Web Usage Patterns of Chronic Fatigue Syndrome Continuing Medical Education Internet Courses

This study identified and analyzed web usage patterns of two chronic fatigue syndrome (CFS) continuing medical education (CME) internet courses in order to better understand the needs of the target audience, improve the process for which persons navigate the courses, and predict dropout patterns.

We collected 2008 web usage data, indentified 10,136 navigation paths from more than 69,000 page views, and constructed a Markov chain model and a fallout diagram.

Our Markov chain model examined the immediate movements from one page to another. It showed the users’ general usage patterns and the probabilities of a user moving from one point to another in each of the courses. It also indicated dropout probabilities at any point in the courses. The fallout diagram examined the eventual reachability and dropout from one page to another.  These results indicate potential steps in the learning process where the visitors were most likely to leave either course.

We categorized visits into three levels of interest groups, Pass-by, Interested, and Serious, according to their viewed web content. They accounted for 75%, 11%, 14% of total visits, respectively. We also separated all paths based on the landing pages, such as the home page, syllabus pages, and introduction pages. Paths with different landing pages reflected different usage patterns. Visits beginning at the introduction or syllabus pages were more likely to be Interested or Serious visits than those initiated from the home page.

One potential design issue identified in this study was corrected in February 2009. The evaluation result shows a positive effect of the modification on keeping users in the course.

In summary, web usage patterns on the two CFS CME Internet courses suggested that most users took the courses in the order recommended by the web site and revealed several potential improvements points in the courses.