Understanding Reference Transactions: Transforming an Art into a Science (review)

This scholarly and theoretical work represents a significant step forward in the ongoing effort to examine reference services as an intermediation process. Through systems analysis and the formulation of a model for an expert system using controlled multilevel independent variables, Saxton and Richardson propel reference service research well beyond the anecdotal and traditional flow chart analysis that has been influential in reference study analysis over the past two decades. The authors derived their model for predicting service outcomes from a set of conditions that includes library setting, librarians, and the person posing the question.

The foundation for this book's theoretical orientation began by identifying 13 volunteer public libraries in Los Angeles and Orange Counties in 1998. Interestingly, this project used real questions actually fielded at the volunteer reference desks. Without using canned questions, the accuracy rate of librarian responses approached the 90% mark. This figure is at variance with the 55% one so often cited as a reputed standard for reference desk accuracy levels. There appears to be value in ongoing research in this area that could prove equally useful for measuring reference outcomes in academic libraries. The authors believe that the quest for an accurate and valid measurement of reference service is one of the more urgent research needs in library and information science.

John V. Richardson Jr. is a professor of information studies at UCLA and editor of Library Quarterly. Chapter 7 in the book under review was published previously in the May 1999 issue of College & Research Libraries, under a variant title, and this fact is duly noted in the preface. Matthew L. Saxton, now assistant professor at the University of Washington Information School, did much of the research for this book while a doctoral candidate at UCLA.

This book represents a paragon of scholarship. The research project, sampling techniques, dependent and independent variables, and outcomes are all explained in language that speaks to both theoreticians and practitioners in managerial and administrative positions. Chapters Two and Three are concise historical reviews of the scholarly literature and of the authors who have contributed groundbreaking research in reference service evaluation over the last century and a quarter. In the future, this work could be included in such a historical roundup. Appendix D, which comprises about a quarter of the book, recapitulates results from previously published works that used similar or related statistical data. This presentation permits comparing statistical results found in the various works.

Saxton and Richardson developed their reference transaction model for predicting outcomes in a public library project. It would appear, however, that there is much in this model germane to the study of the reference process in academic libraries. Independent variables on query difficulty, query currency, user's education level, librarian's experience and education level, and collection size—all seem like integral components of any reference model designed to focus on academic or other libraries.

There are some unanswered questions regarding how the service model constructed here would demonstrate equal validity in the reference service transformation sweeping academic libraries across the United States. Academic libraries are de-emphasizing the traditional walk-in trade and experimenting with a whole range of virtual reference, tiered services, e-mail reference, knowledge-based systems and other online and/or organizational options. How would the various outcomes in this model work within a regionally, nationally, or even internationally collaborative virtual reference environment relying on networked and other emerging digital resources?

While librarianship requires at least an MLS at most public and academic libraries today, changes are in the wind here as well. Currently, librarians obtain their ALA-accredited MLS before they obtain their first librarian position. This model may not withstand scrutiny ad infinitum, leading to higher percentages of staff with skills in information technology, computer systems, networking, and other areas. How...