Library Issues
Editor: Richard M. Dougherty, University of Michigan.
Contributing Editors: Mignon Adams, University of the Sciences in Philadelphia; Eileen Fenton, University of Michigan; Steve Marquardt, South Dakota State University; William Miller, Florida Atlantic University; Maureen Pastine, Temple University; Sarah Pritchard, Smith College;
Vol. 19, No. 2November 1998

Weeding the collection
Academic Libraries and Information Technology

Weeding the Collection—Painful But Necessary

by Evan Farber

The word "weeding" rarely invokes a positive response, but when used in conjunction with library books it has such a negative connotation that one even hesitates to mention it. Substitute terms for weeding a book collection, such as de-selection, de-acquisitions, culling, or even stock control, have been suggested, but none has caught on. Indeed, a periodical, the De-Acquisitions Librarian, was started in 1976 in response to the problem of overcrowded library shelves; despite librarians' recognition of the situation it was meant to address, its title attracted so few subscribers that within its first year it became Collection Management, a term hardly suggesting that libraries ought to get rid of some books. ("Get rid of" may be misleading; some libraries have chosen to keep the weeded books, but in a separate storage facility. That option is discussed below.) As irksome a topic as it may be, weeding a library's book collection is still an important topic for consideration.

Administrative and Educational Factors

If it isn't the most important, certainly the most persuasive (and pressing) reason why college and university librarians and administrators talk about undertaking a weeding project is to avoid the costs of expanding a library building or installing expensive compact storage systems. Some librarians and many administrators once thought that, with the increasing use of electronic sources of information, libraries would need less operating space. That proved to be wishful thinking as libraries are having to accommodate both book collections and the new technology. When one considers that every hundred volumes removed from a collection can recover about forty square feet of building space, and that building costs today can be as much as several hundred dollars per square foot, the economic argument for weeding a crowded collection makes a lot of sense.

But even if there is no immediate danger of running out of space, there are other reasons for weeding a collection. From an educational perspective, a well-chosen, uncrowded collection is most desirable, particularly for undergraduates. The primary purpose of open stacks, after all, is to permit, even encourage students to browse. A tightly packed collection, with many titles on the top or bottom shelves, makes it difficult for students not only to browse, but even to find materials they've already identified since the chances of books being misshelved are much greater. A student encountering either of those difficulties will probably take the course of least resistance and settle for titles that are handier but may not be as educationally valid.

An even stronger educational reason is the danger of misinformation. For many students, the mere fact that their library has a particular title in its collection means that the book is a valid work of scholarship. They don't recognize that some books may be in an academic library not because of their validity, but rather because they are good examples of spurious research—books, for example, that try to "prove" that Shakespeare never wrote those plays or that the Holocaust never happened. While those books make up a very small portion of any collection, that same collection undoubtedly contains many more titles that were once considered useful or even significant, but which have become obsolete or have been superseded by later scholarship. One has only to compare editions of Books for College Libraries, a listing of books recommended for undergraduate collections. BCL was published by the American Library Association for the Association of College and Research Libraries, and the recommended titles were selected by groups of subject specialists representing many institutions. The first edition was published in 1967; that of course was the era of prosperity for undergraduate education, when colleges were being founded or expanding—and so were their libraries. Because of BCL's academic respectability as well as its ease of use, many libraries—venerable as well as newer ones—used it as a convenient guide for checking and/or building their book collections. The usefulness of BCL was so evident that a second edition was published in 1975, and a third in 1988. All editions were highly selective, but less than a third of the 53,400 titles listed in the first edition were also listed in the third. Yet there's no question that in many of the undergraduate libraries that used that first edition as a buying guide, many, if not most of the titles omitted from the later edition are still on the shelves. But how can students know that?

Few undergraduates have developed the skills, the intellectual filters that comprise a typical scholar's ability to quickly, almost automatically discriminate among sources, to separate the authoritative from the spurious, to recognize what is valid and what is mendacious or just obsolete. Consequently, an undergraduate library that contains a significant number of outdated or superseded titles is probably doing its students an educational disservice.

Finally, from a collection super-visor's perspective, overcrowded shelves mean more wear and tear on the books, more staff time spent reading shelves and searching for misshelved titles, and from time to time even having to shift portions of the collection to make room for new acquisitions. One librarian has written that "In the long term, the cost of weeding is not as high as the effective costs of retention and preservation of rarely used volumes."1

Those reasons—economic, educational, and administrative—would seem to be persuasive. Then why the reluctance?

Obstacles To Consideration

First, weeding books from a collection is a negative concept, alien to librarians' ways of thinking. Academic librarians, after all, have been trained to acquire materials, to preserve and service them, and to get them in the hands of students and faculty who can use them. Getting rid of them is inherently contrary to librarians' professional training and instincts. And, of course, the decision to rid a book from the collection is a relatively final decision, and that really gives one pause.

Second, reinforcing the conceptual reluctance is the fact that the process of weeding is a prosaic and tedious one requiring much care and a good bit of staff time. With so many more positive demands on librarians' time, weeding assumes a low place on the administrative list of priorities, and is ignored, postponed, or, if done at all, done on a piecemeal basis.

Third, weeding a particular library's collection means that the size of that collection is going to be reduced, and that may not look so good when comparing institutional statistics. Of course, any knowledgeable reader of such statistics knows that the correlation between quantity and quality is not always direct nor even highly positive. Unfortunately, it is easier to cite numbers, and even some accrediting standards are based on quantity rather than quality.

Finally, the process of weeding a collection can involve serious risks to a librarian's reputation, and even when those risks are avoided, little public recognition is gained at the end of the process. Why, then, would a library director choose to go through this risky process when there are so many other things to be done, so many activities he or she can undertake to improve the library's position on campus?

"Risky?" Yes, risky, and that may be the major factor in the reluctance to weed. On almost any college campus the reputation, and thus the economic support of the library is very much dependent on faculty approval of how that library is administered. If the faculty feels that the library has been at all arbitrary in its weeding process, or if individual faculty members discover that titles were weeded which they were especially fond of or felt were still important, their approval rating of the library is going to be greatly diminished. Assuming that an institution—that is, its top administration and library staff—recognizes the need for weeding the library's collection, what steps should then be taken to assure that the procedure is effective, is done as efficiently as possible, and takes into consideration possible faculty criticism?

A Secondary Storage Facility?

The first step is to decide whether the titles to be weeded are going to be discarded or relegated to a secondary storage facility, usually on or near the campus. There are several points in favor of a secondary storage facility. The first is that a small proportion of the library's collection accounts for a very high proportion of its circulation. One undergraduate library reported some years ago "that over 96 percent of the circulation [for that year] was from a population of books that had been loaned at least once during the last eleven years" and that "approximately 25 percent of the main library's book collection of 250,000 volumes could be categorized as 'little-used'" on the basis of the previous decade's use records.2 With those figures in mind, that library decided to move 50,000 volumes to an on-campus storage facility. That library's experience is hardly unique; numerous studies have shown that a relatively small percentage (usually 15 to 20 percent) of a library's collection accounts for a high proportion (usually about 80 percent) of its circulation.

The second argument in favor of a storage facility is that though there will surely be minor complaints about delays in retrieving volumes, the more serious complaint about individual titles being discarded is obviated. The main negative point is that a storage facility means a considerable expense: the costs of building or acquiring it and then maintaining it (temperature and humidity need to be controlled); the costs of changing and keeping records, both those for staff and users within the library and those accessible to users outside the library; additional staffing will also be needed for retrieving and reshelving items kept in the storage facility. In recent years a number of libraries have gone the secondary storage route, so it's now possible to base one's decision on the literature describing others' experiences.

Selecting Titles

Probably the next step is deciding on the method of identifying which books are to be weeded. There are very few obvious categories that can be disposed of, such as duplicate copies of titles that were once on reserve; on the other hand, because individual opinions as to the usefulness of particular titles and thus as to which should be kept or weeded vary so much, librarians have tried to find methods based on objective criteria. Most of the methods depend on the use pattern of individual volumes—that is, the last time a volume circulated, or the time between uses, or the number of times it circulated over a given period, or a combination of these. Probably the best discussion of the various methods is Stanley J. Slote's Weeding Library Collections: Library Weeding Methods (Englewood, CO: Libraries Unlimited, Inc., 4th ed., 1997). Slote strongly recommends using the "shelf-time period"—the time between a book's uses—as the main criterion, though the determination of that period depends on the availability of circulation records.

For many academic libraries, especially those in undergraduate institutions, one suggested method is comparing the editions of Books for College Libraries. A comparison of the titles in the three editions will provide a very basic listing of titles that are candidates for weeding. The premise of that suggestion is that titles which were selected for the first edition but omitted from later editions were dropped for a reason.3 To make the process of comparing editions even easier, an electronic listing of the titles dropped or retained is being developed. That listing can then be run against a tape of the library's holdings to compile the candidates for weeding.

Final Caveat
That these initial choices are "candidates" cannot be stressed too strongly. Whatever method is used for determining the titles that should be weeded, the initial determination is just that—an initial determination—and the choices should only be regarded as candidates for discard or, if the option has been a secondary storage facility, candidates for relegation to that facility. Before a final decision for removal from the main collection is made, the teaching faculty must be given the opportunity to review the titles. No matter how seldom a particular title has been used, or how outdated it may be, or how flawed it may seem, there may be a reason an individual faculty member will want to retain it. Even where a secondary storage facility is being provided, so that no volume will actually be discarded, individual faculty members will still want to choose where the titles will be kept. Providing faculty members with that opportunity should blunt any potential criticism of the process.

While every commentator on weeding the collection of an academic library has noted the importance of the process, even the obligation to undertake it, they have also been aware of the obstacles, difficulties and risks. Careful planning, however, can overcome the difficulties and avoid the risks. The result, a more appropriate and accessible collection, is well worth the effort.—Evan Farber is College Librarian Emeritus, Earlham College, Richmond, IN.


1Terrance J. Metz, "An Examination of Weeding Methods for Academic Libraries Based on Recorded Use of Library Materials With an Application on a Small Liberal Arts College Collection." Unpublished paper, Graduate School of Library Science, University of Minnesota, May, 1985, p. 49
2Bart Harloe, "The Politics of Weeding: New Myths and Old Realities," Academic Libraries: Myths and Realities; Proceedings of the Third National Conference, of the Association of College and Research Libraries (Chicago: Association of College and Research Libraries, 1984), p. 263, 264
3Evan Farber, "Books NOT for College Libraries," August, 1997, Library Journal, pg 44-45.

Academic Libraries and
Information Technology

by William Miller

Academic librarians today frequently hear three questions:

These are well-intentioned questions and deserve a serious response.

Most of the printed material which libraries own today will never be digitized. It simply would not be economically feasible to do the labor involved in such digitization. Most libraries of any size could not even inventory their collections physically, let alone digitize them page by page. And even if such libraries had the staffing, space, equipment, and other resources necessary to do such digitizing, they would not have the legal right to do so, given the copyright laws of the United States and other countries. Distinguishing what is copyrighted from what is not is difficult, and doing the legwork of obtaining permissions is a nightmare. As a result, the digitizing projects which are currently under way, at the Library of Congress and elsewhere, are dealing with clearly non-copyrighted or out-of-copyright, hard-to-obtain materials. These projects are modest, yet even so are costing millions of dollars each, underwritten by foundations and the government.

"To the casual observer, there appears to be an enormous amount of information available without charge on the web. On closer inspection, however, the vast majority of this material is not useful for academic purposes, and that which is useful tends to be expensive."

Most new information, of the kind that scholars need, is still not available in digital form. Current estimates are that approximately 15 percent of such information is available electronically. Nor is it by any means certain that today's electronic information will still be accessible to scholars in the future. For instance, the tapes of the 1970 U.S. Census are now unreadable by any current technology, and only the printed volumes are accessible. Print and microform are still the only proven, long-term storage technologies. Libraries acquire and preserve for the future as well as for the present.

To the casual observer, there appears to be an enormous amount of information available without charge on the web. On closer inspection, however, the vast majority of this material is not useful for academic purposes, and that which is useful tends to be expensive. The profit motive and the economics of publishing remain in force in the electronic age. If publishers cannot make a profit out of generating reference books, science journals, or novels, they will simply cease to publish these materials.

As more and more information becomes available only in electronic format, libraries remain the agencies which pay for the materials (the costs of which are increasingly based on FTE) and make them available to the academic community. Moreover, libraries are the only agency in which staff spend a lot of time instructing students in how to find and use both traditional and electronic information, and how to distinguish that which is trustworthy from the sea of flotsam which is not.

The short answers to the questions posed above, therefore, are:


Mountainside Publishing, Inc. Library Issues: Briefings for Faculty and Administrators (ISSN 0734-3035) is published bimonthly beginning September 1980 by Mountainside Publishing Co., Inc., 321 S. Main St., #213, Ann Arbor, MI 48104; (734) 662-3925. Library Issues, Vol. 19, no. 2. ©1998 by Mountainside Publishing Co., Inc. Subscriptions: $75/one year; $140/two years. Additional subscriptions to same address $25 each/year. Address all correspondence to Library Issues, P.O. Box 8330, Ann Arbor, MI 48107. (Fax: 734-662-4450; E-mail: Subscribers have permission to photocopy articles free of charge for distribution on their own campus. Library Issues is available online with a password at
last modified: