| The purpose of this study was to organize menu items based on a user-process model and implement a new version of current software. Dervin & Nilan's time-line method is used to develop a user-process model to compare the usability of two different interfaces for Augmented Seriation software, a menu-based system for data mining developed by Gluck, McRae, & Yu.; The research questions for this study focus on which type of menu interface for Augmented Seriation, categorized or user-based, is more usable in terms of task completion time, accuracy (success rate of completing task), and user satisfaction (perceived satisfaction, ease of use, and usefulness).; The two major research methods were used to address the research questions. First, the researcher developed a user-process model drawn from actual users' understanding of their goals/strategy to solve their information needs by using Dervin's sense-making interviews.; Secondly, usability testing was employed to compare two interfaces. Task completion time, accuracy, and user satisfaction were measured to test three hypotheses. The subjects' general computer experience was measured through technology & computer experience questionnaire to observe any main effects on subject performance. User preferences and comments were gathered through post-test questionnaire. The data collected from the usability testing was analyzed by using the General Linear Model Repeated Measure to observe between subject factors and within subject factors of the performance.; The results from the usability testing did not confirm the hypothesis that either the proposed menu or the traditional menu is more effective in terms of task completion time, accuracy, and user satisfaction. However, the types of task yielded some significant results in task completion time and in accuracy. This study didn't reveal statistically significant result for supporting the proposed interface. However, in spite of the advantage of having familiarity in using the traditional menu, the subjects did not have superior performance with the traditional menu. Consequently, it is reasonable to assume that the proposed menu interface in this study is more usable and effective considering a large effect size. The apparent tie in performance should indicate to designers the long-term improvements possible with a user-based timeline menuing system. |
