College and Research Libraries Research Notes CONTENTS Norman Desmarais ModMekkawi NORMAN DESMARAIS 393 396 COllEGE & RESEARCH liBRARIES Losses in a Theological Library The ARL Library Index as a Decision-Making Tool Losses in a Theological Library Librarians often assume that the greatest number of book losses occur in medical and legal libraries. Some librarians would add theological libraries to the list. A search of the literature has revealed that no adequate study of losses in theological libraries has been published, thereby making it impos- sible to draw valid conclusions. St. Mary's Seminary and University Library, Balti- more, Maryland, studied its loss rate, com- pared it with that of other libraries, and con- sidered the security measures that might be taken to reduce such losses. NEED FOR THE STUDY Prior to 1978, the library at St. Mary's had poor exit controls. Circulation often relied on the honor system as the desk was often unsu- pervised and members of the small faculty and student body were expected to sign out for the books themselves. A new library di- rector found this system unsatisfactory and began considering methods to increase secu- rity. Statistics were needed to substantiate adopting new methods and to verify their ef- Norman Desmarais is head of the Order and Cataloging Division, Catholic University of Amer- ica, Washington, D. C. , and former director of li- brary services, St. Mary 's Seminary and Univer- sity, Baltimore, Maryland. fectiveness. The library hired several stu- dents to assist library personnel in inventory of the entire collection during May and June of 1978. A year later, the library staff con- ducted a sample inventory of 11.28 percent of the collection to determine annual loss rate. FIRST STUDY Two-thirds of the collection was still clas- sified according to the Dewey decimal classi- fication (DDC); all acquisitions after 1971 and recataloged materials followed the Li- brary of Congress _ classification (LC). The director assumed that newer books were more likely to be used and therefore more likely to be lost or missing. 1 On this basis, he decided to inventory every fifth shelf of the LC collection and every eighth shelf of the DDC collection (which was denser than the LC collection). The reference collection was inventoried completely. Of a total of 2,588 shelves with more than five books on them, a sample of 292 (11.28 percent) was inventoried-152 DDC and 140 LC. Book losses amounted to a total of sixty vol- umes, or .69 percent of the collection, while periodical losses amounted to twenty-three items, or .17 percent of that collection. In ad- dition to this partial inventory, the staff con- ducted a book census (counting every book and bound periodical) to determine the exact size of the collection, and to serve as a basis I 393 394 I College & Research Libraries • September 1982 for the efficient establishment of loss rate in the future. The library adopted the following security measures: (1) the hiring of extra personnel for more consistent staffing of the circulation desk; (2) the elimination of the honor system; and (3) general tightening of some of the lax procedures. SECOND STUDY The same sampling procedure was fol- lowed in May 1980. The staff inventoried 12.96 percent of the LC collection (27,537 volumes) and 12.5 percent of the DDC col- lection (45,944 volumes) and the entire refer- ence collection (1,966 volumes). This re- vealed that the heaviest losses occurred in the LC collection (1.09 percent of the sample, or thirty-nine volumes). Next came the refer- ence collection ( .457 percent of the sample, or nine items), followed by the DDC collec- tion (.24 percent, or fourteen volumes). During 1980-81, the library staff expected to merge part of the undergraduate liberal arts library from another campus with the theology library. They anticipated reorgan- izing the library and having better exit con- trols; but this reorganization came too late in the year to allow them to draw valid conclu- sions. As long delays in the merger of the shelflists were expected, the staff had to de- vise an alternative method of data gathering. THIRD STUDY It was assumed that, by taking the base figures for the size of the collection (gathered through the book census) , adding the yearly acquisitions, and subtracting the with- drawals , the librarians could determine book losses more quickly and economically on a yearly basis. This would also eliminate skew- ing due to shelving errors. Such a method would not account for human errors in counting, however. This procedure indi- cated a discrepancy of 928 items or . 935 per- cent of the collection, which totaled 99,245 volumes at this point. At a cost of $22 (the li- brary's average cost per book for 1980-81), this would yield a total loss of $20,416. COMPARISONS St. Mary's loss rate compares favorably with studies done at other libraries. C. W. Post Center Library of Long Island Univer- sity (a large library of 460,000 volumes) ex- perienced an average annual loss rate of 1.35 percent prior to installing a book detection system, which lowered the rate to .38 per- cent. 2 Bristol Community College's Learning Resource Center Library (Fall River, Mass.), with a collection of approximately 43,000 volumes, saw its losses climb from .3 percent in 1973 to 1.3 percent in 1974 to l. 7 percent in 1975. 3 A Carnegie study found that the un- dergraduate libraries at the University of California at Berkeley, Northwestern Uni- versity, and the University of Washington re- port annual losses of 4 to 5 percent. 4 "Tufts University found that almost eight percent of the books in its libraries disappear after just one year on the shelves .... A 1976 inven- tory at Claremont Colleges (California) pin- pointed losses in the past 20 years at 15,000 books. And an inventory at the University of Maryland found losses of more than 30,000 volumes. "5 J. W. Griffith reports that the Lewis Cen- tral High School in Council Bluffs, Iowa, ex- perienced annual losses of 3 percent of there- ligion collection and 3. 06 percent of the philosophy collection, with the highest rate occurring in the applied sciences (6.48 per- cent). 6 Public libraries seem to fare the worst. The New York Public Library pegs its annual losses at about 10 percent of the collection. 7 The situation in Britain is much the same, where " the rate of loss reported in the Li- brary Association College Library Survey showed that the actual annual losses from open shelves was 2.6 % . A national average frequently quoted is 1 o/o per annum. "8 CoNCLUSIONS AND RECOMMENDATIONS Ungarelli's study reemphasizes that "there is a correlation between loss rate and publi- cation date, and that there is also a correla- tion between loss rate and the use of materi- als. " 9 We may infer from this that the loss rate of a given library does not depend so much on the type of library as it does on the expectations of its users, and the use they make of its collections. Libraries noted for having good theology collections will experi- ence more use (and loss) of those collections, while libraries with strong literature collec- tions may expect corresponding use and loss in their literature collections. As the student body increases and as circu- lation and use continue to increase, aca- demic libraries can expect that losses will in- crease proportionally. In 1979, when St. Mary's losses totaled .69 percent of the collec- tion, circulation had increased 5.6 percent over the previous year. The following year, the loss rate dropped to . 55 percent of the col- lection, even though circulation increased by 81 percent. This decrease may be partly ac- counted for by the great amount of publicity and consciousness-raising devoted to the problem. The 1981 loss rate jumped to .93 percent of the collection, while circulation increased by 33.5 percent. It appears, then, that the security measures already imple- mented have done little to reduce losses. Even though the percentage of losses at St. Mary's is small compared to other studies, it is still unacceptably high considering that most volumes are irreplaceable. Even a 1 percent annual loss from a large collection represents a large number of volumes. As a theology library whose primary pur- pose is to support the education of future priests and ministers, St. Mary's has been re- luctant to propose installation of a book de- tection system, partly because of the initial costs, and partly because of the negative im- age this might project to its users and sup- porters. However, as other security measures do not seem very effective, as libraries in- creasingly become targets for casual and pro- fessional thieves, 10 and as losses and replace- ment costs continue to increase, th~ time may be ripe to consider the installation of such a . system. Researchers have assumed that, in · order to be considered adequate, the effec- tiveness level of book detection systems should be around 85 percent, and the results of actual studies have been higher than this. Most libraries find that the system begins to pay for itself in the second or third year of its operation. By preserving their collections, li- braries are able to spend more funds for new acquisitions, rather than replacements, and are thereby better able to satisfy their users by providing the books they need. REFERENCES 1. Herman H. Fussier and Julian L. Simon, Pat- terns in the Use of Books in Large Research Li- braries (Chicago: Univ. of Chicago Pr., 1969), p.15. 2. Donald L. Ungarelli, "The Cost-Benefit of a Book Detection System: A Comparative Research Notes I 395 Study," in Ching-chih Chen, ed., Quantita- tive Measurement and Dynamic Library Ser- vice (Phoenix, Ariz.: Oryx Press, 1978), p.153-54. 3. Harvey Varnet, "An Analysis of the Effective- ness and Cost-Benefit of the Checkpoint Mark II Book Detection System Installed in the Li- brary Section of Bristol Community College's Learning Resources Center," in Ching-chih Chen, ed., Quantitative Measurement and Dynamic Library Service (Phoenix, Ariz. Oryx Press, 1978), p.161. 4. "Carnegie Study Finds Theft Rising in College Libraries," Library ]ourna/104:1206-1 Qune 1979). 5. Ibid., p.1207. 6. J. W. Griffith, "Library Thefts: A Problem That Won't Go Away," American Libraries 96:226 (April1978). 7. "Library Security: Book Thefts Are Up," Li- brary ]ourna/100:352 (Feb. 15, 1975). 8. D. H. Revill, "Theft Problem in Libraries," New Library World 76:123-24 (June 1975). 9. Ungarelli, "The Cost-Benefit of a Book Detec- tion System," p.155. 10. Beverly T. Watkins, "For Campus Libraries and Museums, Thievery Is a Growing Prob- lem," The Chronicle of Higher Education 23:8 (Oct. 21, 1981). ~MCGREGOR~ "PERSONALIZED SUBSCRIPTION SERVICE" Every customer is assigned an experienced "Home Office" representative. You correspond direct; any title needs, changes, cancellations or problems can be handled promptly by lettlr or phone. This makes your job easier and keeps you abreast of your subscription needs at all times. With over 45 years expe~ce. McGregor has built a reputation of prompt and courteous service on both domestic and International titles. We prepay subscrip- tions ahead of time. Our customers, large and small, like the prompt attention we give them. We think you would tool Ask about McGregor's "Automatic Renewal" plan de- scribed In our new brochure. Write today for your free copy. OUR 49th YEAR Mount Morris, llllnolsl1054