Temporal Learning Analytics for Adaptive Assessment

Abstract

Accurate and early predictions of students’ performance could significantly affect interventions during teaching and assessment, which gradually could lead to improved learning outcomes. In our research work, we seek to identify and formalize temporal parameters as predictors of performance (“temporal learning analytics”-TLA) and examine students' temporal behaviour (i.e. in terms of time-spent) during testing. The goal is to specify a functional set of parameters that will be embedded in an adaptive assessment system in order to contribute to personalization of feedback services. We adopted the Partial Least-Squares (PLS) analysis method for formulating the causal dependencies between latent variables and the relations to their indicators. In this paper we present the motivation and rationale of our work, along with the followed methodology, initial results and contributions so far, and our plans on future work.