Sequential Mastery Detection of Multiple Attributes With Response Accuracy and Response Time Data

• In Event: Assessment in the Advent of Process Data and AI

Abstract

Cognitive diagnostic models provide a structured approach for harnessing a range of fine-grained skills to address items. Once these skills can be identified, the primary goal of an adaptive tutoring system is to facilitate learners to acquire all target skills. This research focuses on detecting acquired mastery for a particular set of skills when learning progress is reflected in response accuracy and response times. This can be viewed as a problem of detecting sequential change points, with the aim of minimizing the overall detection delay across all skills while maintaining a fixed Type I error. The simulation study showed that incorporating the response time data to the existing inference solely on response accuracy can hasten the detection process using a modified version of G-DINA for response time.

Author

• Sangbeak Ye, Florida Atlantic University