E-Mail Reference Responses from Academic ARL Libraries: An Unobtrusive Study
Stacy-Bates, Kristine, Reference & User Services Quarterly
E-mail reference connects patrons with reference service providers who can find answers for them. How accurate are the responses to e-mail questions? How firm are libraries' guidelines for time of response, type of questions addressed, and category of user whose questions will be answered through e-mail reference? This article examines e-mail reference services provided by academic libraries in the Association of Research Libraries (111 sites). Sets of three pretested e-mail reference queries were sent to each of these libraries during February 2001. Twenty-one assessment questions were used to record characteristics of the e-mail reference service pages and characteristics of responses to the reference queries; comparisons were made between service policies as stated and as practiced. Assessment questions covered the topics of e-mail reference polities, services to non-affiliates, elements of reference interviews, features of reference service enabled by the e-mail medium, and accuracy of query answers.
Thank you for using the [name] University Library's Ask-a-Librarian service. Your question has been forwarded to the liaison librarian for Not Sure.
This message appeared on the computer screen as a result of filling out an online form for an e-mail reference service and selecting "Not Sure" from a dropdown menu to indicate the subject of the submitted query. Providing "Not Sure" as a choice allowed for the amusing response above--perhaps suggesting an alternate job title for reference librarians. More important, the presence of this option was an acknowledgment that some patrons are unclear on how to categorize their information needs. One aim of reference service is to alleviate our patrons' uncertainty, and e-mail reference service can be a means for achieving that aim.
E-mail reference service is now a common component of large academic libraries' services, and has been an established service for long enough that simply providing it no longer suffices. It is time to look at the quality of the service provided in this area. Quality can be affected by a library's choices in setting policies for email reference service, planning it, staffing it, and advertising for it. But none of these things is the actual service. To measure the quality of the output of e-mail reference, it is necessary to examine the e-mailed query responses. Patron satisfaction with e-mail reference service has been measured for individual libraries, but this does not allow for comparisons among libraries or give a picture of the overall quality of electronic reference service provided by these academic libraries as a group. (1) A means for making comparisons and broader statements about quality was needed.
After a 1999 study of e-marl reference pages on Association of Research Libraries (ARL) academic library Web sites, the author conducted an unobtrusive study of e-mail reference practice as performed by these libraries in early 2001. (2) The objectives of the 2001 study were to examine common characteristics of responses to e-mail reference queries, and to collect data on these characteristics as expressed in responses received from the reference service providers.
Expectations for the state of e-marl reference service were:
* Libraries would give all Web patrons a clear statement of their policies concerning e-mail reference service, and then meet or exceed the goals for service mentioned in those policies.
* Libraries would provide more service to their affiliated patrons than to non-affiliates.
* Reference service providers would extend accepted good practices for reference interviews into the e-mail environment.
* Libraries would take advantage of technology to provide features not applicable to face-to-face or telephone reference situations.
* Reference service providers would respond to clearly worded queries with direct answers, or with URLs of Web pages containing direct answers, and would answer with high levels of accuracy. …