One library's experience with live, virtual reference

INTRODUCTION

The McGoogan Library of Medicine offered a live, virtual reference service to the University of Nebraska Medical Center (UNMC) community and to the general public from July 2002 to February 2004. This communication is a brief account of our library's experience with this service and the reasons for discontinuing it.

CASE STUDY

During the early years of the twenty-first century, we, like most librarians, noticed the dramatic increase in the number of library-based, virtual reference services offered across the United States and the world. We decided to try using virtual reference technology to reach our distance faculty and students. We obtained funding for this service through the UNMC Educational Technology Small Grant Program.

We chose the Virtual Reference ToolKit software from Library Systems & Services (LSSI) <http://www.lssi.com>. Like all other synchronous virtual reference products, the LSSI software would allow us to provide real-time, Internet chat service. The LSSI product also provided a combined chat and co-browsing feature that would allow the librarian to escort the user to the appropriate electronic resources. In addition, LSSI's product included a virtual classroom feature that would make it possible to provide synchronous instruction to multiple remote users, regardless of their location.

We named our virtual reference service AskMac. To promote the new service: (1) we placed an announcement in UNMC's all-campus newsletter; (2) we distributed AskMac promotional and informational items during the student orientation for fall 2002; (3) we gave AskMac demonstrations in library classes held both on and off campus for students, faculty, public librarians, and residents of Nebraska; and (4) we placed an AskMac button, linked to the AskMac service entry page, on every page of the library's Website.

Initially, we offered AskMac service twenty-two hours per week with two-hour shifts rotated among six librarians. In May 2003, we expanded the service to thirty-five hours per week. Commitment from those directly involved and their belief in the service was a major factor in the expansion of the service. No additional staff members were used to expand the service hours.

In June 2003, we noticed a marked increase in technical problems. These problems occurred concurrently with the purchase of the Reference Division of LSSI (including Virtual Reference ToolKit) by Tutor.com [1]. The librarians who staffed AskMac were frequently dropped from the Virtual Reference ToolKit server and forced to spend half an hour or more trying to log back in. During these times, the service was unavailable to users. To make matters worse, the co-browsing feature of the software did not work consistently, and librarians were frequently forced to rely on the “chat” function of the software alone. By September 2003, the vendor appeared to have resolved these issues.

In July 2003, we attempted to conduct a library training class using the software. Unfortunately, the bibliographic databases and presentation slides needed to teach the class did not work with the software at class time. Perhaps most significantly, the invited users were not able to enter the “classroom.” The failure of this virtual classroom experiment was a real disappointment. One of our primary reasons for exploring a virtual reference service had been our desire to provide synchronous instruction in the many homes of our distance students.

Much of our displeasure with the LSSI product stemmed from unexpected difficulties like the never-improved, extreme slowness of the product's chat function. However, some of our irritation was generated by known shortcomings of the product. For example, we learned that the software was incompatible with pop-up boxes when we were trained on the software. Initially, this inconvenience was not major. Then, we implemented two pop-up heavy improvements in our library's electronic services. We added Java-based, pop-up menus to our home page. The pop-up menus provided quick access to the library's most frequently used electronic resources and to our contact information. Because these menus did not even appear inside the AskMac co-browsing feature, librarians were forced to teach AskMac patrons more laborious methods for accessing our electronic resources or, alternatively, were forced to communicate instructions for use of the pop-up menus in the chat mode.

The second, problematic, pop-up-heavy resource was our new electronic journal finder. AskMac librarians could guide patrons through initial steps in use of the electronic journal finder but had to relate final steps in the slow-responding chat mode. We should note that although the librarians were frustrated with the technical problems, feedback from AskMac users indicated that they were generally happy with the service.

Finally, the difficulties with the service and low usage statistics led us to reexamine the costs and benefits of the service. According to statistics provided by the vendor, from July 2002 through February 2004, 144 sessions involved AskMac users posing actual reference inquires. These 144 sessions included disconnected calls, as well as sessions in which the patron failed to respond to chat messages from the librarian. The virtual reference service, therefore, brought in an average of 7 reference questions per month.

An analysis of the session transcripts indicated that most users were either on-campus faculty and staff or consumers and professionals from outside Nebraska. While we were pleased to provide our on-campus patrons with a new way to use the library, we had initiated the service to serve UNMC's growing number of distance students and faculty, UNMC-affiliated users unable to come into the library, and residents of Nebraska. None of these populations proved to be frequent users of the service.

Many of the questions asked through the virtual reference service were either ready reference questions that could be answered more quickly via email or telephone or were complex reference questions that were too time consuming to be answered satisfactorily using the slow, co-browsing/chat format.

Additional work with the virtual classroom portion of the software showed that this function did not work consistently and was too cumbersome to have a realistic use in provision of library education.

Before discontinuing the service, we wanted to know what sorts of virtual reference services other academic medical libraries provided. In December 2003, we reviewed the Websites of the medical libraries at the 126 US medical schools in the Association of American Medical Colleges (AAMC). We noted that 34 of these libraries offered a virtual reference service, 73 promoted an email reference service, and 19 did not noticeably promote any Internet-based reference services. Of the 73 medical libraries that offered an email service, 24 were on campuses with affiliated, nonmedical libraries that offered a chat service. It would be interesting to know why these medical libraries chose not to use virtual reference, when it is available on their campuses.

The McGoogan Library of Medicine decided to discontinue the AskMac service at the end of February 2004. The total number of questions received from our targeted distance users and our on-campus population, when considered in light of the high cost and problems with the software, was too low to justify renewing our service contract.

We, of course, considered switching to a less expensive form of live, virtual reference service. Other commercial vendor products that would be more compatible with our resources were considered. Free instant messenger or chat programs were also considered but dismissed because of the difficulties that would arise in collecting statistics and helping patrons perform software downloads. Based on our experience, we felt that regardless of the product, any chat service would bring in a relatively low number of questions, a number too low to justify the effort involved in maintaining the service. We considered a consortial or collaborative approach to providing virtual service unsuitable, as our service tended to draw questions specific to the McGoogan Library's collections and services.

With the discontinuation of the virtual reference service, we began to promote the email reference service and our toll-free telephone number more heavily. These services had served as back-up methods for reference help but have now been brought to the forefront as reliable, alternative paths to reference assistance.

CONCLUSION

The next generation of health sciences students (now in high school and junior high) use instant messenger and chat software on a regular basis [2]. When they enter college, they might expect and demand assistance in this same venue. Although we have found virtual reference service to be a costly and problematic way to handle a low number of reference questions at the present time, we will continue to monitor the virtual reference service arena in academic health sciences libraries and routinely reevaluate the need for such service at the McGoogan Library of Medicine.