WeBWorK log files as a rich source of data on student homework behaviours

ABSTRACT

The online homework system WeBWorK has been successfully used at several hundred colleges and universities. Despite its popularity, the WeBWorK system does not provide detailed metrics of student performance to instructors. In this article, we illustrate how an analysis of the log files of the WeBWorK system can provide information such as the amount of time students spend on WeBWorK assignments and how long they persist on problems. We estimate the time spent on an assignment by combining log file events into sessions of student activity. The validity of this method is confirmed by cross referencing with another time estimate obtained from a learning management system. As an application of these performance metrics, we contrast the behaviour of students with WeBWorK scores less than 50% with the remainder of the class in a first year Calculus course. This reveals that on average, the students who perform poorly on their homework start later, have shorter activity sessions, and are less persistent when solving problems. We conclude by discussing the implications of WeBWorK analytics for instructional practices and for the future of learning analytics in undergraduate mathematics education.

KEYWORDS:

• WeBWorK
• online homework
• learning analytics

Acknowledgements

We thank the Carl Wieman Science Education Initiative at the University of British Columbia, supported in part by a generous donation from Professor David Cheriton (Stanford University), that funded this work. We would also like to thank Costanza Piccolo for feedback on the manuscript, and the reviewers for their additional helpful suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.