![]()
Introduction
Discovery tools are now becoming more common in the academic library landscape, and more products are now available from vendors. While librarians are advocating and promoting their use by students and faculty, they are also evaluating their searching capabilities, their usefulness, and on-going maintenance requirements. This special issue of College and Undergraduate Libraries provides a comprehensive review of the literature published on discovery tools, which indicates how new these developments are in the field of librarianship (Thomsett-Scott and Reese).
A number of issues are raised by the authors published in this special issue. Key questions addressed include how Web-scale discovery is changing the conversation about database and catalog searching in academic libraries (Kornblau, Strudwick and Miller). The quality of searching queries is an important aspect of evaluating discovery tools (Meadow and Meadow). New discovery tools, such as a Javascript bookmarklet, are being created by librarians to enhance searching library resources on the fly (Arnett and Forrestal).
Usability testing is a time-tested strategy to help librarians understand how users respond to resources such as discovery tools (Comeaux). While using new search tools is always a challenge for students, librarians are also creating new best practices for introducing discovery layers via library instruction (Fawley and Krysak). At the level of the end user, Emory University has developed an assessment framework to standardize feedback from users (Durante and Wang).
Moving beyond federated searching into new ways of searching and discovering resources is a key strategy for many libraries that are embracing discovery tools (Wang and Mi). As early adopters, many libraries were challenged by the requirements of cataloging maintenance and authority control for discovery systems (Harpel-Burke).
Case studies included in this issue address needs analysis and effective decision making regarding what works and what doesn’t at colleges such as Davidson (Milberg). Discovery tools often go through a beta testing phase and may be implemented at multiple campus libraries, which is a focus of the article about Seminole State College of Florida (Kaufmann, Larsen, and DeSalvo).
Two articles in this issue address discovery tools at Historically Black Colleges and Universities (HBCUs)—one highlights the library landscape at HBCUs (Guthrie and McCoy), while the other details the discovery tool implementation at two distinctly different colleges (Walker and Sims).
Another case study covers Primo’s implementation and production phases using evaluation tools as part of the Franklin College experience with ExLibris (Mahoney and Leach-Murray). An additional Primo case study at a small regional consortium examines how the process of implementation and configuration works for libraries in a consortium (Mortimore, Dunn, McNaughton, Novicki, Wade, and Whittington). A case study of Summon at Eastern Michigan University explores the challenges and successes of essential timetable elements for implementation (Wrosch, Rogers-Collins, Barnes, and Marino). Lastly, the evaluation of discovery services at Lynchburg College is addressed using informal testing and formal surveys (Ours).
Overall, this special issue will provide librarians with a better understanding of the initial implementation challenges and the on-going questions to be addressed by academic libraries that initiate a discovery tool.
John S. Spencer
Coordinator of Reference and Chair of Public Services at Foley Center Library, Spokane, WA
Christopher Millson-Martula
Director of the Library, Lynchburg College, Lynchburg, VA
![]()
Academic Libraries and Discovery Tools: A Survey of the Literature
BETH THOMSETT-SCOTT and PATRICIA E. REESE
Discovery Park Library, University of North Texas, Denton, Texas, USA
Many thanks to Diane Wahl at the UNT Libraries for her valuable comments.
Discovery tools are becoming increasingly popular in academic libraries, yet there is little information available concerning the tools used regarding their implementation, maintenance, and evaluation, or the tools’ effects on public service librarians. Key questions considered through a literature review include the most common tools used, how their effectiveness is evaluated, and user satisfaction data. Readers will learn how public service librarians are involved in discovery tool processes; the usability of discovery tools; and how to incorporate discovery tools into information literacy endeavors.
INTRODUCTION
Many libraries want to identify “hidden collections” in order to make them available to users. In order to accomplish this goal, libraries primarily needed to enter these items in their catalog or create some sort of finding aid. Of course, this approach does lead to discussions about “findability” of items in the catalog in general, particularly as libraries have moved to include a wide array of digital and electronic content in their collections.
Substantial work on metadata had been done to make these collections and other items Web accessible, but it seemed very cumbersome when users were required to perform a catalog search as well as a Google search to locate library resources. Library catalogs tend to be designed using outdated ways of thinking, and they are well known to be difficult for users. Most librarians would agree that, out of all library resources, users have the most difficult time using the catalog. Catalogs are not searchable by Google, and, therefore, they leave a huge resource of information unavailable to users. Users know the library has things they need but they can’t find the right pathway. Combining a “Google-like” search box with the wealth of a library’s information resources may help us recover users who left our “walls” for the ease of Web searching. Hence, discovery tools!
S. R. Ranganathan’s fourth law is “save the time of the reader,” and libraries would do well to place this at the top of the priority list when designing products. Howard and Wiebrands (2011, 1) state that “… it is the behavior of our information seekers that should drive our services.” The library should be built into user workflows rather than trying to fit the user into the traditional library (Liu and Liao 2009). The focus should be on the users, and our systems should fit the information seeking behavior of our patrons (Howard and Wiebrands). Lougee (2009, 612) adds that the “… library is no longer the archive that deals primarily with the products of scholarship, rather there is a role to be played in all aspects, in all stages, of the processes of scholarship.”
Allison (2010) emphasizes also that library tools need to follow (or lead) innovations on Internet search engines. She adds, “Librarians need to compromise between powerful search interfaces that are too complex for students and featureless interfaces that are too basic to be useful when drill-down is needed for large results sets” (377). Gross and Sheridan (2011) note that libraries need to make themselves more a part of the search process again. Since digital media has become so accessible, fewer and fewer searches start at the library; thus, we need to maximize resource use by reducing user frustration. Howard and Wiebrands suggest that the focus on information literacy is due to the need to teach students how to use complicated resources that cause unnecessary frustration due to poor system design. Anderson (2006, 33) writes that “… tutoring is a far less effective and efficient solution to the problem than fixing the search interface, which is exactly what Google has done.”
In a 2004 discussion panel, Kenney notes that libraries need a Google or Google-like system rather than nonintuitive databases in order to provide the value of all the best information versus all that can be found in a shorter time. Library resources need to be made available through Google so that users can find it, especially since not all people have access to the powerful library databases. Users need to find items in order to use them. It is well known that users will not use a Web site if it is not user friendly. Now that options are available, this can be applied to library resources, especially the library catalog. Perrett (2010) reports that even adding additional copies of textbooks does not increase user satisfaction since patrons are still unable to find items in the catalog. Extra copies are only useful if students can find them. Providing a discovery tool that may make it easier for students to find items will have an impact on their overall satisfaction with the library.
McDonald and Thomas (2006) note that there are disconnects between current users’ wants and needs and the traditional values of libraries. They classify the disconnects into three areas: technology, policy, and unexploited opportunities. Patrons become frustrated when having to use library systems that are limited by library culture rather than by technology; hence, they seek systems that are easier to use and freely available. Michael Gersch, senior vice president and general manager of Serials Solutions, states, “… we’ve all seen the statistics that show users want the quality of the library but prefer the simplicity of Google.” (Luther and Kelly 2011, 4) Wilkes and Gurney (2009) report that, based on two surveys of freshmen, more than two-thirds of respondents preferred using the Internet to find information. The majority of respondents used Google scholar as their preference for a known interface, but they were frustrated by the lack of full text and the inability to limit to peer review. Students using a database reported that easily limiting to full text and peer-reviewed sources was a significant benefit. There was a significant increase in comfort using Google Scholar over two semesters and only a very small increase in comfort using library databases. Discovery tools take a “Google” approach and include limiting features to help define the topic and get the best of the best (Kenney 2011). Way (2010) reports that users found databases and catalogs somewhat confusing, particularly the aggregated content and embargoed items, together with the problematic direct linking to many full text articles. In addition, there are many choices of databases, and each database has a variety of limits. Often the limits, as well as catalogs and databases in general, do not utilize terminology familiar to patrons.
Liu and Liao note that 89 percent of students use search engines first, with only 2 percent starting at a library site. However, patrons will use library resources if they are found through the Web. The results of the survey indicated that users want a single point of discovery for all resources, speed, an easy to use interface, relevant results (facet, natural language, topic clusters), and everything electronic (full text, multimedia download, e-Books) and seamless (and accurate). Users also desired widely available means to contact library staff, self-service, personalized services based on needs, librarians with technological skills to allow the user to exploit functions of resources, and staff with the “… interpersonal skills to enable them to communicate effectively and sympathetically with users …” (Liu and Liao 2009, 303). The authors added that “nowadays the number of visits to the library Websites is no longer as important as it was, the most important thing is that your users find your library and its services in their favorable environment” (304). The authors recommend focusing on promoting information reliability and special collections but that these must be easily available and findable. The authors also advocate for the promotion of librarians as one of the strengths of a library, whether through in-person or virtual means of contact.
LITERATURE REVIEW
A concept called the “next generation catalog” appeared when librarians came to the conclusion that the library catalog was no longer working for users (Wisniewski 2010). Yang and Wagner (2010) compare proprietary and open source tools according to the checklist of twelve features desirable in the next generation catalog according to Breeding (2010). This is a very detailed report and presents several charts that were among the very first to compare the different types of discovery tools available.
McHale (2009) notes that metadata is vital to retrieval in federated searching and next gen catalogs. This also applies to discovery tools. If meta-data is not present in sufficient quantity and quality, retrieval is jeopardized regardless of the system. Additionally, functionality is limited depending on the items included in the search system. Reasons for items not being included are vendor prohibition, limited number of seats, prices based on pay per view, and no current “translator” for the item.
Federated Search
The federated search engine represents an attempt to help users in their discovery process (Keene 2011). This technology allows users to select the databases that they found useful with a single search box query that is sent to each of the chosen databases separately. However, the average user is often unaware of the best databases to use as discovered through evaluation of federated search systems. Williams and Foster (2011) investigate how federated search engines were incorporated into Association of Research Library member Websites in 2010. They conclude that more and more libraries were implementing federated search engines and that on-going research should explore librarian and user opinions. McHale (2009) reports that, after a usability test of their federated search system, graduate students and faculty would recommend it to undergraduates in their class, but not use the tool themselves. Several respondents did note that the availability of previously unknown resources was helpful. These results show the value of the one box search method despite a variety of concerns.
While federated search systems have issues, such as response time, they do serve as a precursor of discovery tools (Bhatnagar et al. 2010). Pradhan, Trivedi, and Arora (2011) provide a good review of the pros and cons of federated searching. Slow response time and time outs can be quite frustrating to users. The results can lead to information overload since limiting is not generally performed. It can be difficult to analyze results from all of the different databases. This article offers a very good explanation of all of the components that go into resource discovery, and it introduces the major players in the game. Stone (2010) has an interesting discussion on federated search engines and their lower than expected success. R. F. Woods (2010) shares her concerns with using federated searching in a law library. Her findings show that subject specific databases are vital to specialized areas of research and practice. Thus, regardless of the technology, specialized tools need to be in the forefront of advanced research.
Discovery Tools
After discovering that federated searching did not satisfy many of our users’ needs, notably speed and thoroughness, we may have a better prospect in the discovery tools that have been developed from the next generation catalogs. Generally, these products offer a single point of access utilizing a centralized consolidated index that combines library catalogs, e-journals, databases, and Web-based resources, as well as digital archives. Since the content is preindexed, response time is significantly improved. The development of Google Scholar in 2005 was definitely another impetus for librarians to move from federated searching to something faster and more comprehensive.
Breeding (2010) discusses his idea of what discovery tools should be and the progress that libraries have made in developing an interface for today’s library users. Vaughan (2011, 8) describes Web-scale discovery and states, “The primary goal of this work is to provide a valuable foundation to libraries that wish to know more about library-focused web scale discovery services and to aid libraries contemplating a marketplace review for their local environment.” He does an excellent job describing concepts that librarians should know about, and he provides important information on all of the major players on the field. Luther and Kelly (2011) discuss what discovery tools are and the areas that need to be considered before purchase. These areas or factors include content, the search process, the fit, and the cost. Discovery tools are changing and expanding continuously. Some libraries have combined different services in various combinations, customizing the tools to meet their individual institutional needs (Vaughan 2011). Bhatnagar et al. (2010) note that more vendors are being more cooperative by allowing coordination among products; this allows for the full potential of Web-scale discovery services. It is possible that vendors have become aware that their products are no longer meeting the needs of libraries, and, thus, they see their inclusion in discovery tools as essential to their survival.
The most popular discovery tools are Summon by Serials Solutions, WorldCat Local by OCLC, EBSCOhost Discovery Service by EBSCO, Primo Central by Ex Libris, and Encore Synergy by Innovative Interfaces. Open source discovery tools, such as VuFind, Blacklight, and eXtensible Catalog allow institutions to tailor their tools for different users and collect...
