Conducting remote, Internet-based experiments on Web design

This paper presents an Internet-based study of the effect of intra-article navigation affordances on Web browsing behavior. Navigation was assessed by randomly assigning participants to one of four experimental conditions: the existing intra-article navigation bar, one of two redesigned intra-article navigation bars, or an intra-article navigation bat-combined with a global navigation bar. Data about participants' browsing behavior were logged by using our experimental design toolkit to instrument the Web site's pages with PHP scripts (J. Barrick et al., 2004). Demographics, perceptions, and comprehension were measured through a post-browsing survey. The study is ongoing and results are presented at the conference. We also report on the further development of our toolkit that has made conducting Internet-based research possible.