Studies indicate that a lean reference collection is the ideal, but how does a librarian determine what to pare? A small academic library did a five-year reshelving study to guide in collection management. Dots were applied to books as they were reshelved, with different colors for each year. Data indicate that, while many items were heavily used, many others were not used at all in five years. As a result of the study, reference staff are reconsidering the nature of the reference collection, beginning to develop a collection management policy, and determining the disposition of the good, but unused, items.
The library that does a use study is attempting to know which items it currently holds will be used next year, and we know from Fussler and Simon’s study that “the single best predictor of future use of a book” is past use.1 If the reference staff at Columbia International University (CIU) could learn which books are used, we could then create a lean, efficient reference collection by weeding unused books and utilizing the data to select new acquisitions by established use patterns. This study is an investigation into the use of a reference collection at a small academic library focused on what has been used, with an eye to weeding the unused collection of items.
The method used was suggested by Eugene Engeldinger’s challenge to make known use the primary criterion for weeding reference collections in his report of a use study done at the McIntyre Library, University of Wisconsin, Eau Claire (UWEC).2 The Engeldinger study involved placing one adhesive dot inside the back cover of a reference book each time it was reshelved, up to five times. At the end of five years, the dots were counted, providing the reference staff with quantitative information to supplement instinct as they made weeding decisions. The Eau Claire study was one that the reference staff at CIU’s G. Allen Fleece Library could easily replicate.
According to Robert Broadus, use studies are differentiated from usage studies by the study’s object. Use studies focus on the materials and how they are used. Usage studies focus on the patrons and how they use the materials.3 Use studies, of course, do not–or should not–exist in a vacuum. The professional literature indicates that the most common reason librarians performed use studies was to support collection management decisions. Use studies rightly inform these activities, but they also provide quantitative information to guide the deselecting, pruning, and weeding process. Therefore, literature supporting this research centers on use studies and weeding as each intersects the print reference collection.
Broadus examined conclusions of the library use studies he considered most consequential.4 He offered an annotated list of five generalizations that might be drawn from these studies. Even though the Broadus article does not address reference collection use, two of his generalizations are relevant to this study. Broadus’s first generalization suggested that a substantial percentage of many libraries’ holdings receive no recorded use. The second relevant generalization states that current use is a predictor of future use; that is, items recently used are more likely to be used in the near and distant future.
In 1981, librarians at the San Luis Obispo campus at California Polytechnic State University conducted a title-by-title review of their reference collection.5 The review had nine objectives, which included providing an inventory and serving “as a means of purging the reference collection of seldom-used or obsolete books… .”6 The results of this study were, perhaps, more dramatic than most librarians would prefer: “… the reference collection was reducible from 16,000 titles and 25,000 volumes to a manageable size of 5,540 titles and 14,331 volumes. Furthermore, librarians became more familiar with the scope and depth of the entire collection, as well as in their own subject areas, and with the location of missing items.”7
In 1982, Engeldinger surveyed 377 academic reference librarians about various aspects of their libraries’ practices relating to weeding the reference collection.8 He found that frequency of material use was a factor for 54 percent of the librarians responding. He also sought to elicit information on how frequency of use was determined. Responses were few, but virtually all who did reply offered only subjective means to assess use.
Biggs and Biggs surveyed 471 reference heads in academic libraries regarding collection development.9 Reference collection size was a focus “especially because our perception–which turned out to be confirmed by our findings regarding numbers and frequency of use of reference volumes–that reference collections tend to be too large for thorough exploitation by librarians in the service of information delivery.”10
Librarians at Iowa State University’s Parks Library considered the types of reference materials most necessary to meet patron information needs, as well as reference staffing and training needed.11 For four weeks, the librarians tallied usage of reference books, differentiating between usage by reference staff and library patrons. Results were further distinguished by Library of Congress (LC) call number ranges. Usage was defined as “referral to, or reshelving of, a particular reference item,” despite the authors’ awareness that this method overlooked those items reshelved by the patron and undercounted those used by more than one patron but reshelved by library staff only once. The study informed the library staff about usage of books within each subject (LC classification), which, in turn, informed them regarding potential collection weaknesses.