Using Data Mining Techniques to Guide Academic Programs Design and Assessment

Abstract

This paper presents a novel application of data mining techniques to guide academic programs design and assessment. More specifically, it propose using association rule mining techniques to discover a set of rules that govern the relationship between two core components of an academic program, program educational objectives (PEOs) and students outcomes(SOs). As a case study, this paper demonstrates how association rule mining techniques are applied to mine mapping rules between the PEOs and a predefined set of SOs adopted by the American Board for Engineering and Technology-Engineering Accreditation Commission (ABET-EAC) for engineering programs. To this end, a set of 152 ABET-EAC accredited engineering programs’ self-study reports have been collected and the mapping data between their PEOs and ABET-EAC SOs have been extracted. This dataset has been pre-processed and transformed into a representation suitable for applying association rule mining techniques. This involves identifying a set of PEOs labels, annotating data instances with PEOs labels, and projecting each multi-label data instance into a set of single-label instances. Apriopi algorithm is then applied to discover the rules that govern the mapping between PEOs and ABET-EAC SOs. The discovered rules are of particular importance for guiding the design and assessment of engineering academic programs. Besides that, the discovered rules unveil of a number of interesting correlations between PEOs and ABET-EAC SOs that need further investigation by pedagogists.