Influences of Content Location, Navigation, and Internet Self-Efficacy on Task Performance and Website Evaluation

Previous research has identified navigational elements and site information architecture as important factors in how effectively users can complete Web tasks, such as locating specific content within a Web site, yet little research has focused specifically on how site depth and breadth impact user task performance (e.g. Krug, 2000; Nielsen, 2000; Palmer, 2002; Yu & Han, 2001). Past research has also examined the impact of self-efficacy on Web user behavior, suggesting low efficacy users tend to accept rather than question computer systems, have poorer searching strategies, and tend to use the Internet less than high efficacy users (e.g. Liaw, 2002; Tsai & Tsai, 2003). The current experiment focused on the effects of Web site navigational elements, the location (depth) of content, and Internet self efficacy on users' ability to complete a task by reaching target content within a Website, correctly answer questions based on this content, and Web siteusability ratings. We manipulated site design and the depth of target content in the site. Each of fifty-two participants viewed one of three Web site designs: (1) plain context menu, (2) emphasized context menu, (3) no context menu. Participants were asked to locate specific content on the site and answer 10 questions based on this content. The answers to the questions were located on pages wide (two or fewer links away) or deep (three to five links away) in the Website. In a post-test, we measured participants' subjective ratings about the site and information about their past experience using the Internet, focusing on Internet self-efficacy and years of Internet experience. It was hypothesized participants would have more incomplete tasks and more question errors when searching for information located on pages deep in a site and when using the site without a context menu. Additionally, it was believed low Internet efficacy participants would complete fewer tasks and make more question errors, but would provide higher Web site ratings than the high efficacy group. Two ANOVAs indicated the participants had significantly more incomplete tasks and more question errors when searching for information located deep in the site. Site design had no impact on participants' ability to successfully reach the target page or question accuracy; however, participants with six or moreyears of Internet experience tended to rate the design with the plaincontext menu as more usable than the other two designs. High Internet efficacy participants gave significantly higher usability ratings, regardless of design. Internet efficacy was significantly correlated to years of Internet experience and to participants' ease in locating the question information on the site. Internet efficacy had no impact on participants' ability to successfully reach the target page or question accuracy.