User behaviour – what the librarian should know
By Margaret Adolphus
Ten years ago, the term "user behaviour" in a library context would have conjured up images of things forbidden by the library's code of conduct, such as talking and eating.
Now librarians are more concerned that their patrons should feel at home, and provide not only comfortable sofas and (in some cases) refreshments, but different areas dedicated to different types of study – such as quiet areas for individual study and more open, discussion areas for groups.
But for most librarians, the term "user behaviour" is most interestingly applied in the virtual area, particularly in how users seek, access, evaluate and use information.
User behaviour studies
There is no doubt that the move from a print-based to an electronic environment has given rise to a huge change in the way that people discover, access, use, and communicate information – who would have imagined tweeting about a useful article 20 years ago for example?
Mohamed and Hassan (2008, p. 419) define "user behaviour" as the way that people think, perceive, behave and feel about information retrieval systems when they interact with a software interface.
Not surprisingly, there have been a large number of research studies over the last 10 years which look at they way users interact with information. Some of which are outlined below.
The digital information seeker
Perhaps the most useful is a synthesis compiled by the UK's Joint Information Systems Committee (JISC): The digital information seeker: Findings from selected OCLC, RIN and JISC user behaviour projects (Connaway and Dickey, 2010).
This report takes 12 studies from the UK and the USA, originated by key players in the area: JISC itself, the Centre for Information Behaviour and the Evaluation of Research (CIBER) at University College London, the UK's Research Information Network, and the OCLC.
The studies span the years 2005-2009, and cover general perceptions of libraries (including the non-academic user), the behaviour of undergraduates, graduates and faculty, catalogues, e-journals, e-books, resource discovery, and virtual reference services. In other words, a fair range of academic library user types, products, tools and services.
The real strength of the report lies in its meta-analysis: two researchers analysed the studies, and compared their findings. Thus there is triangulation not only of one study against others, but also of research methods (studies employ a variety of methods, qualitative and quantitative). It is also honest in admitting that some of the evidence is contradictory.
Ithaca is a non-profit organization based in the USA which researches use of technology in higher education, and regularly conducts a survey of North American faculty. The most recent, in 2009, surveyed over 3,000 people, and dedicated a chapter to discovery and the evolving role of the library (Schonfeld and Housewright, 2009).
Both the Ithaka study and the digital information seeker report cover impressive numbers, but they originate from the USA and the UK (even though at least two of the latter's studies had participants from Australia, Singapore and India).
Emerald has published a couple of pieces of research from further afield: three scholars from Wuhan University in China surveyed the use of electronic resources in Wuhan universities (Zhang et al., 2011), while in Egypt, Mohamed and Hassan (2008) examine scholars' use of federated search engines.
Several studies are ongoing. CIBER, the pioneer of deep log analysis (see below), is looking at usage, information seeking and searching behaviour on Europeana, a multilingual and multimedia online collection of digitized objects from cultural and heritage collections (Nicholas et al., 2010).
The University of Oxford and OCLC Research are carrying out a short-term study of students in the transitional stage between senior school and university: Visitors and Residents: What Motivates Engagement with the Digital Information Environment?
Most of the studies referred to above use a variety of methods, both qualitative and quantitative such as surveys, in-depth interviews and focus groups. A couple also use deep log analysis, which analyses raw transaction logs to provide a lot of very rich data on how the user behaves, for example, when, who, and how long do they spend on the site, which can then be statistically analysed.
What do the studies tell us about user behaviour?
Here are some of the findings from the studies mentioned above.
- One study (De Rosa, 2005, quoted in Connaway and Dickey, 2010, p.6) reported that users associate libraries with books, and do not think of using them to access electronic resources. However, an American Library Association report The State of America's Libraries (ALA, 2010) maintains that academic libraries are experiencing increased use, both physical and virtual (p. 19), with 20.3 million visits during a typical week in 2008.
- Several of the studies reported in the Digital Information Seeking Report indicate (Connaway and Dickey, 2010, p. 38) that users prefer digital over print content, wanting more digitization (e.g. of older literature, images, etc.). One study on the information behaviour of the future researcher, from CIBER (Connaway and Dickey, 2010, p. 15) predicts that by 2017, the information environment will be that of a "unified web culture", with e-books becoming more and more important, and mass digitization of print books.
- There is a widespread use of search engines such as Google for information seeking, which are often preferred to library-based search systems, such as the online catalogue, federated search, electronic databases, etc. Google is often used to access e-journal content, and can account for a third of traffic, according to a study on e-journals carried out in 2009 by the Research Information Network (Connaway and Dickey, 2010, p. 20).The publisher Springer noted that in the first quarter of 2010, 43 per cent of its article downloads came from search engines (Springer, 2010). Typing a keyword into a single search box is a popular search behaviour, while there is little support for the advanced search options favoured by OPACs (Connaway and Dickey, p. 44). The CIBER Europeana study also found that users favoured that site's big search box found on its home page, and rarely used the advanced search option (Nicholas et al., 2010, p. 135).
Figure 1. The Europeana website (note the social software touch with the "People are currently thinking about" feature)
- However, evidence reported in the Digital Information Seeking Report would seem to indicate that in a scholarly context, all sorts of electronic resources are used. For example, a 2006 Research Information Network study (Connaway and Dickey, 2010, p. 12) found that users generally value a variety of tools, and that the most utilized resources were general search engines, library portals and catalogues, specialist search engines, and subject specific gateways. Another study (Hampton-Reeves et al., 2009, quoted in Connoway and Dickey, 2010, p. 22) found a high degree of awareness of the qualitative difference between the basic types of Internet content retrieved by search engines and more formal resources, and that many preferred the library's catalogue. Thus pre-conceptions about the Google generation may not always be supported by the evidence.
- The Ithaka report (Schonfeld and Housewright, 2010), while confirming the importance of the search engine as a gateway to knowledge, also showed that faculty would often use discipline-specific resources as a first port of call (p. 3). The latter might be licensed by the library, but faculty would nevertheless bypass more obvious library service points in favour of a networked resource. In response to a survey question which asked them to rate the importance of the library building, the library catalogue, a general search engine or a specific research resource, there was a general bias towards the latter. In other words, the library's "added value" was as a purchaser and negotiator, rather than a gateway to information.
- Whatever the resource, users prefer to have desktop access and use the virtual more than the physical library (Connaway and Dickey, p. 33). 24/7 access is important, and one librarian from the University Library of Utrecht in The Netherlands described the "ultramodern user" as thinking of computers and other virtual devices as another limb (Elsevier, 2011).
- In the online environment, literature searching has changed dramatically, with users "bouncing" through content, scanning rather than reading in depth, demanding "instant gratification", according to a CIBER study on the information behaviour of the future researcher (Connaway and Dickey, p. 15).
- Users are not just looking to electronic systems that help them discover resources: they want immediate and seamless access. "Discovery to delivery" is now the issue, and users want library systems that support this, and get frustrated when they do not (Connaway and Dickey, 2010, p. 34). Immediate information gratification has been encouraged by search engines, and users become frustrated when they have to pass through further virtual walls requiring password access, etc. Library functionality must also be in place to help them manage results, as well as enhanced catalogue content (in the form of metadata) to help them evaluate (Connaway and Dickey, 2010, p. 4).
- There may be some users who view virtual devices as an additional limb, but their greater digital ability is not matched with increased information literary. The CIBER study on the information behaviour of the future researcher reports that users are not good at evaluating search results, put too much trust in Google, and tend to use natural language searching (Connaway and Dickey, 2010, p. 41). Many studies indicate that students are mostly self taught, although some also indicate a respect for the librarian's role as information broker (Connaway and Dickey, p. 41).
- Disciplinary differences exist, at both student and more advanced researcher level. This is a central conclusion of Connaway and Dickey (2010), but it is also supported by the Ithaka survey (Schonfeld and Housewright, 2010), which analysed faculty according to whether they were scientists, humanists, or social scientists, finding that scientists were least likely to start their research from library specific starting points. Zhang et al. (2011) found that users' jobs correlated with the literature services they needed.
- E-journals were vitally important for research, and difficulty of access, whether to current content or to back issues (Connaway and Dickey, p. 38), caused frustration.
- E-books are definitely part of the information landscape, and are often "discovered" through library catalogues, so that libraries are a key player, according to the JISC National e-Books Observatory Project (Connaway and Dickey, p. 21).
- There is no firm evidence as to the impact social networking (e.g. recommendations from Twitter, social bookmarking, etc.) on research or study patterns and information seeking (Connaway and Dickey, p. 48).
- Although the above picture presents one of the user studying in isolation from the physical library, confident in use of digital resources although not necessarily skilled at evaluating them, human resources are still valued. Studies indicate that librarians will have an important role in the new information environment (Connaway and Dickey, 2010, p. 13), and that users value human resources such as colleagues, peers, family and teachers in their information seeking (p. 39). Virtual reference services (VRS), if users had had a positive experience, were also recommended.
- The CIBER Europeana study reported that users were very keen on multimedia content: they were ten times more likely to select video content when viewing thumbnails (Nicholas et al., 2010, p. 135).
What are the implications for libraries and librarians?
Here are some of the implications of the above findings on user behaviour for librarians.
- The librarian must understand the information seeking behaviour of users, and also that this may differ according to demographic group, age, discipline, and knowledge seeking context. Zhang (2011) suggests offering customized literature services that correspond with users' professional and disciplinary requirements, also, publicizing library services at times of greatest demand (March, September, October and November).
- They must also understand that user behaviour has changed, and that now users require quick access to items they have discovered, and have developed new methods of reading involving scanning.
- Both discovery and delivery systems need to improve. For discovery, high-quality metadata is an important aid to discovery, because it helps the user to know whether or not something is useful without a detail perusal of the resource itself. The user should be able to discover as much as possible from the catalogue entry, and it is particularly important to adequately describe legacy materials (Connaway and Dickey, 2010, p. 49). Resources should include e-journals (including back issues), foreign language materials, e-books, chapters in authored volumes, specialist search engines, and publishers' databases.
- Users require delivery as well as discovery: seamless access to resources, which should be only a click away. Access needs to be less complex (i.e. authentication should be required at one point only, and not every time a new database is entered). Metasearch engines, which help search multiple sources, and which provide additional services, such as authentication, merging and de-duplicating results, go some way towards meeting this need, but it is very important that developers analyse user behaviour when creating software (Mohamed and Hassan, 2008).
- Users, both academic and otherwise, tend to believe that digital is better, so libraries need to provide a greater variety of digital formats and content. E-journals are a particularly strong investment, but libraries should also offer e-books, open source materials, non text-based and multimedia objects, as well as blogs and virtual research environments (Connaway and Dickey, p. 47).
- Library systems need to function more like search engines and services, such as Google and Amazon, with their one-stop-shop search box. The large search box in the culture gateway Europeana proved very popular (Nicholas et al., 2010), with far more people using that than the advanced search option.
- Libraries need to do better at publicity, advertising their resources and services, and their brand, to the communities they serve. For example, if services are available via a network (and can be accessed without going through the library gateway), the library should make it clear that it has purchased these services.
- Information literacy training is important, and the implication of specific disciplinary needs mentioned above is that it should be discipline specific.
- The UK eInformation Group suggests (UKeIG, 2010) that libraries follow the example of search engines, which analyse users' navigational behaviour and use the information to recommend sites. It cites LibraryThing, BibTip, and bX as examples of recommender services based on the borrowing or research habits of users.
Figure 2. BibTip's home page
- Virtual Reference Services (Ask a Librarian etc.), if well run, are much welcomed and users recommend them to peers. Appendix Q of the report Seeking Synchronicity: Evaluating Virtual Reference Services from User, Non-User, and Librarian Perspectives (Radford and Connaway, 2008), provides some best practice guidelines for running a VRS.
Developing user friendly library systems
Much of the above concerns not just the library's human resources, but also its software. Outward-facing library systems (i.e. the ones that patrons, as opposed to uniquely librarians, use), must behave in the way that the user has come to expect: fast and fleet. There are a couple of library systems that do that.
Summon is a discovery service from SerialsSolutions (a Proquest company) which allows the user to search the entire contents of the library – the catalogue, e-journals, print journals, databases, the university repository, etc. – by means of a single search box.
It thus goes beyond federated search, where users have to search individual databases. All content is mapped and normalized, so metadata is enriched, and results are sorted and relevance ranked. It is also easy to integrate with other systems, for example virtual learning environments, and is hosted so can be supported without involving much time from librarians.
Figure 3. Interface of the University of Huddersfield library's Summon tool, showing results for the term "brand management"
Summon relies on cooperation with publishers to aggregate their content; Ex Libris's Primo Central Index works on a similar principle. A cloud service, Primo Central Index also offers one-stop-shop searching, and easy integration into other library systems.
Libraries can offer the modern information seeker what he or she wants, providing they ensure that their services are fast, efficient and resemble the most often used search engines and services in everything except the quality of results.
They also need to market themselves and their brand, so that people are fully aware that the services they are using are paid for by the library, even though they may not be accessing them via the library interface. In the quest to be helpful, the library needs to avoid being unassuming. Tips on marketing your library are available in the "Marketing your library" section of this website.
Librarians also need to avoid being too caught up with the behaviour of today's user, because this will change as new Internet products and services become available. In the virtual world, nothing stays still for very long.
ALA (2010), The State of America's Libraries, American Library Association [accessed 22 June 2011].
Connaway, L.S. and Dickey, T.J. (2010), The Digital Information Seeker: Report of the Findings from Selected OCLC, RIN, and JISC User Behaviour Projects, OCLC Research [accessed 22 June 2011].
Elsevier (2011), "How is user behaviour at your library changing?", Elsevier Library Connect, available at: http://www.elsevier.com/wps/find/librarianshome.librarians/LCN040110 [accessed 22 June 2011].
Mohamed, K.A. and Hassan, A. (2008), "Web usage mining analysis of federated search tools for Egyptian scholars", Program: electronic library and information systems, Vol. 42 No. 4, pp. 418-435.
Nicholas, D., Clark, D. and Rowlands, I. (2010), "Unique perspectives on user behaviour for multimedia content: Europeana case study", paper presented to Online Information 2010, 30 November to 2 December, London, UK, published in Conference Proceedings in association with Information World Review.
Radford, M.L. and Connaway, L.S. (2008), Seeking Synchronicity: Evaluating Virtual Reference Services from User, Non-user, and Librarian Perspectives: IMLS Final Pperformance Report, Report on Grant LG-06-05-0109-05, to Institute of Museum and Library Services, Washington, D.C. Dublin, Ohio: OCLC Online Computer Library Center, available at: http://www.oclc.org/research/activities/synchronicity/reports/20080626-final.pdf [accessed 28 June 2011].
Schonfeld, R.C. and Housewright, R. (2010), Faculty Survey 2009: Key Strategic Insights for Libraries, Publishers, and Societies, Ithaka S+R, available at: http://www.ithaka.org/ithaka-s-r/research/faculty-surveys-2000-2009/Faculty Study 2009.pdf [accessed 28 June 2011].
Springer (2010), "Changing online user behaviour – what do we know so far?", Society Zone, Issue 12, Summer, available at: http://www.springer.com/societies/society+zone?SGWID=0-173202-12-742912-0 [accessed 27 June 2011].
UKeiG (2010), "Mining information-seeking behaviour data to enhance library services", eLucidate, Vol. 7 No. 3, June, UKeiG, available at: http://www.ukeig.org.uk/elucidate/issue/mining-information-seeking-behaviour-data-enhance-library-services [accessed 22 June 2011].
Zhang, L., Ye, P. and Liu, Q. (2011), "A survey of the use of electronic resources at seven universities in Wuhan, China", Program: electronic library and information systems, Vol. 45 No. 1, pp .67-77.