Social Psychology Network

Feedback is just as essential in web development as it is in teaching and research, yet there is currently no formal system of peer review for web sites. As a result, it falls to individual web developers to gather feedback on the quality and usefulness of their work. There are at least three ways to obtain this information, none of which are mutually exclusive with the others: (a) ask people for feedback, (b) install digital counters to record the traffic on specific pages, and (c) analyze the log files of your web server.

Asking other people. The easiest way to gather feedback is to ask colleagues, friends, and family members to bluntly appraise the strengths and weaknesses of your site. One limitation of this approach, though, is that these individuals may differ from the people who actually use your site. To solicit feedback directly from visitors to your site, you might consider adding an interactive survey form to your web site, or even simpler, you might add a "mailto" link on your home page with an HTML statement such as:

Please send additions, corrections, and any other feedback to
<A HREF="mailto:me@school.edu">me@school.edu</A>

In the previous statement, me@school.edu would be replaced in both locations with your e-mail address. The advantage of soliciting information in this way is that virtually all feedback will be useful. In my experience, however, fewer than one visitor in a thousand sends feedback, so this method will not generate a large or representative set of comments. If you are using your web site in connection with an academic course, a better technique for gathering representative feedback is to anonymously survey students in the class. For example, you might ask students to rate how much they enjoy having the site, how often they use it, how easy or difficult the site is to navigate, how much they like the layout and appearance, and what they would suggest to improve the quality of the site.

Installing digital counters. Perhaps the single most important piece of feedback to a webmaster is the number of visits each page gets. After all, there is no point in developing a web page that does not get visited. By the same token, if certain pages receive an unusually large number of visits, this information may suggest the value of developing similar pages in the future. To monitor the number of visits to a page, you can either ask a technical support professional at your institution to help install a digital counter, or you can install a counter yourself through a commercial counter service on the web (a local counter is usually preferable to a commercial service, in part because commercial counting services often slow down the page loading time).

Analyzing log files. Another way to monitor the traffic on your web site is to analyze the log files of your web server (servers typically keep a log of the web pages they serve, the time they served them, and the characteristics of the computers that made page requests). This method of gathering information will probably require the assistance of technical support professionals at your institution, but the necessary software can usually be installed within a day or two, and the results are much more informative than the information provided by digital counters. For example, the first log analysis program I used, AccessWatch (Maher, 1998), was installed in less than three hours and displayed not only the number of visits to each page, but also information on the sequence of pages that were visited, a general profile of who was accessing the pages, hourly information on the flow of traffic, and the mean number of pages viewed per visit. This information helped me tailor the site to the needs of visitors, schedule web maintenance at off-peak hours, and devote my time to the pages that receive the highest amount of traffic. It also eliminated the need to install separate counters on each page, and it avoided the slow-down in page loading caused by many digital counters.