College and Research Libraries CHRISTINA E. BOLGIANO and MARY KATHRYN KING Profiling a Periodicals Collection Lib.raries faced with budget cuts and rising book and periodical prices can- not afford to be without solid information upon which to base collection de- velopment decisions. Evaluative methods are needed to determine the scope, quality, accessibility, and usefulness of the existing collection so that de- velopment can proceed in a direction parallel with the current needs and future goals of those whom the library serves. Specific methods for obtain- ing such data about a periodicals collection are described. Several ap- proaches were used, including analysis of unmet user needs as shown in interlibrary loan requests, comparison with major bibliographies, analysis of the accessibility of the collection through available abstracts and indexes, and determination of the relationship between the collection and the academic programs it is designed to support. IT SEEMS A PARADOX that librarians these days must often use their expertise not so much to develop their collections as to iden- tify those areas that can safely be left unde- veloped. The days of openhanded spending are over; few institutions can now afford to purchase materials without a carefully for- mulated, comprehensive developmental plan. The problem remains: how to evaluate the collection in terms of assigning priorities for the purchase of materials. Since the price increases of periodical subscriptions have been among the most staggering of the blows to library funding in recent years, much of the interest in collec- tion development has focused on this area. · A growing body of literature has developed describing procedures for identifying lesser used periodicals for the purpose of subscrip- tion cancellations. Studies such as those by Holland, Windsor, and Maxin attempt to objectively measure actual periodical use and relate user patterns to the cost of ser- vicing apparent needs.l.2·3 Various methods of citation analysis, such as those described Christina E . Bolgiano is coordinator of techni- cal services, and Mary Kathryn King is peri- odicals librarian in the Madison Memorial Li- brary, james Madison University, Harrisonburg, Virginia . by Chambers and Healey and Scales, meas- ure ascribed use to determine the research needs of the user community. 4 •5 While literature of this nature provided a background of ideas for methods of deter- mining specific use characteristics in a given collection, our objective was to obtain a more comprehensive view of our periodicals holdings. Cancellation of subscriptions was one welcome result of our study, but the major goal was to evaluate our collection as a functioning unit. Such a study seems particularly appro- priate at this point in our institution's his- tory. In the early 1970s James Madison University entered a period of rapid growth and change: within five years the student population almost doubled to reach 8,000, the faculty increased accordingly, and many new academic . programs on both the bache- lor's and master's level were developed. The collection grew from 160,000 to 232,000 volumes; periodical subscriptions increased to 1,875. When statewide budget cuts made every penny importanr for collection main- tenance, the need for reappraisal became pressing. The profile of the periodicals collection was composed of four major elements: data on the library's currently received peri- odicals; data on the periodicals holdings /00 100 I College & Research Libraries • March 1978 supporting each of the twenty-six academic departments on campus; an analysis of the interlibrary loan periodical transactions for the previous calendar year; and the analysis of journal citations in master's theses in three academic departments over the past five years . Each of these phases, described in detail below, entailed various kinds of re- search which produced a large amount of very specific information. The periodicals office was responsible for compiling the data for the first two phases, while other staff members were called upon to help with the third and fourth phase~. The project was coordinated by the dean of libraries. Since the periodicals department is perhaps the most understaffed area in our library, the project necessarily often had to be set aside and, therefore , took close to eight months to complete; less time would have been needed, however, had more in- tensive work been possible. DATA ON CURRENTLY RECEIVED PERIODICALS The first phase of our study attempted to gain an overview of our currently receiyed periodicals (1,875 titles) th~ough several kinds of analysis: a definition of the extent to which these periodicals were accessible through the subject indexes and abstracts available in the library; a comparison of our current holdings with Farber's and Katz's bibliographies; 6 •7 and a delineation of our back-file situation. ·we began by assigning code letters to each of our subject abstracts and indexes, and as . the lists of periodicals covered by each abstract/index were checked against our shelf list, the appropriate code letters were marked on the shelf list. As we pro- gressed in this endeavor, we found that we were actually identifying two sides of the same coin: not only were we tallying the number of titles in our current collection covered by specific abstracts or indexes, but we were also discovering the specific abstract/index in which each title could be found. Having this last bit of information on the shelf list is of continuing benefit, since it enables us to help patrons who need to know which index to use in locating articles in a particular journal. At the same time that the shelf list cards were coded, the list of periodicals covered by each abstract/index was marked, so that we were able to determine the percentage held of the titles in each abstract/index .. For the checking of Farber and Katz, the shelf list cards were photocopied (forty en- tries to a sheet) and entries were marked F or K. Our holdings were also put on these sheets, and from that information we were able to color code each currently received title according to whether we had two to five years of back-files, more than five years of back-files, or complete back-files. Fun- damental to statistical accuracy in phases 1 and 2 was the determination of a cut-off date (September 1975) after which newly ordered or newly arriving titles were not considered. Data obtained through these procedures were concisely stated on one page, (see fi- gure 1). While only the more general or comprehensive abstracts and indexes were included on this sheet, those that pertained to specific subject areas were listed on the data sheets for related academic depart- ments. In reviewing the results of this phase of the study, it was apparent that the science and technology journals were the least well represented in the collection. The staff had felt . intuitively that this was the case; that feeling was now documented. DATA ON THE PERIODICALS SUPPORTING ACADEMIC PROGRAMS The second phase of the study focused on the support our current titles gave each academic department. To begin to deter- mine this, we needed to assign each title to an academic department. This proved to be the most difficult part of the study because so many journals fit equally well into several areas of study. Records for subscriptions that had been entered in the last several years contained the name of the requesting department, but those formed only a small percentage of the current titles and did not always help since faculty sometimes re- quested titles outside their field. For each academic department, three lists were prepared: titles assigned primarily to that department, with the expiration date and the cost (cost was broken down to an annual rate) ; titles supportive of the de- partment's program but primarily assigned \ Profiling a Periodicals Collection I 101 Number of periodical titles currently received on subscription 1,875 No. of titles with more than 2-year back files No. of titles with more than 5-year back files 480 1,280 No. of titles with complete back files 437 Cost of current subscriptions on annual basis Number of currently received titles not $81,480.30 indexed by any subscription service ( 448 Number of currently received titles indexed by at least one 1,427 Number indexed by 3 or more subscription services 883 Percent held of titles listed in the following: Farber - Classified List of Periodicals for the College Library 62 % Katz - Magazines for Libraries 20 % Readers' Guide 78 % Humanities Index 72.5% Social Sciences Index 64.3% Applied Science & Technology Index 17.8% Biological & Agricultural Index 34.7% Chemical Abstracts - 100 Most Cited Journals Education Index 53 % 79 % Business Periodicals Index 49.6% Fig. 1 Profile of Periodicals Collection of James Madison University (Sept. 1975) to other departments;· and supportive titles received on government depository and other free items. Once the lists had been made, a data sheet similar to that for the entire current collection was prepared for each depart- ment. The number of currently received ti- tles primarily supporting that department, the number and kind of back files, the an- nual cost, and the percentage held of titles covered by the appropriate abstracts and indexes-plus the alphabetical lists of titles-were sent to each department. Thus each department was provided with some means to judge the adequacy of ' the peri- o