May08.indb What’s in a Name? Using Card Sorting to Evaluate Branding in an Academic Library’s Web Site Peter Hepburn and Krystal M. Lewis Libraries are pressed to effectively promote use of the tools they provide users as well as their role in creating, selecting, and purchasing them. Applying “brand names” generated within the library is one promotional strategy. Usability testing at one academic library demonstrated how the card sortingtechniquecan beusedtoevaluate branding efforts.Thestudy found that library users do not recognize or comprehend library brand names in the absence of a consistent approach to branding even if they do use the services that have been branded. ibraries are pressed to dem- onstrate their worth, both to their parent organization (the college or university) and to their users. Librarians frequently observe that users are accustomed to relying on Internet resources found through freely available search engines such as Google or Yahoo. As a result, users do not o en understand that the various electronic subscriptions to scholarly materials that they use have been ve ed by the library, are generally more reliable, and are paid for through the library’s budget, or that the library offers services to obtain scholarly materials they do not own. The library provides access to excellent scholarly material, but users are o en unaware of these resources. To promote the use of the resources and services they provide—and to establish the library’s role in the development, selection, and provision of them—libraries will name or “brand” their resources and services. For example, a database will be known as X Library’s Database Y or something similar. The library’s Web site then must balance the seamless provision of such resources and services with the library’s desire to underscore its role in creating, choosing, and offering them. Evaluating the presentation of resources and ser- vices through library Web sites has been the subject of a number of usability tests in past years. While Web site usability tests usually uncover a host of informa- tion about the placement of links, color schemes, and site architecture, testing can also measure responses to library branding efforts. Peter Hepburn is Digitization Librarian and Assistant Professor in the Richard J. Daley Library at the University of Illinois at Chicago; e-mail: phepburn@uic.edu. Krystal M. Lewis is Assistant Reference Librarian and Assistant Professor in the Richard J. Daley Library at the University of Illinois at Chicago; e-mail: kmlewis1@uic.edu. The authors thank Francis Kayiwa for his contributions as one-third of the card sort test team. They are grateful to Lisa Wallis for her contributions to the process and to the UIC Library’s Web Oversight Commi ee for their support of the undertaking. They wish to thank Ann C. Weller and John M. Cullars for their considerate comments and suggestions. 242 mailto:kmlewis1@uic.edu mailto:phepburn@uic.edu Using Card Sorting to Evaluate Branding in an Academic Library’s Web Site 243 Literature Published results from a number of us- ability studies have measured the efficacy and appeal of library Web pages. Usability studies are generally well known in li- braries including UIC,1,2 but the card sort technique is not common. Card sorting is a method in which subjects sort index cards labeled with Web site content into groupings that make sense to them. An ex- amination of the library science literature for models of this technique uncovered four major studies. In 1999, a er running a task-oriented usability study of their site, the MIT Libraries conducted a card sorting study of Web site content.3 This was followed by a category description survey and a “reverse category survey” in which users were asked which of five categories they would choose first to find a specific item from the Web site. The results of the card sort and surveys were used to group and (to some extent) label content according to the user’s perspective in the redesigned Web site. Cornell University was also an early adopter among academic libraries of the card sorting technique. The Faiks and Hyland study at Cornell was confined to a section of the Cornell Library Web site, the Gateway Help, not the whole site. Moreover, the Cornell study was less concerned with terminology specifically than with organization of help topics.4 In a study at SUNY-Buffalo (now the Univer- sity at Buffalo), participants grouped tasks commonly performed using the library’s Web site. The study focused almost exclu- sively on site nomenclature, though not so far as branding might be involved.5 A University of Arizona study underscored some of the guidance the MIT and Cornell studies offered, but neither it nor the rest of the card sort literature discusses issues of branding in libraries.6 In non–card sorting usability studies, where issues with terminology and nomenclature were important parts of the findings, branding of library resources and services has not been closely examined. There is li le wri en on branding in academic libraries, whether or not usabil- ity studies played a role. One exception is Duncan and Fichter’s case study in which the authors describe the process of nam- ing a new real-time electronic reference service at the Health Sciences Library, University of Saskatchewan. As part of that study, the authors employed a us- ability technique similar to card sorting called “preference testing” in which test participants selected from among a set of alternative labels or suggested other wording of their choosing.7 In this case, preference testing was conducted prior to the introduction of the new service at the University of Saskatchewan. In an ar- ticle from the professional (nonresearch) literature, Rowley examines branding techniques in libraries and the impera- tives that drive them.8 In her survey, she linked commercial branding initiatives to what libraries could and already do. For example, she identifies the brand logo as shorthand for everything the brand should mean to users. In the case of libraries, this means display of the library logo on all Web pages where its presence adds to the user experience or understanding. A couple of key points surfaced in the literature: that “[libraries] already have what every Web site wants: a trusted and respected ‘brand’ identity [and that they should offer] a highly positive user experience that results in frequent return visits—or what marketers call ‘stickiness’ to a brand”9 and that “the choice of cutesy names and logos—sometimes based on the region’s mascot or something similar … may be of possible value for branding and marketing, but of no inherent mean- ing to new users.”10 Methodology In 2005 a new commi ee was convened at the UIC Library to redesign the library’s Web site. In recent years, the Web site had grown to hundreds of pages and sprawled across many subdirectories, partly as a result of decentralized creation and maintenance of the site and pages. The commi ee was charged with harmo- 244 College & Research Libraries May 2008 nizing the range of styles evident in the site and ensuring that the pages conform to newly established university visual identity standards, as well as assessing the information content and redesigning the architecture of the site. As a first step in the UIC Library’s Web site redesign project, the commi ee cre- ated a Web survey to gather information regarding the library’s Web site such as ease of navigation, quality of content, visual appeal, terminology, most-used features, and frequency of use. The survey was posted on the library’s Web site for a month, and three hundred fourteen re- sponses were received. An analysis of the survey results indicated that the two areas that presented the most problems for us- ers were the navigation of the Web site and the terminology used within it. Thus, a subgroup of the commi ee decided to conduct a card sorting usability test to determine the architecture and terminol- ogy of the new library Web site. Having received approval from the university’s Institutional Review Board, the card sort team sent out, in March 2006, an e-mail advertisement for participation in a card sorting test to the participants in the earlier Web survey. In addition, an announcement seeking participants for the card sorting was posted on a campus electronic bulletin board. An incentive of a $10 gi card to Barnes & Noble was of- fered at the completion of each test, and refreshments were served at the sessions. More than 50 persons expressed interest in participating in the study. In the end, the team scheduled 18 tests and, because of participant dropout, completed 15. The number and types of participants recruited for the study were based on the recommendations of Jakob Nielsen and adjusted to accommodate the organiza- tional structure of the UIC campus and libraries.11 Nielsen states that the five us- ers recommended for traditional usability tests will garner a correlation of only 0.75 in a card sorting usability test, whereas a 0.90 correlation will be achieved by test- ing 15 users. Tullis and Wood, meanwhile, recommend testing more users, as many as 20 to 30.12 However, given limited resources, deadlines for delivering the findings to the Web redesign team, and the marginal gains in correlation that would come from a larger study popula- tion (0.93 for 20 participants and 0.95 for 30), the researchers decided that testing 15 users would be sufficient. Participants recruited for card sorting usability studies should be representa- tive of the users of the test site, according to Nielsen,13 so the researchers limited participation to the core library users: current students and faculty at UIC. UIC includes two Chicago campuses: one for arts and humanities, social sciences, and professional studies, and another for the health sciences. Additionally, there are health sciences campuses in Peoria, Rockford, and Urbana that are also part of UIC. The library’s organization mirrors this structure. Because the two largest site libraries (Chicago’s Richard J. Daley Li- brary and Library of the Health Sciences) have comparably sized but otherwise very different user bases, the team drew approximately even numbers of respon- dents from both those user populations. Graduate students responded in greatest number to the invitation, a result that echoed the library’s earlier Web site sur- vey results and may also reflect a higher level of use of and concern for the library’s Web presence by that subpopulation. Faculty, meanwhile, responded in modest numbers to the earlier Web survey and almost not at all to the call for card sorting participants, resulting in only one faculty participant in the testing. Due to logistical considerations, all card sorting sessions were conducted at the Chicago campuses at the participant’s choice of either the Richard J. Daley Li- brary or Library of the Health Sciences. Each test participant was given a set of 93 index cards, each one numbered and labeled with the name of a library resource, service, or Web page content either existing or yet-to-be-developed (Appendix A). The 93 labels were pre- http:libraries.11 Using Card Sorting to Evaluate Branding in an Academic Library’s Web Site 245 dominantly drawn from existing links on the library’s Web site and augmented by suggestions from the Web site survey. The participants also received a pad of adhesive notes, felt marker, pen, and blank index cards. Using the guidelines developed by the MIT Libraries,14 partici- pants were instructed to: • sort the cards into piles represent- ing categories of information that made sense to them, making as many or as few categories as they chose; • speak aloud, stating the rationale for sorting cards into the piles created; and • put a post-it note on top of each pile and write a name for that category on it. The name could be as long and descrip- tive as the participant chose. Additionally, the team asked partici- pants to: • use the blank index cards to create new cards for anything the participant felt was missing or to create duplicate cards where the participant felt that a card belonged in more than one category; • set aside cards that had no meaning to them in some sort of “discards” pile; and • make a category called “other” or “general” for cards that seemed not to fit into any category but still should be retained somehow. Two team members monitored each session, with one providing most of the instructions and both taking notes. Because Nielsen suggests combining correlations with users’ comments to in- form Web site architecture,15 participants were asked to explicate their mental processing of the task at hand while the team members wrote notes on what was said and made other observations. Users were permi ed to ask questions during the test, but the monitors could neither define the label on a card nor indicate whether a card referred to existing or potential Web content. Once the sorting was completed, the participant filled out a brief survey about general Internet site usage and preferences. At the end of the session, the team members debriefed the test participants, inviting questions about the card labels, the study, and the library in general. Findings Although the card sort testing was in- tended to determine the architecture of the library’s forthcoming Web site and the language used within it,16 the tests inadvertently provided some helpful measure of the efficacy of library brand- ing. A surprising, unanticipated finding was that in no case was the branding widely effective; and, in one case, the brand name was unknown to a majority of the test participants. This has some hard implications for the delivery of library resources and services.17 Branding at the UIC University Library is neither comprehensive nor consistent. The library did not actively embark on a unified campaign to brand its resources and services. Instead, the services that were branded were done case by case. Each time a new service was introduced, a name was devised by those who managed it rather than according to a librarywide marketing plan. Among the branded services, there were variant naming conventions and no standard gateways to or other graphic expressions (like watermarks) of the connection between the resource and the library. The branded resources and services all used the initials “UIC” and followed one of two models: • In apparent emulation of the American Library Association’s pervasive “@ your library” campaign, “@UIC” has been appended to the name of the service: MyILL@UIC, ERes@UIC. • The university’s initials have been incorporated directly into names of resources: qUICsearch, InfoqUIC, UIC- CAT. Of the 93 cards used in the test, users discarded 47 cards (51%) at least once. Out of those 47, 10 cards were set aside at least 20 percent of the time (3 or more instances in the 15 tests). Of the six most discarded cards, five featured some sort of brand- http:services.17 246 College & Research Libraries May 2008 TABLE 1 Cards Most Frequently Designated as “Discard,” “Unknown,” or “Redundant” Card label Incidence Percentage % ERes@UIC 10 67 InfoqUIC 7 47 MyILL@UIC 6 40 qUICsearch 6 40 Resources 4 27 UICCAT 4 27 Note: An additional 41 cards were designated “discard” one to three times each. ing. Leading the way was ERes@UIC, rejected 10 times in the 15 tests, a rate of 67 percent. Table 1 shows how the other branded labels ranked. User recognition varied from one brand to the other. The comments, however, show that even when users recognized the name of the resource and what it stood for, they still found faults with the name. Of the two “@UIC” styled brands, MyILL@UIC was the be er recognized. MyILL@UIC is the library’s service that permits interlibrary loan users to place and track requests and view their request history. Five participants (33%) under- stood that the “my” part of the name was connected to a personal account and one (7%) categorized the card in a manner that connected it to interlibrary loan. The card was otherwise discarded 40 percent of the time, and it remains unknown whether users who categorized the card actually understood what the service is. Two participants (13%) referred to this service as “My Illinois at UIC.” Although one stated, “My Illinois? I don’t know what you mean by this,” the other did not indicate what she thought the name meant. The test team speculates that the student may have been thinking of a state government information service or documents repository (the library does have Illinois government documents in its collection) or may have interpreted it as a personal service related to the University of Illinois. Another participant included the card in a pile labeled “portal features,” and noted that MyILL@UIC really sounds like a patient history. Whether partici- pants understood the meaning or not, the comments and discard rate demonstrate that the name is not necessarily obvious. The other brand that used the “@UIC” naming convention was more frequently discarded. ERes@UIC, the library’s electronic course reserves service, was discarded 67 percent of the time. Those participants who kept the card did not generally categorize it in a way that sug- gested understanding of the meaning. In the five tests (33%) wherein the card was categorized, two (13%) had it grouped in “Other” and one each (7%) in “Search- ing in General,” “Electronic Resources” (which might be logical considering how similarly “resources” and “reserves” may be abbreviated), and “Collections/hold- ings.” In the last instance, the participant (the last to undergo testing) surprised the team by knowing exactly what ERes@UIC was. Although high usage statistics indi- cate that UIC’s electronic reserves service is a success and that students must be familiar with the concept of electronic course reserve readings, the evidence suggests that users have trouble with course reserve terminology in general. In SUNY-Buffalo’s card sort test, participants used many names to refer to the service, such as “reserve,” “course readings,” “class readings,” “readings for outside class,” and “reading for reserve.”18 Since the service lacks a name that is easily understood by users, branding may not account for the total number of discards. Certainly, though, branding did not make the service more recognizable. One participant thought that ERes@UIC was a page and hold or recall service and another commented that it should fit in with other “acronym-y” names though it did not seem to. The “acronym-y” names to which the participant referred use the other brand- Using Card Sorting to Evaluate Branding in an Academic Library’s Web Site 247 ing convention to integrate the initials “UIC” into brand names at the library. To incorporate the three le ers sequentially in a word is challenging. “Quick” is one of the few words in which they appear, and it also describes a desirable characteristic for many services. Consequently, two of the labels play off that word: InfoqUIC and qUICsearch. Test participants did not always connect the idea of “quick” (or “quic”) with the names. Several spelled out the “quic” part in pronouncing the names (in other words, q-u-i-c), perhaps because it’s lacking the final le er “k.” Within the library itself, there has been confusion over the spelling of InfoqUIC and qUICsearch. Correct spelling sets off the “UIC” in capitals, but variations have appeared on Web pages, with the whole of “quic” in capitals. Given the inconsistency within the library, that users spelled out the “quic” rather than pronounce it as one syllable was not entirely surprising. Other than the indications of speedi- ness and institutional branding from the names, users were confronted with “info” and “search,” two terms that apply to many things in the library. Users know these words, but they did not always understand them in the context of the ser- vice. Some comments and observations: • One participant thought InfoqUIC might relate to the catalog, saying, “I have no idea what that is.” • One participant thought that Info- qUIC meant “Information Quick” so it should go with “Reference.” • Of qUICsearch, one said, “Must everything be named ‘UIC?’…[Is it] searching libraries? The whole campus? I’ll put it by all the searching stuff.” • One participant put qUICsearch with InfoqUIC because of the similar names, pronouncing the la er as InfoQ- U-I-C instead of as “info-quick” and wondering why the name included the le er “Q.” The two services have very different functions. InfoqUIC is a fee-based article delivery service operated through interli- brary loan and qUICsearch is the library’s application of WebFeat’s federated catalog and database search tool. These examples demonstrate how unhelpful branded names can be to users. The third example of the use of “UIC” within the name of a resource is the library’s catalog, UICCAT. The term is a contraction of “UIC Catalog” and is similar to catalog names at other academic libraries (for example, AUBIECat at Au- burn University and MadCat at the Uni- versity of Wisconsin–Madison). The pro- nunciation of UICCAT by library staff is “u-i-c-cat,” but the team heard several test participants refer to it as “u-cat,” which suggests that users might not discern the branding. Indeed, the UIC acronym is ubiquitous around campus. The “CAT” part is not especially well understood either. Users do not necessarily make the link between the three le ers and the word “catalog.” One participant renamed the card “UICCAT catalog search” and suggested a aching an explanatory note, speculating that non-UIC users will not know what function UICCAT serves. The SUNY-Buffalo study revealed that users grasp the concept of a catalog, but two- thirds do not use the term, opting instead for descriptive phrases.19 One of the UIC participants likewise asked whether the library might simply rename UICCAT as “find a book” and another participant wondered why the library does not sim- ply name the service or resource for the actual function (“Catalog Search” instead of “UICCAT”). UICCAT as a brand name was not as useful to some test participants as expressions that more explicitly indi- cate what the actual service or function may be. The card sort test team did not antici- pate any particular, strong, or consistent feedback on the use of the assorted brand names on the library’s Web site at the outset of the testing. The branded cards accounted for only 5 of the 93 cards overall (5%). With such a large number, there were many other possible trends that could have emerged. Moreover, none of the branded resources was especially http:phrases.19 248 College & Research Libraries May 2008 new: each had been in place for more than two years, and one (InfoqUIC) for over ten years. It was striking, then, how many of the participants raised concerns with the branded names. Conclusions Very real concerns about establishing the value of library services to users and to the parent institution drive branding decisions. In times when tight budgets result in careful assessment of the efficacy and efficiency of services, a situation in which users do not know what a service is because of the name is problematic. The UIC Library’s card sort usability study re- vealed several weaknesses in the library’s branding efforts that provide guidelines for future branding initiatives: • Use descriptive language. Use of generic terms and library jargon within brand names is not helpful. For example, “info” (or “information”) is so general that it might apply to almost everything at the library. Abbreviating jargon to yield a term like “ERes” (for electronic reserves) further sacrifices clarity for the sake of brevity on the library’s Web site. Naming a service with explicit descriptive language that the user will also under- stand—even if brevity is lost—helps the user identify a library service. • Use distinct names. Use of the same term (“UIC” in this case) within multiple brand names muddles the distinctions be- tween the various resources and services. Users may think that the resources are related (and so they are, by virtue of the library providing them) when they may not help the user in similar ways at all. • Provide marketing and instruc- tion for services. Branding alone is not enough. The names of the services at the UIC Library may be confusing to the users, but the team was le wondering whether marketing of the brands would have ameliorated that state. Indeed, it is supposed that some awareness-raising and instruction may always be required in the use of library services and re- sources. If efforts other than branding are re- quired, this last point brings up larger questions about why libraries feel it is necessary to communicate their respon- sibility for services when they have user loyalty and whether there might be a bet- ter way to impress this upon users. There are other methods to indicate ownership, mostly visual. Employing standard col- ors, watermarks, and graphic logos are alternatives that may be effective. Though card sorting highlighted the limitations of the branding done at UIC, it had several positive outcomes. Partici- pants were very curious about the labels they did not recognize, and the debriefing gave the test team an opportunity to edu- cate them about the library. The feedback was constructive, not simply critical, and the card sort method allowed users to propose terminology that makes sense to them, giving the commi ee a good start in the Web site redesign. Of course, it was expected that the card sort usability test would inform the architecture and terminology of the UIC Library’s new Web site; that it also provided an analysis of branding was a surprise and suggests a new application of such testing for the future. The test results and comments indicate that subjects do use a number of library services and do value the role of libraries on campus. Notes 1. Susan Augustine and Courtney Greene, “Discovering how Students Search a Library Web Site: A Usability Case Study,” College & Research Libraries 63, no. 4 (July 2002): 354–65. 2. Steve Brantley, Annie Armstrong, and Krystal M. Lewis, “Usability Testing of a Customiz- able Library Web Portal,” College & Research Libraries 67, no. 2 (Mar. 2006): 146–63. 3. Nicole Hennig, “Card Sorting Usability Tests of the MIT Libraries’ Web Site: Categories from the User’s Point of View,” in Usability Assessment of Library-Related Web Sites: Methods and Case Studies, LITA Guide #7, ed. Nicole Campbell, 88–99 (Chicago: LITA, 2001). Using Card Sorting to Evaluate Branding in an Academic Library’s Web Site 249 4. Angi Faiks and Nancy Hyland. “Gaining User Insight: A Case Study Illustrating the Card Sort Technique,” College & Research Libraries 61, no. 4 (July 2000): 349–57. 5. Brenda Ba leson and Jane Weintrop, University Libraries Website Nomenclature Test using the Card Sort Method: Summary Report Presented to the University Libraries Web Access Team State University of New York at Buffalo (Spring 2000). Available online from www.jkup.net/BuffaloNomen- clatureTest-Spr2000.rtf. [Accessed 3 July 2006]. 6. Ruth Dickstein and Vicki Mills, “Usability Testing at the University of Arizona Library: How to Let the Users in on the Design,” Information Technology & Libraries 19, no. 3 (Sept. 2000): 144–51. 7. Vicky Duncan and Darlene M. Fichter, “What Words and Where? Applying Usability Test- ing Techniques to Name a New Live Reference Service,” Journal of the Medical Library Association 92, no. 2 (Apr. 2004): 218–25. 8. Jennifer Rowley, “Online Branding,” Online Information Review 28, no. 2 (2004): 131–38. 9. “Building Brand Loyalty: Using Your Library Web Site to Promote Return Visits. Sugges- tions made by John Lubans at the 2000 Illinois Library Association Conference,” Illinois Library Association Reporter 18, no. 6 (Dec. 2000): 7–8. 10. “Naming Conventions,” LISWiki. Available online from h p://liswiki.org/wiki/Nam- ing_conventions. [Accessed 5 July 2006]. 11. Jakob Nielsen, “Card Sorting: How Many Users to Test,” Alertbox: Current Issues in Web Usability, July 19, 2004. Available online from www.useit.com/alertbox/20040719.html. [Accessed 22 November 2006]. 12. Tom Tullis and Larry Wood, “How Many Users Are Enough for a Card-Sorting Study?” Paper, Usability Professionals Association (UPA) 2004 Conference, Minneapolis, Minn., June 7–11, 2004. Available online from h p://home.comcast.net/%7Etomtullis/publications/ UPA2004CardSorting.pdf. [Accessed 22 November 2006]. 13. Nielsen, “Card Sorting: How Many Users to Test.” 14. Hennig, “Card Sorting Usability Tests of the MIT Libraries’ Web Site,” 91–92. 15. Nielsen, “Card Sorting: How Many Users to Test.” 16. A full analysis of the card sort results will be published by the authors in a separate, forthcoming paper. 17. The authors have chosen “resources and services” to refer to the branded items while recognizing the vagueness of the terminology. One card sort participant noted that everything in the library is a resource. The discard results in table 1, meanwhile, demonstrate the meaningless- ness of that label. MIT and SUNY-Buffalo also found that participants didn’t recognize the terms “resource” or “electronic resource.” 18. Ba leson and Weintrop, University Libraries Website Nomenclature Test using the Card Sort Method, 2. The authors do not indicate whether “course reserves” are electronic or physical in format. 19. Ibid. www.useit.com/alertbox/20040719.html www.jkup.net/BuffaloNomen 5 10 15 20 25 30 35 40 45 250 College & Research Libraries May 2008 APPENDIX A Card Sort Labels 1 access to other libraries 2 alerts 3 articles 4 ask a librarian audio/visual equipment 6 audio/visual materials 7 a–z list of article databases 8 a–z list of electronic journals, newspapers, and magazines 9 books calendar 11 catalog search 12 circulation policies 13 citation style guides 14 comments and suggestions contact the library 16 copyright policies and procedures 17 course reserves 18 digital collections 19 disability services document delivery 21 electronic journals, newspapers, and magazines 22 electronic reference collection 23 electronic reserves 24 electronic resources ERes@UIC 26 exhibits 27 FAQ 28 floor plans 29 give to the library glossary of library terms 31 government documents 32 help 33 hours 34 how to find articles how to print 36 InfoQUIC 37 information for non-UIC library users 38 interlibrary loan 39 i-share catalog jobs in the library 41 laptops 42 library collections 43 library instruction and workshops 44 library news features Library of the Health Sciences–Chicago 46 Library of the Health Sciences–Peoria 47 Library of the Health Sciences– Rockford 48 Library of the Health Sciences–Urbana 49 locations and addresses 50 maps collection 51 microfilm/microfiche 52 mission statement 53 my account 54 MyILL@UIC 55 National Network of Medical Libraries Greater Midwest Region 56 new books 57 off-campus access to library databases 58 online tutorials for research 59 other library catalogs 60 page and hold materials 61 patents and patent information 62 PDAs 63 phone numbers 64 photocopying 65 photography collections 66 qUICsearch 67 rare books & manuscripts 68 reference books 69 reference desk hours 70 refworks (citation management) 71 renew books 72 request materials from another library 73 research assistance 74 research consultation 75 resources 76 Richard J. Daley Library 77 scanning and photographic services 78 Science Library 79 search the Web site 80 services 81 site map 82 special collections 83 staff directory 84 study rooms 85 subject guides 86 subject list of article databases 87 subject list of electronic journals, news- papers, and magazines 88 suggest a purchase 89 tours & orientation 90 UICCAT 91 university archives 92 virtual tour of the library 93 wireless access