College and Research Libraries Letters Cost-Benefit Analysis To the Editor: The article by Bommer and Ford (C&RL, July 1974) on a cost-benefit analy- sis for determining the value of an electron- ic security system, contains the statement: "Of the 55 documents not accounted for in December 1971, 7 have been accounted for after an exhaustive one-year search. It is doubtful that more of these documents will be accounted for in the future. Thus we es- timate that approximately 7 I 55 or 13 per- cent of the documents estimated to be lost will be accounted for in the future" (p. 273). Thus the assumption is made that 87 per- cent of the documents found missing will not be returned. If this assumption is not true, then the basis of the cost-benefit anal- ysis models which are suggested are likely to be unsound. The policy at the University of Bradford is to carry out a 10 percent stock check ev- ery year. In order to establish how long it is necessary to continue searching for miss- ing material before it is safe to amend the catalog, an annual recheck has been carried out each year at the University's Social Sci- ences and Management Centre Libraries. The figures are given in the table below for three and two years respectively: Total missing % still missing after 1 year % still missing after 2 years % still missing after Soc. Man. Sci. Lib'y Cent. Lib'y 251 312 71.3 76.9 59.8 66.3 3 years 55.4 It can be seen that books continue to be returned in significant numbers two and three years after they are first reported missing. It would appear, therefore, that it would COLLEGE & RESEARCH LIBRARIES be dangerous to use the cost-benefit analy- sis model suggested by Bommer and Ford until more exhaustive tests have been carried out to establish the proportion of documents which are lost forever, as op- posed to those which are returned to the library after periods of up to three years and even longer. Response F. H. Ayres Deputy Librarian University of Bradford Bradford 7, Yorkshire England The authors of this article recognize nothing sacrosanct about the figure 87 per- cent as an estimate of the number of docu- ments which will never be accounted for of those judged to be lost. Obviously dif- ferent libraries employing different search methodologies will need to develop proce- dures particular to their own situation for estimating this figure. It is our belief that some missing docu- ments at the University of Pennsylvania are recovered, not so much as a result of being returned by "borrowers," but rather as a re- sult of being found within the library. In addition, we are convinced that the search procedures employed in this study were so thorough (not merely a stock check) both in the initial search to verify the lost con- dition of the initial group of 55 documents and in the subsequent year-long search which recovered 7 of these documents, that it would be most unlikely that additional documents of this group would be recov- ered in the future. However, the major point to be made in reply to Mr. Ayres' comment is that ques- tioning the accuracy of a particular data input for a model does not discredit the va- lidity or usefulness of the entire cost-bene- fit analysis model. Although a different es- timating procedure might be need~d to ob- I 153 154 I College & Research Libraries • March 1975 tain various data inputs for different librar- ies, the credibility of the basic model has not been affected. Michael Bommer Clarkson College Potsdam, New York and Bernard Ford University of Pennsylvania Libraries Philadelphia Library Decision Making To the Editor: Jeffrey Raffel's article in the November issue of C&RL, "From Economic to Polit- ical Analysis of Library ·Decision Making," contained many interesting concepts, but I strongly object to his statement on page 417 that " ... a high-level [MIT], library acquisitions department staff member had not only made no effort to buy books from the Harvard Coop but also had never even been to this store .... " I cannot imagine how Mr. Raffel obtained this startling bit of misinformation, but to set the record straight, the "Coop" is, and has been for a number of years, our fifth largest vendor. For years it has been our practice to make daily shopping trips to the Tech Coop (the MIT branch of the Harvard Coop) to fill orders by picking books off the shelves. Furthermore, since 1966 an average of 2, 000 books have been purchased from the Coop each year for the MIT Student Cen- ter Library alone. In 1968 three different experiments were undertaken by the MIT Libraries acquisi- tions department in an effort to determine the best way to utilize the resources of the Harvard and Tech Coops to fill our daily orders for current American publications. One experiment, for example, of personal shopping tours to the Harvard branch of the Harvard Coop was abandoned as too expensive when it was determined that it took an average of three hours daily for one clerk to fill only 12.4 percent of the orders. Robert L. Hadlock Head, Acquisitions Department The Libraries Massachusetts Institute of Technology Cambridge Response The "startling bit of misinformation" which has upset Mr. Robert L. Hadlock, head, Acquisitions Department, M.I.T. Li- braries, grew out of an interview I had with an Acquisitions Department staff member in the summer of 1967. The subject was the Harvard Coop, not the Tech Coop branch, w·hich had a far more complete inventory in the nonphysical science areas. Thus the Acquisition Department's use of the Tech Coop, as described in detail by Mr. Had- lock, is irrelevant to the statement quoted in my paper. I am glad that three experiments took place in 1968 to investigate the "best way to utilize the Harvard and Tech Coops. ... " I would like to think that my original conversation or research encouraged this inquiry. I am afraid, however, that Mr. Hadlock's letter misses the key point of my article. His letter leaves unstated the value choices associated with the conclusion that "per- sonal shopping tours to the Harvard branch . . . [were] . . . too expensive." If it costs $35 per book purchase at the Harvard Co- op, including personnel time, to get a book to serve as required reading for an under- graduate seminar, is it too expensive a method? What if the book is for a senior faculty member's research project? Who de- cides which means are too expensive? These are the kinds of questions which I think should be addressed. I should take care not to miss Mr. Had- lock's major concern. My article was not in- tended to criticize any individual or depart- ment associated with any library. My re- search was undertaken almost a decade ago and times, libraries, and people have changed. In the context of the article I re- ported an interview I had had years ago to illustrate a series of more general points which are as appropriate now as they were in 1967. I effrey A. Raffel Assistant Professor Division of Urban Affairs and Political Science University of Delaware Newark As your budget gets tighter you need the best advice you can find. You just found it. Consumers Index and Media Review Digest provide the advice you need, compiling evaluative information on media software and equipment of value to libraries, schools, and consumers. Each is the most significant reference and selection tool in its field. Purchasing based on advice from these works could save many times their modest cost. * Digests the contents of articles from over 100 consumer interest and general information sources. * Indexes and codes all product tests and evaluations from these sources by brand name. * Digests the contents of new books, pamphlets and other publications related to consumerism. * Covers cars, cameras, stereo equipment, tape recorders, boats, camping trailers, insurance, fur-- nilhings, and equipment for business, educational and library use, personal investments, health care and much more. * Designed for use by the general consumer, edu- cational institutions and business offices. * Published quarterly and cumulated annually: Quar-- terly subscription, $25.00; Annual cumulation, $25.00; Combined subscription, $45.00 * 10% discount to libraries ordering 10 or more subscriptions direct from Pierian Press. Foreign •. ''"'"'""'"\ :•~ to Produl"l haluation' and lnformatiun Sc~t~rcc' postage extra. ---------~ Is the most comprehensive and current selection, acquisitions, cata- loging and reference tool in the field of non-book media. Covers both educational and feature films, film strips, records and tapes (popular, classical and spoken-word), and miscellaneous media including slides, transparencies, overlays, games, kits, etc. Catalogs all educational items, providing: descriptions of subject content, Library of Congress headings, Dewey decimal numbers, grade level indications, and award citations. Includes an estimated 50-60,000 review citations per year, and di- gests many thousands of critical and evaluative reviews from major reviewing sources. • Both an alphabetical Library of Congress and a classified Dewey decimal subject approach are provided for all educational media. • Annual cumulations are updated by quarterly supplements, all of which include subject indexes. MRD 1973n4, $65 plus postage. MRD 1974n5, $65 plus postage. OTHER SUBSCRIPTION PACKAGES AVAILABLE'. Serials Review and Reference Services Review provide similar advice on serials and reference materials. 30--day examination privileges. Send for complete information. • • • • an exc1t1ng • new serv1ce package The computerized MARC- based book cataloging and processing service system that wraps up everything you need from a book order s-ervice into one neat package BRO-DAR T, INC. 1609 MEMORIAL AVENUE · WILLIAMSPORT . PA 17701 1236 SOUTH HATCHER AVENUE · CITY OF INDUSTRY . CA 91749 BRO -DART 6 EDMONDSON STREET · BRANTFORD . ONTARIO N3T 5M3