Inferring learners' affinities from course interaction data

TL;DR

This paper introduces a data-driven model that infers learners' affinities for different learning patterns from course interaction data, using non-negative matrix factorization and bootstrap inference to analyze and compare student groups.

Contribution

It proposes a novel approach combining stylized learning patterns and affinities with NMF and bootstrap methods to analyze course interaction data.

Findings

Identified meaningful learning patterns from student interaction data.

Connected learning patterns to a learning style system.

Demonstrated differences in affinities between passed and failed students.

Abstract

A data-driven model where individual learning behavior is a linear combination of certain stylized learning patterns scaled by learners' affinities is proposed. The absorption of stylized behavior through the affinities constitutes "building blocks" in the model. Non-negative matrix factorization is employed to extract common learning patterns and their affinities from data ensuring meaningful non-negativity of the result. The empirical learning patterns resulting from the actual course interaction data of 111 students are connected to a learning style system. Bootstrap-based inference allows to check the significance of the pattern coefficients. Dividing the learners in two groups "failed" and "passed" and considering their mean affinities leads to a bootstrap-based test on whether the course structure is well-balanced regarding the learning preferences.