Bulletin, February/March 2006


Should Libraries Acquire Books That Are Widely Held Elsewhere?  A Brief Investigation with Implications for Consortial Book Selection

by William H. Walters

William H. Walters is assistant professor of librarianship at Millersville University. He previously served as collection development coordinator and acquisitions librarian at St. Lawrence University. He can be reached by email at william.walters<at>millersville.edu.

Traditionally, librarians have identified the most important books within particular subject areas by relying, at least partly, on lists of the titles most often held by other libraries. This is especially true where retrospective selection is concerned. For example, a selector attempting to fill in the gaps in his library’s collection of books on East Asian art might well ask, “What books are we missing? What do other libraries have that we don’t?” Of course, this approach is based on the idea that the most important books are often those that have met the selection criteria in place at many different institutions. There are at least three variants of this selection strategy: (1) counting only the holdings of the top specialist libraries – presumably, those with the most knowledgeable selectors and the most fully developed selection criteria; (2) counting only the holdings of comparable institutions (other private liberal arts colleges, for example) on the assumption that only comparable institutions are similar in their resources, needs and goals; and (3) counting the holdings of a wide variety of libraries (all the OCLC member libraries, for example) on the assumption that local idiosyncrasies in collection development policy will cancel each other out when combined into an aggregate indicator such as total library holdings.

            More recently, some librarians working within close-knit library consortia have adopted a distinctly different approach to book selection. This approach attempts to maximize the number of titles held within the consortium by minimizing the degree of overlap among the member libraries’ collections. If we assume that the consortium requires just one (or at most, a few) copies of any particular title, then the books most widely held within the consortium are those least appropriate for selection by other member libraries. For instance, participants in a cooperative resource sharing program might legitimately ask, “If libraries X and Y have this book, why does library Z need it as well?”

            Although both these approaches seem intuitively reasonable, they lead to essentially opposite outcomes. To some extent, the choice of either strategy depends on a conceptual distinction: Are the member libraries building a single consortial collection or a set of related collections? Moreover, both approaches may coexist within a single institution depending on the preferences of individual selectors and the activities they perform. The procedures and standards used in the evaluation of recently published books may be very different from those employed during retrospective selection, gift evaluation or the management of endowment funds.

            Only a handful of studies have addressed the relationship between aggregate library holdings and other indicators of importance, quality or appropriateness. Virtually all of them have focused on context-independent measures of quality such as inclusion in Choice magazine’s annual list of Outstanding Academic Books (now called Outstanding Academic Titles). For instance, Calhoun (1998) reports that Outstanding Academic Books are more widely held than most other current titles. At the same time, however, other evidence suggests that these same books are acquired by a relatively small and declining number of academic libraries (Budd & Craven, 1999; Sweetland & Christensen, 1997). Overall, we know little about the nature or magnitude of the relationship between library holdings and quality or the circumstances under which such a relationship can be found. (It is possible, for instance, that holdings and quality are linked only among certain kinds of libraries, only in certain subject areas or only for books that were published at a particular time.) We know even less about the relationship between library holdings and appropriateness for selection, since collection development policies can vary dramatically from one library to another.

            In the absence of more reliable evidence, the acquisitions staff at St. Lawrence University conducted a brief investigation of one particular question: Does the fact that several other ConnectNY libraries hold a current title suggest (a) that we should have it as well or (b) that we don’t need to get it? ConnectNY is a nine-library consortium that facilitates quick interlibrary loan (often within one or two working days) through a shared union catalog, a user-friendly ILL interface and rapid delivery mechanisms. At the time of our analysis, the consortium had seven members: Colgate University, Rensselaer Polytechnic Institute, Rochester Institute of Technology, Siena College, St. Lawrence University, Union College and Vassar College. (Bard College and West Point joined ConnectNY later in 2005.)

            The St. Lawrence University Libraries receive and process book requests from librarians, departmental faculty and other library patrons. Normally, the acquisitions staff searches our own Innopac catalog to see if requested titles are already held or on order at St. Lawrence. In October 2004 we began using the ConnectNY union catalog for this purpose. By using the union catalog, we were able to identify all requests for books already held (ordered or received) by two or more of the other ConnectNY libraries. Because we rely on a slow but rigorous process of soliciting book requests from departmental faculty, it is likely that the other members of the consortium would have placed their orders for a particular title before a request for that same title was processed by the acquisitions staff at St. Lawrence.

            As collection development librarian, I evaluated each requested title held by two or more of the other ConnectNY libraries. Specifically, I placed each title into category A (something St. Lawrence must get regardless of who else has it) or category B (something we’d consider not getting if we knew it were readily available through expedited interlibrary loan). Although information on the number of ConnectNY holding libraries was provided to me along with each book request, I attempted to assess the merits of each title without considering the number of copies already available within the consortium. Each assessment was based on the book’s bibliographic description, any other information provided with the request (on the Choice card, in the publisher’s catalog, etc.) and my own knowledge of the university’s collections, curricula, faculty and students. For about one-third of the titles, I checked our Innopac catalog to see what other books we already had by the same author, in the same series or on related topics. I looked for published book reviews in approximately 5% of cases and searched disciplinary indexes (to evaluate the author’s expertise) in approximately 2% of cases.

            From October 12, 2004, to April 27, 2005, the St. Lawrence University Libraries processed 2,232 book orders, excluding approval books, standing orders, blanket orders and gifts. Of that number, 623 (28%) had been ordered or received by two or more ConnectNY libraries other than St. Lawrence. Specifically, 254 titles were held by two other libraries, 200 by three other libraries, 109 by four other libraries, 42 by five other libraries and 18 by six other libraries.

            Among those titles held by two or more ConnectNY libraries, there is a moderate inverse relationship between the number of holding libraries and the proportion of books in category A (something St. Lawrence must get regardless of who else has it). (See Table 1.) We can conclude from these data that books held by two, three or four other ConnectNY libraries are more likely to be appropriate for St. Lawrence than those held by five or six other ConnectNY libraries. Although the reasons for this are not entirely clear, I did notice a general pattern when evaluating these titles. Books held by two, three or four other ConnectNY libraries are often noteworthy due to the author’s argument, opinion or approach. They are not necessarily controversial, but it is often easy to see how a particular title might add unique content not found elsewhere in the collection. A good example is Pay Without Performance, by Lucian Bebchuk and Jesse M. Fried, which asserts that corporate boards often fail to represent shareholders’ interests due to the absence of systematic mechanisms that would make them more directly accountable. Bebchuk and Fried’s book is not simply another critique of executive pay, but a work that fills a particular niche within its subject area – a work that is likely to add unique content even to a collection that already includes many books on this topic. The topic itself is popular but not overwhelmingly so and is likely to be represented more fully in some academic libraries than in others.

            In contrast, books held by five or six other ConnectNY libraries are often works on hot topics such as eating disorders, the war in Iraq or the death penalty. Many such titles are popular chiefly because of their subject matter rather than the author’s viewpoint or approach, and at least some of them seem unlikely to add anything unique within their subject areas. As I evaluated these works I often found myself thinking, “Yes, St. Lawrence needs a sufficient number of books on this topic, but we don’t need this particular book.” Books of this type also tend to cover topics that are popular among college students in general – students at Vassar College as well as those at Rensselaer Polytechnic Institute.

            I would not claim that these findings are valid for any institution other than St. Lawrence University. Nonetheless, these results suggest that for some current books, consortial interlibrary loan is an adequate substitute for ownership. In this particular case, approximately 8% of requested titles – 27% of those held by two or more ConnectNY libraries – were placed into category B (something we’d consider not getting if we knew it were readily available through expedited interlibrary loan). Although books held by five or six other ConnectNY libraries tend to be less appropriate for St. Lawrence than those held by two, three or four other ConnectNY libraries, the number of holding libraries does not itself provide an adequate basis for discriminating between titles in category A and those in category B.

            The fundamental question remains: For particular settings, subject areas or types of books, is the number of holding libraries a reliable indicator of importance, quality or appropriateness? Here we have considered only a single institution and one subjective indicator of appropriateness. This brief investigation does demonstrate, however, that individual libraries can generate useful information that addresses their needs without dramatically changing or disrupting their usual acquisitions procedures.

Acknowledgements

I am grateful for the advice and assistance of Renée Dominie, Bart Harloe and Esther Wilder.

For Further Reading

Budd, J.M. (1991). The utility of a recommended core list: An examination of Books for College Libraries, 3rd ed. Journal of Academic Librarianship, 17 (3), 140–144.

Budd, J.M., & Craven, C.K. (1999). Academic library monographic acquisitions: Selection of Choice’s Outstanding Academic Books. Library Collections, Acquisitions, and Technical Services, 23 (1), 15–26.

Calhoun, J.C. (1998). Gauging the reception of Choice reviews through online union catalog holdings. Library Resources and Technical Services, 42 (1), 21–43.

Calhoun, J.C. (2001). Reviews, holdings and presses and publishers in academic library book acquisitions. Library Resources and Technical Services, 45 (3), 127–177.

Hardesty, L., & Mack, C. (1994). Searching for the holy grail: A core collection for undergraduate libraries. Journal of Academic Librarianship, 19 (6), 362–371.

Harloe, B. (Ed.) (1994). Guide to cooperative collection development. Chicago: American Library Association.

Line, M.B. (1995). Access as a substitute for holdings: False ideal or costly reality? Interlending and Document Supply, 23 (2), 28–30.

Sweetland, J.H., & Christensen, P.G. (1997). Developing language and literature collections in academic libraries: A survey. Journal of Academic Librarianship, 23 (2), 119–125.

Wood, R.J. (1997). The axioms, barriers and components of cooperative collection development. In G.E. Gorman & Ruth H. Miller (Eds.), Collection management for the 21st century (pp. 221–248). Westport: Greenwood Press.

 

 

Table 1

Percentage of requested books that were placed into category A (something St. Lawrence must get regardless of who else has it) by number of ConnectNY holding libraries

73%

of the titles held by 2 to 6 other ConnectNY libraries were placed into category A.

72%

of the titles held by 2 other ConnectNY libraries were placed into category A.

77%

of the titles held by 3 other ConnectNY libraries were placed into category A.

76%

of the titles held by 4 other ConnectNY libraries were placed into category A.

67%

of the titles held by 5 other ConnectNY libraries were placed into category A.

50%

of the titles held by 6 other ConnectNY libraries were placed into category A.