College and Research Libraries Research Notes CONTENTS COLLEGE & RESEARCH LIBRARIES Andrew Peters 149 Evaluating Periodicals ANDREW PETERS The literature of periodical evaluation dis- cusses many critieria for judging the useful- ness of individual periodicals to a particular library. The most important factors are sub- ject r~levance, * usage, general availability' r accessibility (indexing coverage), cost, for- mat, publisher reputation, and citation fre- quency. Depending on the needs of the indi- vidual library, the language of the journal may also be important. All of these factors are naturally interdependent, but they can still be grouped in ways that will make them more coherent and easier to manage. The Central State University Library serves a student population of 8,000-9,000 with a collection of approximately 600,000 books and 4,000-5,000 current periodical subscriptions. The periodicals collection oc- cupies the third floor of the library and offers open access. Back files are not bound but are purchased in microform and shelved with the *Subject relevance refers to the importance of the journal to the curriculum in an academic li- brary. Thus the Journal of Accountancy may be vital to a business curriculum but of little or no importance to the fine arts. l Availability can refer to the availability in other libraries close by or in the state, the extent of the holding within the library, or the availability of a title on microfilm. Andrew Peters is head of the reference services department, Central State University Library, Ed- mond, Oklahoma. Evaluating Periodicals current issues. Escalating costs prompted this study to determine not only which individual titles could be canceled most easily, but also a collection development system that would ensure the most effective collection with whatever funds are available. In managing a serials or periodical collec- tion, it is essential to know precisely which titles are most important to the library and which would not really be missed very much if cancellations become necessary. One way of designating the relative value of periodi- cals uses cost/benefit ratios. These figures can be determined simply by dividing the cost of the journal by its use, or it can be computed through a more complex mathematical for- mula involving all possible factors. Some- thing in between is probably more desirable, because managing the information about all factors may be very time-consuming and it- self not cost-efficient. Also, it must be noted that we are only indicating relative impor- tance; a certain amount of subjectivity is in- herent in evaluation processes. It should also be emphasized that evaluation procedures and the information derived from them are aids to management. The process should be helpful in limiting the number and types of decisions necessary. Donald Kraft and Richard Polacsek devel- oped a formula that, with the proper group- ing of factors, can be used to develop a simple yet carefully structured evaluation system. 1 The system developed at the Central State I 149 150 I College & Research Libraries· March 1982 University Library draws on this formula and employs the following structure: the cost/benefit ratio for each paid title is equal to the cost of that journal divided by the relative worth of that journal (r. C/W., as in Kraft and Polacsek). Worth= is 1 equal to the weighted sum of other important factors: (1) relevance, which is again subdivided into ac- cessibility, subject relevance, and format/ journal reputation; (2) usage; and (3) avail- ability, subdivided into microform availabil- ity, completeness within the library, and in- state availability. The only factor eliminated from the original list is citation frequency. At this point citation statistics are rather diffi- cult to ascertain and maintain. For those who can manage this information, it could easily be attached to the format/journal reputation factor in the equation. Weights are an inher- ently subjective standard, which should be determined by the periodicals or serials li- brarian or an evaluation committee. At CSU, the factors are weighted as follows: 1. Relevance. Since this factor is intended to be the theoretical measure of a journal's usefulness, we weighted it most heavily (.7), thus ensuring that a standard field of materi- als will be available for subjects appropriate to the university. Subdivisions are weighted by varying the size of the scales applied dur- ing the evaluation process. The scales applied at the CSU Library are: A. Accessibility (Indexing): if the title is indexed, it is given five points; if not, zero. Intermediate points could be assigned if dif- ferentiation between indexing services is de- sired. B. Subject Relevance: points are assigned on a scale of 0-10, 10 being those journals most important to a subject area. The library maintains the collection of general interest and "alternative" publications, which are as- signed points on the same scale according to their importance to the collection. C. Format: points are given on the scale 1-3, 3 being the best format of the "best" pub- lisher. No journal can be assigned 0 on this factor. Relevance then equals the sum of all points assigned. 2. Usage. This factor can be tabulated two ways: points can be assigned relative to the number of uses/time period (we have chosen one year), for example, 0-10 uses/year = 0 points, 11-25 = 1 point, 26-50 = 2 points, etc., or the usage figure can be tabulated "straight." If the usage figures are not used straight, this factor should be weighted more heavily to ensure a broad range of compari- son. At CSU, the weight assigned may seem small (.2), but because the usage figures are tabulated directly, this factor assumes a very important role in the differentiation · of im- portant and unimportant titles. Maintaining the relevant statistics for this factor may be a problem for some libraries. Only a closed pe- riodicals section will provide truly accurate figures. However, a strategy devised at Case Western Reserve University2 can be adapted for most collections, perhaps even those that interfile books and periodicals. At Central State University, periodicals are housed on the same floor in an open collection shelved alphabetically by title. Microfilm back files are kept on special shelves with the current issues, so that the entire holdings of a title are openly available together. Use statistics are maintained by placing stickers on shelves during the reshelving procedure. Though ad- mittedly not exact because of those users who choose to ignore the "please do not reshelve materials" signs, these statistics are still suf- ficient measures of relative importance. Also, the effects of whatever margins of error exist are lessened by using the "straight" figures. If the use figure indicates a journal was used 28 times during the year while it was actually used 31 (or 25), the effect on the overall for- mula will be very small, though it would still be sufficiently differentiated from other jour- nals which were used 0-5, or 75-100 or more times. It should again be emphasized that the whole evaluation procedure serves merely to indicate relative importance. Use statistics as well as all other factors should be kept in that perspective. 3. Availability. This factor is not as im- portant as the others in determining the hy- pothetical importance of a journal and so should be weighted smaller than the previous factors (the CSU weight is .1). It can help provide fine distinctions between titles that are close in every other aspect. As in the rele- vance factor, subdivisions are awarded points according to their importance. Be- cause the CSU Library has a large microform collection, the availability of a title in micro- form is of more importance than the other divisions. Points (5 at CSU) can be awarded if it is available (or if storage costs are low for a title in libraries without microforms); 0, if it is not (or if storage costs are high). Whether or not the library has the complete holdings of a title may be of some importance, if a deci- sion for or against cancellation must be made. It might be more difficult to cancel the subscription of a title if the library already owns the complete run. Points (1 at CSU) can be awarded for complete holdings; 0 points, for incomplete. Finally, if a title is available to the user in another library nearby, that title may not be as necessary to the holdings of the library, but it may be much more impor- tant if it is the only subscription in the state. This factor should only be used in libraries that can equitably cooperate with surround- ing iibraries or where comprehensive systems of cooperation have been established. Other- wise, that other library might just cancel its subscription, too. Points can be awarded if the title is not immediately available; 0, if it is available nearby. The whole availability fac- tor may not be important to many libraries. If it is not worth maintaining the necessary information, this factor could be eliminated without detrimental effects to the overall evaluation system. The other factors are much more important, but this factor might help make the decision if two titles are virtu- ally alike, even down to 0 uses, and one must be canceled to meet the budget. So, the worth of a journal is equal to a weight times the relevance, plus a weight times the use statistic per year, plus a weight times the availability.* At Central State Uni- versity the equation looks like this: W; = . 7R + .2U + .lA, where the relevance (R) = indexing points + subject relevance points + format points, usage (U) = the number of uses per year, and the availability (A) = mi- croform points + holdings points (other li- brary status is ignored at this time). Evalua- tion cards are made for each paid title including all of the appropriate information for that title. The cost/benefit ratio (r; = C;fW;) can be helpful in determining the relative impor- tance of periodical titles (the smaller the number, the greate~ the importance). At *The weights in the equation must always add up to 1. Research Notes I 151 CSU, the list of current holdings is divided by subject (or academic department) and then ranked with most important at the top and least important at the bottom. It might be easy to presume that this num- ber is the objective fact of a journal's impor- tance. This assumption might be especially tempting if the librarian is confronted by the obstinate patron who demands to know why "his" or "her" journal was canceled two years ago. But the system is only developed to high- light the information and data relevant to what the library sees as the most important criteria for judgment. Knowledgeable deci- sions can then be made with more confi- dence. REFERENCES 1. Donald H. Kraft and Richard A. Polacsek, "A Journal- Worth Measure for a Journal-Selection Decision Model," Collection Management 2:129- 39 (Summer 1978). 2. W. M. Shaw, Jr. , "A Practical Journal Usage Technique ," College .& Research Libraries 39:479-84 (Nov. 1978). .GET IN TOUCH=::.~,.....- • with 9,000 of your colleagues • with developments in the field • with special projects and research • with your profession Get involved in ACRL. Join us! Write: Association of College and Research Libraries American Library Assn. 50 E. Huron St. Chicago, IL 60611 96 pages full of furniture JUST FURNITURE Gaylord introduces the most comprehensive library fur- niture resource available. In it you'll find over 1500 items designed specifically to meet the needs of your library. Featured are 200 brand new products , including spe- cialty wood shelving, computer-assisted furniture, and an expanded line of seating. In addition, the catalog also in- cludes the Sjostrom line of oak furnishings, now available exclusively from Gaylord. Gaylord "Ship from Stock" assures furniture denvery in 4 fo 6 weeks. Only Gaylord gives you this "fast company" promise! And it's free! Best of all, the new Gaylord furniture cat- alog is absolutely free. If you haven't already received your copy, just call toll free : 800-448-6160 GAYLORD Gaylord, Box 4901 , Syracuse, NY 13221 (315) 457-5070 Gaylord, Box 8489, Stockton, CA 95208 (209) 466-2576 TWX 71 0 545 0232 We 're involved! American Library Association 50 East Huron Fund. New Scholarly Publications in the Life Sciences from 1111 Dr. W. Junk Publishers • Boston/The Hague • A Kluwer Company Biochemistry Hydrobiology The Biological Effects of Glutamic Acid and Its Derivatives edited by V.A. Najjar, Tufts University (Developments in Molecular and Cellular Biochemistry 1) ISBN 90-6193-841-4 December 1981 384 pp. $85.00 Immunologically Active Peptides edited by V.A. Najjar, Tufts University (Developments in Molecular and Cellular Biochemistry 2) ISBN 90-6193-842-2 December 1981 135 pp. $55.00 Ecology Biogeography and Ecology of New Guinea by J.L. Gressitt, Bishop Museum , Honolulu (Monographiae Biologicae 42) ISBN 90-6193-094-4 January 1982 1008 pp. 2 vol $195 .00 Coastal Research in the Gulf of Bothnia edited by K Muller (Monographiae Biologicae, 45) ISBN 90-6193-098-7 December 1981 480 pp. $87.00 The Physical Environment of the Faeroe Islands edited by G.K Rutherford, Queen's University, Kingston, Ontario, Canada Lakes and Water Management edited by V.K.E. Ilmavirta, Mai and Tor Nessling Foundation, Helsinki, Finland, R.I. Jones, Loughborougb University of Technology, Loughborougb, UK. , and E.P.E. Persson, University of Helsinki, Finland (Developments in Hydrobiology 7) ISBN 90-6193-7 58-2 March 1982 226 pp. $54.50 The Biology of Turbellaria edited by Ernest R. Schockaert and Ian R. Ball (Developments in Hydrobiology 6) ISBN 90-6193-757-4 December 1981 316 pp. $69.50 Ophthalmology Pathophysiology of the Visual System edited by L. Maffei, Director of the Institute of Neurophysiology, Pisa, Italy (Documenta Ophthalmologica Proceedings Series 30) ISBN 90-6193-726-4 November 1981 304 pp. Ultrasonography in Ophthalmology $53.00 edited by J.M. Thijssen, Institute of Ophthalmology, University of Nijmegen, the Netherlands, and A.M. Verbeek (Documenta Ophthalmologica Proceedings Series 29) ISBN 90-6193-724-8 January 1982 560 pp. $99.50 (Monographiae Biologicae 46) ISBN 90-6193-099-5 March 1982 160 pp. Please use ISBN when ordering these Dr. W Junk book s from your bookseller or direct from our $39.50 North American distribution center: M !9a~~~~L~~!?!J1~f· Please be seated to load, operate, view Dukane's new Mode/27 A66 Microform Reader is convenience engineered! Please be seated and enjoy these Model 27A66 benefits: quality optics and controlled illumination to give desired eye comfort ... highly visible instructions to make operation easy, even for the first time user ... film handling, loading and all controls conven- iently reached from the viewing position ... eye-level screen for comfortable viewing . The 27 A66 projects 35mm and 16mm roll film with either manual or motorized film drives. It also offers the option of projecting microfiche or aperture cards. AV-7047-1 Next to its tasteful styling and trouble- free performance, the economical price and minimal operating costs are appealing, too! Covered by Dukane's one-year warranty on parts and labor. UL approved. For added information , contact us or your local authorized Dukane micro- graphics dealer. DUKANED DUKANE CORPORATION/ AUDIO VISUAL DIVISION/ DEPT . CR3/ST. CHARLES , IL 60174