are likely to be most useful in making intrainstitutional rather than interinstitu­ tional decisions. In sum, there are no simple so­ lutions, no ready panaceas, no easily available substitutes for intelligent analysis o f available data. Exam ple o f Ratio A nalysis Table 1 below demonstrates the application of ratio analysis to library materials expenditures as a percentage of total library operating expendi­ tures. It is based on the latest (1976-77) ARL data. For the sake of brevity and because this is simply used as an example, only twenty of the total applicable n in ety -th ree institutions have been included. T he L ib ra r y R esou rces Index The Library Resources Index is a specialized index devised by Allan M. Cartter and published TA BL E 1 L i b r a r y M a t e r i a l s E x p e n d i t u r e s a s a P e r c e n t a g e o f T o t a l L i b r a r y O p e r a t i n g E x p e n d i t u r e s ( V a l u e ) f o r T w e n t y U n i v e r s i t y L i b r a r i e s , 1976-77 Rank O rder Institution Num ber Num ber Institution Name Value 1 31 Houston 50.61 2 3 Arizona 44.63 3 82 Texas A & M 44.05 4 87 VPI & SU 42.84 5 81 Texas 42.69 6 28 Georgia 42.21 7 35 Iowa 42.15 8 71 South Carolina 42.08 9 68 Rice 41.67 10 42 Louisiana State 40.19 11 20 Connecticut 40.04 12 60 Oklahoma State 39.51 13 53 Nebraska 39.30 14 80 Tennessee 39.22 15 52 Missouri 38.93 16 4 Arizona State 38.62 17 22 Dartmouth 38.30 18 24 Emory 38.23 19 1 Alabama 38.08 20 57 Notre Dame 37.87 109 T A BL E 2 T h e L i b r a r y R e s o u r c e s I n d e x A p p l i e d t o T w e n t y A R L L i b r a r i e s , 1976-77 Overall Rank O rder Total Volum es Library Overall Institution Name Volum e Added Serials R esources Index Index Index Index Index 1 Harvard 4 .4 9 2 .25 3.8 9 3.54 2 Illinois 2.74 1.95 3.43 2.71 3 Yale 3.24 2 .4 0 2.44 2.69 4 Calif., Berkeley 2.31 1.75 3.9 0 2.65 5 Texas 1.91 2 .87 2.41 2.39 6 Indiana 2.07 2 .3 9 1.71 2.05 7 Columbia 2.22 1.57 2.31 2.03 8 Michigan 2.31 1.81 1.92 2.02 9 Stanford 2.05 1.67 2.13 1.95 10 Toronto 1.87 2 .1 5 1.66 1.90 11 Calif., Los Angeles 1.84 1.44 2.26 1.84 12 Washington 1.52 2.16 1.64 1.77 13 Cornell 1.87 1.33 2.08 1.76 14 Chicago 1.83 1.60 1.76 1.73 15 Wisconsin 1.52 1.30 1.92 1.58 16 Ohio State 1.53 1.50 1.15 1.39 17 Minnesota 1.58 0.93 1.48 1.33 18 Duke 1.35 1.28 1.33 1.32 19 Princeton 1.37 1.18 1.25 1.27 20 Pennsylvania 1.31 1.08 1.10 1.16 in his An A s se ss m e n t o f Q u a lity in G r a d u a t e E d u c a tio n .7 It is an average of three indexes and is computed in the following way. First, the pool of institutions to be compared is determined. (In the example, shown as table 2, this pool is all ARL libraries and the data are for 1976-77). Sec­ ond, three variables are isolated: (a) total vol­ umes; (b) volumes added; and (c) periodicals re­ ceived. A separate index is formed for each vari­ able by finding the average for each variable and dividing the average value into the value for each institution. F or example, assume that the average number of periodicals held in ARL libraries is 15,000, and th re e in s titu tio n s have to ta ls re s p e c tiv e ly of 60.000, 15,000, and 7,500. Dividing the average, 15.000, into each of these figures yields index values of 4, 1, and .5. Similarly, values are found for each institution for the other two variables: volumes added and total volumes. Then the three index values for each institution are summed, di­ vided by three, and sorted into descending order. F or exam ple, refer to institution num ber 8 in table 2. It is Michigan. It has index values of 2.31, 1.81, and 1.92. The sum of these is 6.04. Dividing this by 3 yields 2.0 1 , the overall library resources index. Mr. C artter’s index was based on 1963-64 data. His general conclusion at that time was: “Those lib raries which fall below .5 are probably too weak to support quality graduate programs in a w ide ran ge o f field s, althou gh th ey may be adequate for an institution that specializes in technology or in advanced work in a very limited number of areas. ”8 Table 2 demonstrates an application of the Li­ brary Resources Index to twenty ARL libraries, using 1976-77 ARL data. R e g r e s s io n A n a ly sis T a b le s U sing A R L D a ta , 1975-76 In analyzing data from A R L lib ra r ie s , the strongest statistical relationships are found to exist when these libraries are categorized in some way. Therefore, by way of example, ARL libraries may be grouped in four different ways: 1. All ARL academic libraries. 2. All private ARL academic libraries in the U.S. 3. All public A RL academic libraries in the U.S. 4. All Canadian ARL academic libraries. Further, for each group additional tables may be developed that predict the values of certain different (dependent) variables based upon the value o f other (independent) variables. Six var­ iables, for example, which can be examined are: 1. Professional staff 2. Total staff 3. Gross volumes added 4. Expenditures for library materials 5. Total library expenditures 6. Current periodicals held F o r each library in each of the four groups noted above, the following predictions then can be made: 1. N u m b er o f p ro fession al s ta ff based on number of volumes held 2. Number of total staff based on number of volumes held 3. Number of gross volumes added based on volumes held 110 TABLE 3 E x a m p l e o f R e g r e s s i o n A n a l y s i s A p p l i e d t o S i z e o f P r o f e s s i o n a l S t a f f (Y) Institution Y Value Y Estim ate Residual Display Library A 37 39 - 2 X Library B 52 48 +4 X Library C 63 55 +8 X Library D 60 72 - 1 2 X least squares line normalized 4. Expenditures for library materials based on gross volumes added and volumes held 5. Total expenditures based on volumes held, gross volumes added, and total staff 6. Number of current serials based on number of volumes held Thus, for each table there can be plotted a dis­ play of variables, together with observations for each institution, and which include for each de­ pendent variable its actual value, its estimated value, and the residual, which is the difference between the actual and the estimated value. For example, assume we have the display shown above as table 3, which predicts the number of professional staff a library is expected to have based upon the number of volumes held. The first column identifies each institution; the second shows the actual value for each variable; the third shows the expected value based on the regression equation computation which has been done; the fourth is the difference between col­ umns two and three; and the fifth is a plot of the data. Looking at Library A, we see that it has thirty-seven professional staff, but based on the other libraries in its comparison class, it would be expected to have thirty-nine. The actual value is two fewer than expected, so its position on the graph is plotted to the left of the least squares line. (See any standard textbook on statistics for detailed explanation of this technique.) Libraries B and C have more professionals than would be expected, so they are plotted to the right of the line. Consequently, by inspection, the library manager can note any obvious anomalies between his or her institution and others. R e f e r e n c e s f o r A p p e n d i x 1. Doctoral granting institutions in Carnegie Commission on Higher Education. A C lassi­ fica tio n o f Institutions o f H igher E ducation (Berkeley, Calif.: The Commission, 1973). p. 1-2, 9-22. 2. Herman H. Fussier, R esearch L ib raries and Technology, A Report to the Sloan Foundation (Chicago: Univ. of Chicago Press, 1973), p.61. 3. Fritz Machlup, “ Our L ib raries: Can We Measure Their Holdings and Acquisitions,” AAUP Bulletin 62:303-7 (Oct. 1976). 4. Se e, for exam ple, Morris Hamburg and others, L ibrary Planning and Decision Making Systems (Cambridge, Mass.: MIT Press, 1974). 5. William J. Baumol and Matityahu Marcus, Econom ics o f A cadem ic L ib ra ries (Washing­ ton, D .C .: American Council on Education, 1973). 6. F . W. L ancaster, T he M ea su rem en t a n d Evaluation o f L ibrary Services (Washington, D.C .: Information Resources, 1977), p.323. 7. Allan M. Cartter, An Assessment o f Quality in G r a d u a te E d u c a tio n (W ashington, D .C .: American Council on Education, 1966). 8. Ibid ., p. 114. E ditor s Note: Members may o rd er single copies by sending a self-a d d ressed la b el to the ACRL office. N onm em bers sh o u ld include $1.00 with their order. ■■