Discussion

If we focus on the analysis of all ten questions, the selection list navigation bars did not seem to provide all of the benefits that were predicted. Subjects actually performed the fastest with the control navigation bar that only provided a link back to the main page. The selection list navigation bar did require fewer page loads of the subjects than the standard, but did not perform any better than the control. If we shift our attention to the analysis of only the first five questions, the selection list navigation bars seemed to be even worse, requiring significantly more time for subjects to complete the search tasks. These results do not support our hypotheses.

Support for the benefits of the selection list navigation bars can be found in the analysis of the last five questions of each treatment. The selection list navigation bars required less time than the standard and fewer page loads than either the standard or control navigation bars. The results from the last five questions of each treatment do support our hypotheses.

The differences between the results from the first five questions and the results from the last five questions indicate that subjects were able to learn how to use the selection list navigation bars quickly. This is comforting news for anyone interested in developing or implementing selection list navigation bars. Even though the selection list navigation bars received the lowest subjective scores for providing comfort, subjects felt that they would like this style of navigation bar much more with more practice. Although these two measures are probably not independent, the results indicate that users are willing to take the time and effort to learn how to use the selection list navigation bars.