Early Prediction of Student Self-Regulation Strategies by Combining Multiple Models

Self-regulated learning behaviors such as goal setting and monitoring have been found to be crucial to students ’ success in computer-based learning environments. Consequently, understanding students ’ self-regulated learning behavior has been the subject of increasing interest. Unfortunately, monitoring these behaviors in real-time has proven challenging. This paper explores a variety of data mining approaches to predicting student self-regulation capabilities. Students are classified into SRL-use categories based on evidence of goal-setting and monitoring activities. Prior work on early prediction of these categories pointed to logistic regression and decision tree models as effective techniques. This paper builds on these findings by exploring techniques by which these models can be combined to improve classification accuracy and early prediction capabilities. By improving classification accuracy, this work can be leveraged in the design of computer-based learning environments to provide adaptive scaffolding of self-regulation behaviors