This research attempts to classify time as a key variable that can be used to understand and analyze the many aspects of the system of human interaction, specifically human interaction with a Web-based environment. This is achieved by focusing on relevant factors in the data, rather than on any specific set of variables, and modeling these factors over time (temporal modeling).
This hypothesis is challenged though a series of user testing in which the subjects navigate a set of Web sites. Their progress is monitored with respect to the kinds of interactions they have with the sites, and is modeled to show how time can be used to illustrate successes and failures that the subjects have. Both first-time use and concurrent use of the Web sites is considered.
The findings are correlated with each other to show patterns in the interactions, and with the components of the Web sites themselves to show how elements in the design could help or hinder user interaction. Finally, the techniques illustrated in this research are compared to other forms of user analysis, and suggestions for their use in existing design methodologies are given.
In the end, the intent is to show that by observing and analyzing the temporal states of a user within a given Web-based environment, one can accurately design for desired and successful user behavior, and better predict future behavior.