Rachel Applegate
Print version (Adobe Reader required)
This study examines reference transaction quantities reported through the Academic Library Survey of the National Center for Educational Statistics to explore whether, and the extent to which, academic libraries are seeing a decline: the beginnings of a “deserted library.” Data from 2002 and 2004 shows a decline in reference transactions per week on a per-library basis and on a per-student basis, but this decline differs by the type of institution (Carnegie Class). Librarians at master’s institutions have actually seen an increase in the numbers of questions per librarian. ARL institutions’ patterns differ from those of other universities, which calls into question using ARL experiences as indicative of the wider academic universe.
“The Deserted Library”—a headline in the Chronicle of Higher Education. Much commentary, and some research, has addressed the topic of trends in usage of academic libraries. A decline in use of the physical academic library is seen by some commentators as an inevitable result of changes in the information landscape and in student, particularly undergraduate, information-seeking habits.
An underappreciated aspect of this issue is that just as not all libraries are alike, not all academic libraries are alike. True, one organization, the Association of College and Research Libraries (ACRL), serves libraries at community, baccalaureate, master’s, doctoral, and research institutions. However, some types of libraries are more often the site of published research than others. Association of Research Libraries (ARL) members are identifiable, organized, and have been committed for decades to extensive data-gathering. That makes their existence (for surveys) and their data a treasure trove for researchers interested in academic library issues. However, is the ARL perspective a reasonable representation of all academic librarianship? How representative is this group? Analysis of data about reference transactions from a much broader data pool suggests that ARL experiences may be disproportionately affecting the representation of the “decline” issue.
Literature Review
Writings on the possible decline in use of academic libraries include anecdotal or personal opinions or perspectives or broad, data-based research. Some authors argue or show that there is a decline; others that there is not. The “decline” expression that crystallized and sparked much of the debate was offered in a 2001 article by a reporter, Scott Carlson. The title and the publication venue say it all—“The Deserted Library,” on the front page of the Chronicle of Higher Education.1 The Chronicle is the premier news source for academia in general, and is read by administrators and faculty as a key information source for what is going on in their world. A direct counter argument was provided in Library Journal by Andrew Albanese.2 Drawing upon his contacts with staff at a number of individual libraries, he argues that usage has increased (although he did not conduct any systematic research to verify the representativeness of these observations).
Charles Martell’s 2005 editorial in the academic library journal portal: Libraries and the Academy presents a neat summary of the issue and uses ARL statistics to argue that reference usage has indeed declined.3 On the other side, in a review written specifically for nonlibrarian academics, Barbara Fister argues against the “myth” of the nonimportance of the physical library in the lives of undergraduates on the basis of personal observations (she is a director at a small, liberal arts college with mainly residential, traditional-aged students).4 Currently each side has some data to support its statements about library traffic. The ARL statistics center clearly shows a decline in reference (4.5 percent average annual decline, 1991–2005) and in circulation transactions (1.2 percent decline per year).5 On the other hand, a systematic and comprehensive survey of new or renovated academic library facilities showed consistent gains in usage.6
Interestingly, an ACRL task force looking into academic libraries’ future did not list a decline in use as one of their top ten assumptions for the future. The following items listed do not seem to anticipate any declines and in fact seem to predict increases: “Distance learning will be an increasingly more common option in higher education, and will coexist but not threaten the traditional bricks-and-mortar model,” “Students and faculty will increasingly demand faster and greater access to services,” and “The demand for technology-related services will grow and require additional funding.”7
What is not often a feature of the analysis or a subdivision of the data are the distinctive characteristics of the libraries involved. ARL library researchers and issues are well-represented in library literature. Yet college, university, and research libraries do show distinct differences, and not just in the most elemental and traditional measurement of a “research library”—collection strength. Collection variables represent three (or four) of the five data elements for the membership-defining ARL Index: number of volumes held, number of volumes added (gross), and number of current serials received. The others are total operating expenditures and number of professional plus support staff.8
A recent study of library staffing has demonstrated the reality of these by-type differences.9 It showed that staffing at doctorate granting institutions (both public and private) differs from those offering only master’s and baccalaureate degrees in ratios of librarians to patrons. Doctoral intensive and doctoral extensive (Carnegie classifications) institutions have the lowest instructional employee to librarian ratios of all institutional types (that is, each librarian potentially serves fewer faculty on average). There are more “other staff” per librarian in doctoral institutions than in master’s or baccalaureate institutions (that is, a smaller percentage of a doctoral institution’s library staff are librarians). Those statements are generally accurate for both public and private institutions. For private doctoral extensive institutions, there are far fewer students per librarian; for public doctoral universities, the student–faculty ratio is more similar to other types of institutions.
From these conclusions it is plausible to assume that library traffic also may well be affected by a library’s specific mission and focus, as well as the characteristics of the library’s parent institution in terms of student body, faculty roles, and its own strategic positioning with respect to physical versus virtual education. Much of this institutional differentiation is reflected in Carnegie institution-type classifications and in “control” classifications (the governance structure of either public, private not-for-profit, or private for-profit).
There is a need for more data to answer the question, is library use really declining? If so, does the decline vary by type of library? This study examines reported reference transactions from 2002 and 2004 (the main dependent variable) and analyzes differences among institution types (the main independent variable).
“Reference transaction” quantification is not unambiguous. The recording of reference transaction volume has existed for quite some time (at least since 1947) with the most important and systemic approach taken in 1976–77 by the Library Administration Division of the American Library Association along with the U.S. Department of Education to determine questions for the “HEGIS/LIBGIS” federal survey.10 It defined a reference transaction as one that involves “information sources” and the “library staff” (see appendix for the current and original definitions).
Writing about the process of reference transaction tallying has taken several forms. In one form, different variations for sampling are examined for those libraries not engaged in continuous tallying.11 In another, different categorization schemes for transactions are proposed.12