Presenting a model for the structure and content of a university WWW site

by lain Middleton, Mike McConnell and Grant Davidson
Journal of Information Science, ISSN: 0165-5515, Volume 25 Number 3 pp. 219-227

Abstract

This article proposes a model for the structure and content of a University WWW site, by a process of:

Presenting the content in a manner suited to the characteristics and information needs of the target user groups.

Through an analysis of existing UKHEI web sites and an extensive literature review, a case is made for a non-static approach to web design which identifies users and makes extensive use of themes to promote currently relevant information at the top level. The authors propose a link-rich environment which does not rely on user categorisation and exploration of long sequences of links, and is not constrained by traditional boundaries between departments. The authors contend that such an approach is yet to be adopted in UKHEI, but is beginning to occur at HEIs in the US.

About the authors

Iain Middleton is a research assistant in the School of Information & Media. He has been responsible for the development of pedagogical tools and information resources on the WWW for three years, developing sites which are used both internally across the university and worldwide by educationalists and business alike. He has a background in Human Computer Interaction with experience of graphic interface design. Iain specialises in user support and has published widely on the subject. He is on the review board for the Internet Engineering Task Force's (IETF-TERENA) deliverable on Online User Support and is a team leader on the Universities and Colleges Information Systems Association Software Group (UCISA-SG) Web Tools evaluation group.

Mike McConnell is an Educational Development Officer for the Robert Gordon University's Centre for Learning and Assessment, in charge of learning technology implementation and support across the University. Mike has developed a range of web-based materials for teaching and learning, research, and business and has published on information science research, and education and technology. Mike is a member of the Institute of Information Scientists; the Association for Learning Technology; the Committee of Scottish Higher Education Principals sub-group on Teaching & Learning, and is the Aberdeen Metropolitan Area Network representative for TALiSMAN (Teaching and Learning in Scottish Metropolitan Area Networks).

Grant Davidson is the computer technician for the Robert Gordon University's School of Public Administration & Law and School of Information and Media. He has extensive experience of Internet design, being the site developer for the School of Public Administration and Law, Centre for International Labour Market Studies, the Interface Between Politics and Business Newsletter and European Union Interest Representation.

1.0 Introduction

UK Higher Education is experiencing dramatic changes, reflecting both shifts in society, and government policy. Widening access to education, the increasingly mobile employment market, developing technology, and the growing emphasis on continuing professional development (CPD) and lifelong learning (LLL) all mean that there are new opportunities for those institutions that grasp them, and very real threats for those that do not.

The World Wide Web (WWW) has had an immeasurable impact and will continue to grow: demonstrably it has the power to open up new frontiers and to bring down established entities. New companies have been developed from scratch, successfully exploiting the Internet, while others in high-competition sectors have folded as a direct result [1]. Within the UK, national initiatives aimed at enhancing the technological and information literacy of the nation, and moves to connect schools, libraries and other public spaces to the WWW, suggest that society is increasingly becoming technically literate. Organisations that embrace the WWW and that attempt to exploit to its best advantage, are the ones more likely succeed in the future.

However, the question of how best to exploit the WWW in Higher Education remains a problematic one. It is becoming accepted that it is somehow important, but defining this importance and what to do about it is less well understood. There is no direct comparison with other longer established media, and efforts to date have been built largely on enthusiasm and "best guesses" as to what should be done. WWW development must have a direction, purpose and strategy if it is to succeed. For businesses, the objectives of having a web site are fairly transparent; in Higher Education, however, a heterogeneous audience dictates that these goals are manifold and can become confused.

This paper outlines the major issues which need to be addressed in developing a WWW site for a higher education institution by answering the following questions:

It then discusses the various essential elements of such a web site, why they are important and how they should be presented. (For a discussion of the importance of understanding the "who" and the "what" see Smith, Newman & Parks[2])

2.0 Why have a WWW site?

As the Internet has developed and matured, its potential in a number of areas has been progressively realised:

A university has an interest in all of these: the WWW can be genuinely useful for research, teaching and communication among staff and students, and it can serve as a useful tool for advertising the institution. As such, it does not necessarily do anything new; what it does provide, however, is the potential for performing traditional tasks better, by enhancing speed, convenience and accessibility.

What is placed on the WWW (unless specifically dictated otherwise) is visible to the outside world, and thus represents the university to a market potentially greater than any other medium. The WWW is beginning to replace (at minimum, to augment [3]) more traditional forms of representation to the extent that, "The web site will become an increasingly important 'shop window' for the world and may in time become the primary source of information for those outside the institution"[4] and has the advantage of flexibility and currency. Evidently the better an institution presents itself in any medium, the more successful it will be at attracting students, academics, business or whomever it wishes to attract. A successful WWW site will augment traditional forms of advertising and promotion. For many people the WWW provides the most convenient means for locating information about an organisation - but only if that organisation or information is represented there.

3.0 Whom does the site serve?

Having defined why a web site is important, an institution must understand who its audience is before it can decide how best to serve it. Two categories of user are immediately apparent:

both of which have different, disparate compositions, all with distinct needs. Some generalisations, however, are valid:

3.1 Internal users

Internal users are a captive market, in this case comprising of staff and students. As such there is no need as such to attract them; rather, as existing "customers", an institution should be looking at methods of best serving them. The information bias in this case must be towards utility, facilitation and enhancement of the working environment - perhaps even entertainment. Such information might include telephone directories, downloadable forms and documents, databases, library catalogues, timetables, booking forms, lecture notes, educational materials, news/discussion fora, etcetera.

3.2 External users

External users are target markets - in many cases, these will have no better reason to look at an institution's site than any other UKHEI's; often less so. There is thus a pressing need to attract this audience's attention as soon as possible and then hold it. Information bias must then be towards promotion. This however, should not mean a deluge of advertising; it means providing good reasons for doing business with an institution and enabling users to do so as easily as possible. Indeed, utility may be good promotion to some user groups - a user is likely to be more inclined to return to a site if it provides a useful service than if they are presented with nothing more than slick advertising[5] (often known as "brochureware"). Such information might include course/prospectus information, local information, mission statements, frequently asked questions, travel and accommodation information, press releases, contacts, job openings, downloadable forms, entertainment, feedback information, etcetera.

Additionally an institution can provide its audience with useful generic or discipline-based resources that enhance its reputation and ensure that users return - for example, the UKHEI site map at Wolverhampton[6] provides access to all UK HEI web sites through an interactive graphical map. This has become known as a default starting point for locating UKHEIs on the web. Such a service is not overtly promotional per se, but may nonetheless serve a promotional purpose by illustrating to the audience the high quality facilities an institution offers to its staff/students[7] and its place at the cutting edge of the field.

External users are, of course, a more heterogeneous group than internal users. While it is not possible to exactly state who makes use of any given university web site, it is highly probable that the external audience will include all of the following groups on a regular basis:

4.0 What information do these individuals need?

Many disparate groups of users and different types of information need are suggested above. However, institutions may define content more clearly if they think instead of which types of information they want to ensure that users have. After all, the information provided should, directly or indirectly, benefit the institution. What kinds of information will bring benefit to the institution?

5.0 How can this need best be met?

All of the types of information identified may be used to attract and satisfy an audience. None of these types of information is by definition static: they are constantly changing according to circumstances and the needs of the institution and its potential customers. The WWW site must reflect this immediately without users having to 'dig' in order to satisfy their queries: otherwise it has ceased to provide information, it only stores information. The ethos of the web site must be to reflect what is current. It must also reflect users' needs, without encumbering any set of users with information which is only relevant to another, distinct set of users. For this reason it is recommended that web sites consist of distinct sections relevant to "internal" and "external" users respectively. This does not require mutual exclusivity, it implies only the need to ensure that users entering the site via the "front door", who are likely to be unaware of how the site is structured, are immediately and easily able to find out about the institution, while the internal page concentrates on keeping staff and students informed of pertinent information and providing access to online tools. However, users should not be forced to make the decision as to the category in which their interest lies before they can utilise the site. That is, the default home page should give the option to specialise, along with a range of information of which the institution wants users to be aware: the aim is to communicate important information to users, not to narrowly categorise them.

The primary function of a web site is to provide information, and this is what should be presented at the top level, rather than simply providing links to where information may potentially be found. For this reason, it is not enough to provide links to departments: these links may or may not contain information which the user wants, but there is no way of telling until the link has been explored. This approach also encourages a "many isolated parts" model of development in which individual departments take varying approaches to style, navigation and content, often duplicating effort and ultimately failing to address the users' need which is not necessarily specific to a department. Such fragmentation detracts from the ideal model of the web site as the "one stop shop" for information, and leads to a divided and confusing viewing experience for the user[10]. While access should nevertheless be given to departmental lists if chosen by the user, these alone are an inadequate structure for web information.

Another approach common in UKHEI web site design is to divide links into two distinct sections of information FOR a certain group of users and information ABOUT the university or its services. This however, (as with the 'department' approach) is an attempt to narrowly categorise users and their needs at an early stage, and given the heterogeneity of the WWW, is impracticable. A suitable analogy would be a multiple choice questionnaire: when completing it, respondents often feel that none of the given options applies exactly to them. In the same way, UKHEIs cannot reliably categorise users or their requirements in advance. For example, the University of Strathclyde's home page[11] immediately categorises users into 'Staff', 'Students' and 'Visitors'. This implies a homogeneity among each category, (already illustrated above to be flawed and also) unsafe in the light of a substantial body of HCI research[12].

Given the modern environment of Higher Education with great diversity among current and future students, and new modes of work (e.g. telecommuting, Open and Distance Learning), it is wrong for designers to narrowly categorise users - as is often the case. It is unlikely that visitors will be able to classify their information requirements into the restrictive groupings evident at so many university sites.

5.1 Internally relevant information

Internally relevant information should be concentrated in a separate section devoted to staff and student needs, otherwise the dual purposes of the site may interfere. Allowing access by external users to this section is desirable in order to further capitalise on the online facilities available to staff and students by actively demonstrating their extent and quality; however, there are almost certainly areas that will need to be protected from external access (for example, student email addresses). Institutions should be guided by common-sense principles of privacy, confidentiality and competitive advantage on this issue.

By grouping internal information in a distinct area, staff and students will avoid the need to view promotionally oriented material in the course of their work. Additionally, there can be less emphasis (and therefore time spent) on making the site appear aesthetically pleasing, and more on utility. The institution can afford to assume more of its internal audience, since it is captive and finite, and can thus provide for it more confidently.

Internally provided information on the web site should highlight the kind of information which institutions want their staff and students to be aware of i.e. an emphasis on current awareness. This information should not require to be sought out: as with any bulletin or important announcement, the information should be presented without being requested. Main, or "start" pages should ideally have a bias towards news items - such information as the institution wants to ensure that users will read. Other information should be provided with the intent of eliminating bureaucracy: enabling staff to carry out their tasks without having to locate and ask a specific person "what to do" in a given situation or requesting and waiting for standard forms to arrive in the mail. Common procedures such as research grant applications might be included in this category.

In summary, an "internal" information page should:

5.2 Externally relevant information

Externally relevant information needs the most visibility from the default/home page. As this is the point of access which will be used specifically by those who possess the least prior knowledge of the institution it should have maximum accessibility: users' patience, preconceptions and modem speeds will not necessarily match those of the institution. Choice of syntax is imperative; for example, the distinction between 'courses' and 'prospectus', may seem a fine one, but might prove a significant difference for foreign students. A case in point is the Robert Gordon University[13] where local terminology results in a home page with links to both "Schools" and "Departments", without additional explanation. These links are ambiguous in meaning, as it likely that external users will not be aware that at this institution, "Schools" are the academic departments, while "Departments" are the service, support and administrative departments. The word "Schools", in isolation, is particularly open to interpretation.

Similarly, access is more important than rigid structure: WWW users have little concept or understanding of the hierarchy of the pages they are browsing through: it is not their interest . Institutions should give top level access to information wherever possible, or failing that, categorise, characterise or conceptualise it meaningfully into sections according to purpose.

A substantial amount of recent research indicates that users will not explore very far down complex hierarchies, and prefer a shallow structure to a deep one. McMurdo[14] advocates the use of many links rather than fewer, quoting Lynch and Horton[15], who suggest that 'less-common Gopher menus...provide classic examples of overly nested menus, from which users might have to make a number of selections before encountering any content at all... interface studies show that users prefer menus with a minimum of five to seven links, and prefer a few very dense screens of choices over many layers of simplified menus'. Other researchers concur:

Institutions cannot afford to assume how users will conceptualise or wish to use their site[19], so the fewer decisions made for them the better. Users will go for information which has a label that they recognise: designers are exhorted to design for "recognition rather than recall: Make objects, actions, and options visible. The user should not have to remember information from one part of the dialogue to another"[20]. Users are scanning for information of relevance, and pick up on the most relevant looking keywords. The shallower the hierarchy, the sooner users will find such keywords. These are basic tenets of HCI, and are further confirmed by research into how users "read" (scan for) information on the WWW[21]. Mutual self interest may be served by providing the information which users want to see, for example, clearing places, application forms, newsworthy items.

In summary, an external page should:

5.3 Currency of information

There is one final factor to take into account in providing information (of any sort): information seldom improves with age. Both internal and external users need to have timely and current information. It is a characteristic of users of the WWW that they browse with a short attention span: currency makes information more relevant and thus holds the user's attention for longer. This serves the information provider as well as the user: it is in the interests of both that information is timely and pertinent.

For this reason, top level pages should not be static: while their essential functionality and structure must be constant (i.e. standard navigational and functional links and their overall purpose), they should be reflecting current themes wherever possible. This is valid both for the internal and external user: while there are obviously things an institution should promote to potential customers, there are also themes, ideas, events and technologies of which internal users should be aware. This drive towards currency must be implicit throughout the top level pages: designers should not rely on users seeking it out. News should permeate the entire web site, rather than being segregated into a separate "News" section: this will only attract people who have a predisposition towards reading news - it is archiving, not communicating. For example, the meningitis health scare at UK Universities in 1996 prompted some university web sites to place an advisory/reassurance headline on their front page: placing it in a "news" section would have had little effect. Nevertheless, news should also be collated into a news page for those interested in what the institution has been achieving lately: past news items need to be retained and made accessible for future reference.

In summary, important news will serve its purpose much more effectively when it receives this strategic deployment. It will also mark the institution's site as a dynamic resource, worth revisiting.

5.4 Examples in the real world

An example of an HEI site which does take a dynamic, thematic approach at the highest level is Cornell University[22] in the US. Here the home page includes the standard, static informational, navigational and thematic links, but it also has dynamic features: links to the latest Cornell-related news, a "featured site" which changes on a regular basis and an attractively photographed picture of a featured building on campus, with relevant link, which is also changed regularly:

Image of Cornell home page

It also features interactivity, with the ability to search for 3 different types of information directly from the home page (staff, students and departments). It is an engaging home page, rich in information and links, which conveys what Cornell wishes the user to know without requiring them to seek it out.

However, even the Cornell home page suffers from the common problem of over-large file size, weighing in at nearly 100k, resulting in prolonged download times for users without high-speed access. It is not alone in this weakness: Greenwich's site[23] is nearing 150k (indicating a download time of over 80 seconds on a 14.4k modem[24]), offering only three links to further information, similar to Strathclyde's approximately 85k home page[11]with four links. Without user access to a high speed connection, none of these sites comply with the limits for acceptable download times outlined by Nielsen[25], resulting in potential frustration and/or loss of interest by the user. In a survey of a sample of 38 UK tertiary education web sites (see Appendix 1), average home page size was found to be approximately 39k (with a spread from 2.3k to 145k): at the very upper limits for comfortable browsing with standard equipment. Exuberant, graphical sites with complex animation contrast with dour, text-only pages with a handful of links. Clearly, the desire for complex functionality and attractive graphics needs to be tempered with an appreciation of the technical limitations of the institution's and users' equipment.

If web sites are to use bandwidth-intensive techniques, they must be married to practices of good design, structure and access to information, giving maximum benefit to the user in return for their patience and (potentially expensive) online time. A effective site will utilise strong technical/expertise to manage the variety of information content into a coherent whole. This calls for the institution to centrally co-ordinate efforts to ensure consistency of design and currency of information: neither task should be attempted in isolation.

6.0 Review

In summary, an institution must satisfy users by providing maximum content and facilitating access to it wherever possible, whilst ensuring that the information provided is what the institution wishes the users to know. This will necessitate the adoption of editorial and administrative policies and structures to ensure control is exercised across the university without stifling creativity. Such controls will necessarily vary from institution to institution, and there is no globally prescriptive model. However, some form of central editorial control will be required for top-level pages in order to identify current themes for promotion and to ensure that a coherent face is presented.

6.1 User needs

Users, both internal and external, will need:

  1. Content - there is no point in accessing a site without it
  2. Accessibility - because 1. is meaningless without it
  3. Relevance - materials must suit the users' needs, as well as the institution's
  4. Currency - current information is more relevant and will ensure that the user will return and/or think better of the institution

If user perceive the above elements, either consciously or unconsciously, then the institution will have greater success at achieving its goals.

6.2 Institutional needs

The institution will need to ensure:

  1. Representation - all elements of the institution should be represented, to provide coherence and clarity and to enable dialogue with potential customers
  2. Adequate Promotion - the ability to use the medium as an effective marketing tool in concert with existing means to attract custom, investment and collaboration. This will require commitment to ongoing editorial and administrative policies, structures and support services
  3. Innovation - space for scholarly use and learning new ways of exploiting the new medium. Everybody within the institution should have the ability/opportunity to feed into the web site, provided sufficient editorial guidelines are in place. This promotes a vibrant web culture that encourages usage - seen by many as the raison d'être of the Internet
  4. Satisfaction Of Users' Needs - access to relevant, current content. As stated above, without this there is no point in having a WWW site
  5. Structure - so information may be effectively edited, administered and deployed, and support effectively targeted

6.3 In Conclusion

User needs are not necessarily the same as the institution's needs, but the institution must satisfy the users' needs in order to satisfy its own.

A university WWW site should therefore be one:

The WWW may still be seen as a novelty by many, or at best, an optional extra; however, the reality is that it is becoming increasingly important as a mainstream medium. The image that HEIs present there will have to be more carefully considered. The competitive and international environment of the contemporary higher education market dictates that institutions need to represent themselves effectively to potential customers and allies; as well as providing their staff and students with the means to keep up with their peers, and advances in thinking and technology. The WWW has the potential to achieve all this: it cannot continue to be regarded as something best left to the technical /computer 'experts'.[26]

Instead, representation of the whole institution demands a co-ordinated approach identifying target users and placing their needs at the centre of design. Today, this includes (at least theoretically) all staff and students of an HEI. Such stakeholders need to be equipped with the necessary information and skills to take advantage of the WWW. If UKHEIs develop purpose, strategy and - when justified by these - committed resourcing, they have the opportunity to capitalise on the benefits of the only medium of communication which is open 24 hours per day, 365 days per year.

References

[1] Eoin Licken, Free news providers must woo advertisers to survive, The Irish Times, Friday March 6, 1998. Electronic document, available online at:
http://www.irish-times.com/irish-times/paper/1998/0306/tech6.html Visited: 13/8/98

[2] P.A. Smith, I.A. Newman And L.M. Parks Virtual hierarchies and virtual networks: some lessons from hypermedia usability research applied to the World Wide Web, International Journal of Human-Computer Studies (1997) 47(1), P128, Available online at: http://ijhcs.open.ac.uk/smith/smith.html Visited: 13/8/98

[3] Kelly Russel, Ole Miss using the Internet, info booths in malls to attract students, Mississippi Business Journal, 10/21/96, Vol. 18 Issue 43, p20, 1/2p

[4] Andrew Aird, The Future of Goldsmiths' Website, Lecture: 19th Nov 1997. Available online at: http://www.gold.ac.uk/webmanager/lecture.htm Visited: 13/8/98

[5] M. Phillips and S.Rasberry, Marketing Without Advertising, 2nd edition, Nolo Press 1997. Electronic document available online at: http://www.nolo.com/mwad/1b.html Visited: 13/8/98

[6] University of Wolverhampton: Online document available at http://www.scit.wlv.ac.uk/ukinfo/uk.map.html Visited: 13/8/98

[7] Wayne Rash Jr., It's Not Just The Three R's Anymore, It's The Three W's: The World Wide Web, Communications Week, October 14, 1996, p126

[8]Neil Orman, High technology recruiting competition breeds secrets, Austin Business Journal, 2/23/96, Vol. 16 Issue 1, p24, 1/2p

[9] Michael Stoner & G. Phillip Cartwright, Alumni, Public Relations, Admissions--And Technology, Change, MAY/JUNE 1997, Vol. 29 Issue 3, p50

[10] For a good example of this, browse through the academic Schools at the Robert Gordon University, accessible from: http://www.rgu.ac.uk/topic/theuni/schools.htm Visited: 13/8/98

[11] University of Strathclyde, http://www.strath.ac.uk/ Visited: 4/8/98

[12] (For example) Committee on Human Factors, Commission on Behavioural and Social Sciences and Education, National Research Council: Research Needs for Human Factors, National Academy Press, Washington D.C. 1983. Available online at: http://www.ul.cs.cmu.edu/books/research/needs001.htm Visited: 4/8/98

[13] The Robert Gordon University: http://www.rgu.ac.uk Visited: 13/8/98

[14] George McMurdo, Evaluating Web information and design, Journal of Information Science, 24 (3) 1998, pp.192-204. Available online at: http://jimmy.qmced.ac.uk/~jisew/ewv24n3/ Visited: 13/8/98

[15] P.J. Lynch and S. Horton, The Yale Style Manual (Yale University, 1997). Online at: http://info.med.yale.edu/caim/manual/ Visited: 13/8/98

[16] Peter Morville, Lou Rosenfeld, Information Architecture for the World Wide Web, Designing Navigation Systems, O'Reilly & Associates, 1998. Also available online in: Designing Navigation Systems, Web Review Feb. 20, 1998 at URL: http://webreview.com/wr/pub/98/02/20/arch/index.html Visited: 13/8/98

[17] K. Norman, The Psychology of Menu Selection: Designing Cognitive Control at the Human/Computer Interface, Ablex, Norwood, NJ (1991).

[18] Ben Shneiderman: Designing information-abundant web sites: issues and recommendations, International Journal of Human-Computer Studies, 47(1) (1997). Available online at http://ijhcs.open.ac.uk/shneiderman/shneiderman.html Visited: 13/8/98

[19] L. Hardman, Evaluating the usability of the Glasgow Online hypertext, Hypermedia,
1 (1), 34-63. (1989)

[20] Keith Instone: Usability Heuristics for the Web, Web Review, October 1997. Available online at http://webreview.com/97/10/10/usability/index.html Visited: 13/8/98

[21] Jakob Nielson, How people read on the Web, The Alertbox: Current Issues in Web Usability, October 1997. Available online at: http://www.useit.com/alertbox/9710a.html Visited: 13/8/98

[22] Cornell University: http://www.cornell.edu Visited: Visited: 13/8/98

[23] University of Greenwich: http://www.gre.ac.uk/index.shtml Visited: 13/8/98

[24] Data from "Doctor HTML": http://www2.imagiware.com/RxHTML/htdocs/single.html Visited: 12/6/98

[25]Jakob Nielsen, Usability Engineering, AP Professional, Boston, MA, 1994. Cited passage also available online at: http://www.useit.com/papers/responsetime.html Visited: Visited: 13/8/98

[26] "Bottom Line: IT organisations should not own content, but should focus on application and 'net management policy development and content management infrastructure elements." excerpt from: META Group's Advanced Information Management Strategies, 'Taming End User Discontent' a META Group's weekly META Fax, online at:
http://www.metagroup.com/newweb.nsf/e06e5d7fe70d8a5d852563310072b59d/ Visited: 13/8/98