Learning Analytics in Higher Education: Making the Connections Among Data, Learning, and Outcomes

• In Event: Learning Analytics in Higher Education: Exploring Big Data on Campus

Abstract

University programs can be considered as a chain of courses. Each course experience offers the student key concepts tied strongly to the subject area and other facets revolving around individual growth and maturity. Identifying the concepts covered in a university course is crucial for prospective college students and parents seeking information on individual program and course offerings at a university. Accrediting bodies like Accreditation Board for Engineering and Technology (ABET) require a program to cover a set of pre-specified concept areas within the classes. Additionally, a holistic approach to program evaluation involves the analysis of the prerequisite structure within a department, specifically the conceptual overlap between courses in a prerequisite chain. The data describing university courses is publicly available on university websites in the form of course catalogs and syllabi. However, understanding the text-based course descriptions requires manual inspection and domain expertise.

We developed a tool that automatically infers the sets of concepts within a course extracted from text obtained from course catalog descriptions. We present this application framework, which is publicly available at http://edmine.cs.gmu.edu. The web application provides tools to view the inferred concepts for a university’s courses, compare departments by their topic composition, and visually analyze conceptual overlap in departmental prerequisite structures.

Author

• Huzefa Rangwala, George Mason University