Do the Outcomes Justify the Buzz?: An Assessment of LibGuides at Cornell University and Princeton University Angela K. Horne, Steven M. Adams, Michael Cook, Lance Heidig, and Christian Miller Abstract sities have collaborated on an assessment initiative to Springshare’s LibGuides has inspired significant buzz discover how these guides are valued on each campus. in the library blogosphere. Touted for its “Web 2.0” This study goes beyond the “2.0” dogma to empirical- functions, attractive interface, and ease of use for li- ly determine if LibGuides lives up to its publicity. No brarians, Libguides has transformed the way many li- evidence-based studies exist that explore the impact braries build web-based research guides. Cornell and of this software on the librarians who use it to create Princeton Universities decided to collaborate on an guides, the patrons who use guides for research, and assessment initiative to discover how these guides are the faculty who depend on guides to support their valued on each campus. This study goes beyond the classes. “2.0” dogma to empirically determine if LibGuides In a joint research study conducted during the lives up to its publicity. Fall 2008 semester, researchers at Cornell University and Princeton University conducted user surveys to Overview determine the extent to which LibGuides lived up its Springshare’s LibGuides has inspired significant buzz publicity and marketing. The Cornell and Princeton in the library blogosphere since the company first of- LibGuides implementations were used as case stud- fered their software in Spring 2007. Touted for its ies. Both institutions were early adopters and the buzz “Web 2.0” functions, attractive interface, and ease of quickly grew amongst staff using the tool and admin- use for librarians, Libguides has transformed the way istrators learning about its features. As the numbers of many libraries build, organize, and maintain web- guides being created and overall page views increased, based research guides. Cornell and Princeton Univer- it became clear at both libraries that we needed to Angela K. Horne, Director, Management Library at Cornell University, e-mail: [email protected]; Steven M. Adams is Biological and Life Sciences Librarian, at Lewis Library, Princeton University, e-mail: [email protected]; Michael Cook is Head of Public Computing at Mann Library, Cornell University, e-mail: [email protected]; Lance Heidig is Reference & Instruction Librarian at Olin and Uris Libraries, Cornell University, e-mail: [email protected]; Christian Miller is Reference, Instruction, and Outreach Coordinator at Catherwood Library, Cornell University, e- mail: [email protected] 172 Do the Outcomes Justify the Buzz 173 learn more about how the guides were being used. The Though progress was made towards this goal, usage statistics generated by the software only pro- staffing changes and programming challenges caused vided surface-level data. the project to remain uncompleted. Our research sought answers to a broad range of By 2006 the Cornell Library’s entire web presence questions: Did the social web features such as user- was in need of a major overhaul, one that would align embedded comments, chat, interactive polls, etc. in- the Library’s online environment more closely with crease student-librarian communication and were the needs and expectations of its users, many of whom guides with those features used more frequently? were by then already adept at Web 2.0-type function- Did librarians feel that the software met their needs? ality. A Library Web Vision Team was charged with Were librarians building more course-related guides the research and recommendations towards this goal. than subject guides, and if so, why? Were faculty will- After evaluating the need and usage of existing ingly adding guides to their course sites and were they subject guides at Cornell, the team researched the state satisfied with the new look and functionality of the of subject guide technologies available at the time. Part guides created for their classes? of the group’s charge was to find an emerging or exist- ing “off the shelf ” solution as opposed to developing History of Library Guides at Cornell and a “homegrown” system. Their final recommendation Princeton was to “implement, evaluate and constantly enhance At Cornell University: [Springshare’s library guide creation software] Lib- Across Cornell University Library’s twenty unit Guides.” In their final report they concluded: libraries, subject and course guides have tradition- ally been created as static HTML pages. These have “Library guides are a heavily used and im- been done in a variety of methods, representing ev- portant vehicle for guiding users to the most erything from hand-crafted HTML files, WYSI- useful resources of their field, thus helping to WYG-created guides (using Dreamweaver and develop the research skills of users… users re- other editing software), to content management spond well to and expect subject-based guid- systems such as Commonspot. Uniformity in the ance that is prominently placed on a web look and feel of guides changed depending on the site. Subject guides in the Web 2.0 environ- library; there was little consistency from place to ment (especially LibGuides from Springshare place unless they were from units that were admin- http://www.springshare.com/libguides) can istratively linked. Few if any had what are known do more than our current subject guides can as Web 2.0 features; many resembled traditional offer: they can also become interactive and col- print bibliographical “pathfinders” of subjects. Most laborative and can be integrated with course- of the libraries’ subject guides, research and writing ware such as Blackboard and social networks guides and “How do I find… ?”-types of guides were such as Facebook. Tag clouds can be used for grouped together on one long page on the Cornell visual presentation of subjects and terms. Li- Library’s website. brarians can present themselves in a way that In 2003, a group of Cornell librarians was makes them easier to relate to and that invites awarded a Cornell University Library Internal communication from the users via any IM Grant to develop a production system for the cre- network (Yahoo, AOL/MSN). Research shows ation of library guides. Specifically, they sought to: that the net generation expects experts to be available in their work environment—putting “… create a database management system librarians’ expertise within easy reach is one that allows librarians to create true “custom- of the attractions of LibGuides. Providing ized views” of a single persistent structured another way to promote and highlight our data source, thereby facilitating the presen- resources in relation to the needs of specific tation of information to specific audiences, users is also essential.”2 increasing the efficient use of staff time, and benefiting end-users through the availability At Princeton University: of more, and better, guides.“1 Princeton University librarians began creating on- March 12–15, 2009, Seattle, Washington 174 Angela K. Horne, Steven M. Adams, Michael Cook, Lance Heidig, and Christian Miller line research guides in the late 1990s for one of the versity Library, which traditionally had a conservative, early library websites. Early on, these early guides top-down approach to adopting new technology. were online representations of traditional print re- In the months following this symposium, a Princ- search guides with, hyperlinks to relevant electronic eton librarian heard about Springshare’s LibGuides and resources. Only a few librarians knew how to main- initiated a trial subscription. Librarians immediately tain web pages at that time and most pages were began using it to create course guides for specific built using HTML editing programs like Micro- classes and subject guides for departments. PUL’s soft FrontPage or Netscape Composer. Although the early adopters of LibGuides conducted a demonstra- library’s web presence evolved as a result of major tion to share their experiences with the system at the redesign efforts in 2001 and 2005, little progress was end of October 2007. The reaction to the product was made in making subject and course guides consistent overwhelmingly positive and Princeton subscribed to in look and feel. In 2005, the library’s web devel- LibGuides a few weeks later. opment manager created style sheet templates that subject librarians could use for their guides. Some Methodology librarians elected to use the guide templates, oth- Surveys were conducted of three target population ers designed their own sites with in Dreamweaver or groups at both Cornell and Princeton: faculty for other web design software. The Princeton University whom at least one course LibGuide had been cre- Library (PUL) website had pages dedicated to both ated; students who had taken at least one class for course-specific and subject-specific research guides, which a course LibGuide had been created; and library but the guides themselves were inconsistent in style staff who had created at least one LibGuide. We used and content. Cornell’s online survey tool, Checkbox, to administer As the sophistication of web design increased in and manage all six surveys. Checkbox provides default the early 2000s, the expectations of academic library reports and the ability to download all responses into users also evolved. Experiences with sites like Yahoo, MS Excel for customizable analysis. Amazon, Google, and many others in the commercial Survey questions were identical at both schools sector changed how people seek information, and al- except for one demographic question concerning the tered what patrons expected of libraries. Library users respondent’s academic department / program affili- began find it tedious to navigate webpages that do not ation. Terminology differentiations between Cornell reflect their aesthetic and usability preferences. Librar- and Princeton’s academic structure required a slight ians around the world were looking for ways to lever- modification of that particular question. The survey age the power of the web to avoid the being rendered questions will appear in the final version of the paper obsolete.