Information professionals have featured strongly in the evaluation of the use of commercial online hosts and online public access catalogues, but not so strongly in the evaluation of the use of websites. This paper describes a piece of research that was conducted on The Times/The Sunday Times websites, to determine the most appropriate methods for evaluating use and to establish what forms of analysis could best be derived. A database of one million subscribers and three months’ worth of logs, constituting 65 million lines of data, were obtained and the data were analysed using parsing techniques and the Statistical Package for the Social Sciences. There were problems associated with the analysis, largely because of the difficulties in establishing the identity of Web users, determining what actually constitutes use and measuring the time spent online.
Men in their 30s were the sites’ most numerous subscribers. The majority of subscribers were foreign and came from commercial organisations. Use varied considerably from hour to hour and day to day and from country to country: in the UK, midweek lunchtimes proved very popular. On average, a user conducted 2.35 sessions over the survey period, spent 15 minutes on a search session and 0.5-2.1 minutes on reading a page. Commercial organisations and Americans were the heaviest users and news pages proved to be the most popular.
Get full access to this article
View all access options for this article.
References
1.
[1] ABC Launches Monthly Audit Summaries 17 May 1999. Available at: www.abc.org.uk/presr116.html (visited 7 August 1999).
2.
[2] D. Backman and J. Rubin , Web Log Analysis - Finding a Recipe for Success (1997). Available at http://www.nwc.com/811/811cn2.html (visited 8 July 1999).
3.
[3] J. Fieber , Browser Caching and Web Log Analysis (1999). Available at: http://ella.slis.indiana.edu/%7Ejfieber/papers/bcwla/bcwla.html (visited 9 September 1999).
4.
[4] A. Gutzman , Analysing Traffic on Your E-commerce Site (1999). Available at: http://www.ecommerce.internet.com/solution/techno_advisor/article/0,1281,9561-186011,00.html (visited 17 October 1999).
5.
[5] B. Krishnamurthy and J. Rexford , Software Issues in Characterising Web Server Logs (1998). Available at: http://www.w3.org/1998/11/05/WC-workshop/Papers/bala2.html (visited 18 September 1999).
6.
[6] P. McGlamery , MAGIC transaction logs as measures of access, use and community , Journal of Academic Librarianship23(6) (1997) 505-510 .
7.
[7] E. Meyer , Web metrics: too much data, too little analysis . In: D. Nicholas and I. Rowlands (eds), Impact and Evaluation of the Internet: Conference held at Cumberland Lodge, Windsor Great Park, July 1999 (Aslib, London, 2000).
8.
[8] J. Mullins , Statistical measures of usage of Web-based resources , Serials Librarian36(1/2) (1999) 207-210 .
9.
[9] D. Nicholas , An assessment of the online searching behaviour of practitioner end-users , Journal of Documentation52(3) (1996) 227-251 .
10.
[10] D. Nicholaset al. , Developing and testing methods to determine the use of web-sites: case study - newspaper , Aslib Proceedings51(5) (1999) 144-154 .
11.
[11] D. Nicholas and P. Huntington , Who uses web newspapers, how much and for what? In: NetMedia99: The Internet Conference for Journalists. Conference held at City University, July 1999 (Department of Journalism, City University, London, 1999).
12.
[12] D. Rahmel , Internet systems: log analysis , DBMS (Internet Systems Supplement)10(8) (1997) S12 .
13.
[13] N. Randall , Web Site Analysis Tools: The Results are In (1998). Available at: http://zdnet.com/pcmags/features/webanalysis2/index.html (visited 2 August 1999).
14.
[14] J. Rubin and R. Reimundez , Log-Analysis Tools - Site Server is on the Right Track (1998). Available at: http://www.nwc.com/917/917r1.html (visited 20 November 1999).
15.
[15] J. Sterne , Every click they take , Inc Magazine3 (1999) 117 . Available at: http://www.inc.com/incmagazine/archives/17991171.html (visited 17 October 1999).
16.
[16] M. Stone , Server Logs Help Shape Web Strategies (1999). Available at: http://www.net2b.com/cgi-bin/print_articles.pl/article/940 (visited 17 October 1999).
17.
[17] R. Stout , Web Site Stats: Tracking Hits, and Analysing Traffic ( McGraw-Hill , 1997).
18.
[18] S. Tweddle , P. Avis , D. Davis , N. James and H. Daniels , A method for investigating the usage of a cancer website , New Technology in the Human Services11(1) (1998) 12-16 .
19.
[19] D. Williams , Data Plays a Key Role in Net Success (1999). Available at: www.adage.com/interactive/articles/19990802/article3.html (visited 17 October 1999).
20.
[20] M. Zawitz , Web Statistics - Measuring User Activity (1998). Available at: http://www.ojp.usdoj.gov/bjs/pub/ascii/wsmua.txt (visited 9 October 1999).
21.
[21] Z. Zhang , Evaluating electronic journals and monitoring their usage by means of WWW server log file analysis , Vine111 (1999) 37-42 .
22.
[22] A search of Library and Information Science Abstracts on Dialog (conducted on 18 April 2000) uncovered about half a dozen references, most of which were published in the USA.
23.
[23] At the time of our survey, users of the websites had to register their personal details before they could use them.
24.
[24] A study of health logs, entitled ‘The Web, the kiosk, digital TV and the changing and evolving face of consumer health information provision: a national impact study 2000-2001’, funded by the Department of Health.
