UI Design Newsletter – August, 1999

Dimensions affecting Web site identity, Omanson, R. C., Cline, J. A., Kilpatrick, C. E., and Dunkerton, M. C., Proceedings of the Human Factors and Ergonomics Society 42nd Annual Meeting, 429-433 (1998).

What should Web site designers do to help users know when they are still in the same Web site or have moved to a different site after clicking on a link?

This question was studied by Omanson, et.al., 1998. Their research attempted to determine which dimensions were most important when establishing a Web site identity. They had observed that many Web users are often unaware when they have moved to a different site.

In their first experiment, they had 19 adult Web users interact with 42 Web pages from various sites. The participants were shown the Web pages one at a time for about 20 seconds. They were asked to imagine that these were pages visited while browsing the Web, and to indicate whether they were at the same site or at a different site. The subjects also provided information about which aspects of the site most influenced their judgment.

The participants identified eight within-page dimensions of web pages that they believed were being used to make decisions:

1. Background color and/or background pattern of a page
2. Company or organization logo or symbol
3. Font style used and its characteristics (size, color, bold, italics, etc.)
4. Location of the table of contents on each page (left, right, top or bottom)
5. Organization of text and graphics on a page
6. Style of graphics used in pictures, banners and illustrations
7. Subject matter of the Web page
8. Title or heading at the top of the page

Had they stopped with this first experiment, the resulting behavior of Web site designers would have been wrong for the next few years.

In their second experiment, they had 168 experienced and inexperienced Web users use 23 different Web pages. They evaluated 11 transitions that involved staying on the same site, and 12 transitions that involved going to a different site. Again, each Web page was shown to participants for 20 seconds, and the subjects then indicated whether they were on the same site, or a different site.

After making a judgement, the participants were asked to tell which of the eight dimensions most affected their judgments. For both experienced and inexperienced users, "logo" was the primary dimension used. Even though the subjects in the first experiment tried to identify why they had made their judgements, seven of their eight within-page dimensions seemed to have little, or no, actual impact on the decisions being made. (Test participants frequently try very hard to satisfy testers with logical, but meaningless, explanations for their performance.)

There are two major findings in this study.

• First, users tend to use the Web site logo to help orient themselves on the Web.
• Second, other visual attributes may not be as important as once thought. Many of the website properties that were once believed to have strong "orienting" impact on users, such as background colors, organization of the text and graphics, consistent font style, etc., may be less important.

It is possible that the inability to effectively compare the pages was due primarily to working memory limitations (i.e. many people find it difficult to remember details of the page they were looking at just a few seconds ago).