College and Research Libraries Letters Library Growth Rates To the Editor: The article on growth rates of major aca- demic libraries by Steven Leach in the November 1976 issue of C&RL provides some interesting and useful information, but the conclusions are not very profound. The main generalization: As higher levels of collection size are ap- proached, it becomes increasingly likely that the rate of collection growth will be- gin to decelerate ( p.539) could be further generalized into a law, such as the bigger things become the more likely it is that their rate of growth will slow down or stated more concisely: Things cannot grow at an exponential rate indefinitely. Obviously this law would apply to the production of information too, which would eventually have its effect upon libraries, but as long as library collections grow at a rate faster than their buildings, librarians will have big problems regardless of Fre- mont Rider's exponentiality or Leach's Law of Deceleration.-R. Dean Galloway, Li- brary Director, California State College, Stanislaus. Response To the Editor: I will attest that Leach's Law of Decel- eration does have an alliterative ring, but that is as far as I will dare to go in its de- fense. I happen to agree with Mr. Galloway that my conclusions are hardly profound. As for myself, when I seek profundities, I go to G. B. Shaw. I would like to comment on Mr. Gallo- way's final remark. While we may acknowl- edge that "things cannot grow at an exponential rate indefinitely," we also are discomforted by that realization because it contradicts the widely held value that big- ger is inevitable (or better). Even I am not so provincial to suppose that that value is peculiar only to us Americans, but it must be admitted that by our actions we Americans-librarians included-con- sistently have demonstrated our belief in that value. For most university libraries the tradi- tional solution to the problem of collection growth has been to build a new library or library addition every ten to twenty years. That solution was consistent with the ''big- ger is inevitable" value and, at least until quite recently, probably represented the most economical and, perhaps, only solution to the problem. However, certain trends, e.g., declining student enrollments, lowered expectations about the personal · economic benefits to be derived from a college edu- cation, and higher priorities for resources being accorded to other social needs, may combine to force university libraries to find other solutions to the problem of collection growth. Fremont Rider predicted that eventually it would become impossible to build ever- larger libraries for their ever-expanding col- lections. I suggest that Rider's prediction may be realized-but as the consequence of contingencies far different from those he envisioned.-Steven Leach, Information Services Librarian, Technological Institute Library, Northwestern University, Evans- ton, Illinois. User Frustration To the Editor: The article in the January 1977 C&RL by Saracevic et al. is a useful study, corre- lating circulation and access, and this letter is not intended as a criticism of the research reported but rather a plea for more re- I 249 250 I College & Research Libraries • May 1977 search. In the article there is a brief discus- sion of in-house use, stating that "44 per- cent of books used . . . were through loan and 56 percent were used in-house. For every book borrowed, another was used in the library." There is no indication of how this ratio was obtained, but, most likely, since the ratio is similar to those of other studies, it represents books left on tables or return shelves and does not include books used and replaced on regular stack shelves by users. Unfortunately, because of the difficulty of obtaining accurate data, there have been no well-designed studies of the much larger in-house use represented by books used and replaced in spite of signs in some libraries asking that books not be replaced. Most ob- servers believe that, in open-shelf research libraries, total use is on the order of ten times circulation. Total use appears to be divided approximately 10 percent, circula- tion; 10 to 20 percent, books left on tables or marked return shelves (so-called "sweep- YOU WILL HAVE- "CONFIDENCE" In Our Complete Periodicals Service- All American and Foreign Titles Promptness is a Traditional part of McGre1or S.rvico • , . os well as: • EXPERIENCE • TRAINED PERSONNEL • FINANCIAL STABILITY • AMPLE FACILITIES • RESPONSIBLE MANAGEMENT An attractive IJrotlturo ;, availaiJio lot tlto aMint. OUR 44th YEAR MOUNT MORRIS, IlLINOIS 61054 ings"); and 70 to 80 percent, books taken from and returned to shelves by the user. Without considering this major factor in re- search library use, studies of user frustra- tion (or satisfaction) present only a partial picture. Those using libraries for research will generally agree that a considerable part of the use of materials on the shelves results from following chains of references while working in the stacks of a library, a proce- dure that is effective only when it can be carried out with very little interruption, i.e., without use of recall or interlibrary loan. As the authors of this article point out, "the option of recall is relatively un- used," as is the option of interlibrary loan, which represents only a small percentage of circulation and can never be more than a tiny fraction of total use. (Serendipity is a much-talked-about factor in in-house use but is probably considerably less impor- tant.) With reduced acquisition rates in many research libraries and plans for cooperation dependent on access through interlibrary loan, research studies of the extent of user frustration that will result are long overdue. It would appear from interlibrary loan and recall statistics, and allowing for better ac- cess from interlibrary loan through easily used computerized bibliographical and holdings data, that actual use of books that are available only at a distance, or are not easily accessible on the shelves of a re- search library, will probably be no more than 2 to 4 percent of the use they would have if directly accessible to the user. Thus, if the total use of a book on the shelves would be five times a year, if it is made accessible only through a delivery system, its probable use will, at most, be only once every five years. Put another way, if, over a number of years, one million volumes that are selected as pertinent and important to the fields of research of a uni- versity are either not acquired or are re- moved to another location, and even if the average use of these books would be only once per year (equal to one circulation every ten years), at l~ast 960,000 uses per year are lost ( 1,000,000 minus 4 percent of 1,000,000). It is indeed time for statistically sound research studies that will accurately mea- sure total library use and, not only all of the factors that result in user frustration, but also the disastrous effects on research in American universities of over-dependence on cooperation as a substitute for adequate research collections.-Melvin ]. Voigt, Co- Editor, Progress in Communication Sci- ences, La Jolla, California. Response To the Editor: We would like to make a few comments on Mr. Voigt's letter. The most important has to do with the basic direction of our research. Unlike earlier studies, which have indeed sought to correlate book availability with circulation, our study places "circula- tion" in perspective as one of four nearly coequal factors which generate user frustra- tion. These factors are acquisition policy, "circulation," library operations, and user skills. Our work shows, for example, that even "perfect circulation" would leave many users frustrated. 1 We must also point out that, operationally, the category of "circulation" was explicitly defined to in- clude all cases where the desired books were in the hands of other users, including those being used in-house as well as those checked out. Mr. Voigt raises an interesting question about the effect of the in-house use of books. The relative importance of this effect can be estimated from available data. The crucial concept is that of "total use." This phrase may be interpreted in two distinct ways: ( 1) the total number of use events and ( 2) the cumulative duration of use events. (It is clear that the second may be computed from the first if the average dura- tion of the use events is known.) It is the latter interpretation, and not the former, which is significant with regard to estimat- ing user frustration. Let us compare the "total use" associated with circulation to that associated with in- house use. It is known that, independent of loan policy, borrowed books remain out for about three-quarters of the stated maxi- mum loan period. 2 Hamburg has reported the results of a study of a library system in Letters I 251 which 5.2 million circulations were ob- served over a one-year period.3 If the loan period for the libraries investigated is, on the average, two weeks (a conservative estimate), the circulations resulted in 5.2 x 1.5 (:4 of two weeks) million book weeks of unavailability, or 546 million hours of unavailability (assuming seventy hours of operation per week). During that same period, 1.3 million hours of in-house use occurred. Consequently, the in-house use is the smaller effect by a factor of 420 (546/ 1.3). Our own unpublished studies at Sears Library indicate a ratio of about 300 to 1.4 This does not mean that in-house use is an unimportant service, but it underlines the fact that it contributes relatively little interference to other users. There was some discussion of the effect of networking on availability at the Pittsburgh conference. 5 We hope that librarians will adopt the analytic technique presented in our paper to evaluate the significance of the four principal causes of user frustration and to monitor changes produced by both internal management decisions and the growth of networking.-Paul B. Kantor, William M. Shaw, ]r., Tefko Saracevic, Case Western Reserve University, Cleveland, Ohio. REFERENCES I. T. Saracevic, W. M. Shaw, Jr., and P. B. Kanter, "Causes and Dynamics of User Frustration in an Academic Library," Col- lege & Research Libraries 38:7-18 (Jan. 1977). 2. W. M. Shaw, Jr., "Loan Period Distribution in Academic Libraries," Information Process- ing and Management 12: 157-59 ( 197 6). 3. Morris Hamburg, Leonard E. Rarnist, and Michael R. W. Bommer, "Library Objectives and Performance Measures and Their Use in Decision Making," in Don R. Swanson and Abraham Bookstein, eds., Operations Research: Implications for Libraries ( Chica- go: Univ. of Chicago Pr., 1972), see espe- cially p.117-18. 4. Unpublished research papers compiled by students in LS 509, "Library Performance Measures," at Case Western Reserve Uni- versity. 5. P. Kantor, "Comments at the Pittsburgh Conference on Resource Sharing, September 1976," forthcoming. 252 I College & Research Libraries • May 1977 Book Reviews in History To the Editor: I have just noticed the review of Index to Book Reviews in Historical Periodicals 1972, compiled by Brewster and McLeod, in the January 1977 issue of C&RL (p.53). It is stated that this work covers many "publications of state and local history so- cieties not indexed in existing book review reference works." Evidently you are not aware of our publication, America: History and Life, Part B, Index to Book Review>S. This specialized history tool, first published in 1975, is an index to reviews of books on U.S. and Canadian history, culture, and so- ciety from 113 journals ( 130 journals be- ginning in 1976). From the list of periodi- cals, you can see that we cover the state and local history journals. Comparing the AH L list of periodicals to the Brewster- McLeod 197 4 list, you will see that only 18,000 LIBRARIANS six titles-American Heritage, American ] ournal of Archaeology, American Antiqui- ty, East Texas Historical Journal, The Geor- gia Review, and The Indian Historian- not covered by AH L can be considered U.S. history journals (plus possibly parts of Current History, History and TheQTy, and Commentary). The AHL list of periodicals was devel- oped in consultation with numerous his- torians and librarians and was further re- fined in 1976 after two years of experience indexing the journals. Furthermore, you will notice that where- as we cover 113 (now 130) journals on U.S. and Canadian history alone, the Brew- ster-McLeod work indexes only some ninety journals for world history.-Yvonne Turner, Assistant Editor, America: History and Life, Part B, American Bibliographical Center-Clio Press, Santa Barbara, Califor- nia. KNOW... '~ ... there's more to ~ running a library than checking cf' out books and checking in periodicals ... There's the exhilaration of introducing a five-year-old to Dr. Seuss. Or locating the periodical that gives precisely the information the student is searching for to document his term paper. Running a library also has its problems. Like keeping track of hundreds of subscriptions to periodicals. There are better ways for you to spend your day. Faxon's automated subscription service will keep track of your subscriptions for you. You place one order and receive one invoice. Clear and accurate. There are so many exciting facets of library work. Now you'll have more time to discover them. Write for a copy of our Service Brochure and our annual Ll BRAR- IANS'GUIDE. Or call today toll-free. Then you will know too! Library business is our only business-since 1881 . ffiB F. W. FAXOn COmPAnY, InC. Library Magazine Subscription Agency 15 Southwest Park, .Westwood, Massachusetts 02090 Tel : 800-225-7894 (toll-free) • 617-329-3350 collect in Mass. and Canada only)