A Joint Modeling Framework Using Responses and Response Times to Track Skill Acquisition: Model Estimation and Application

• Activity Number: 394 - Recent Advances in Cognitive Diagnosis Modeling
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Mental Health Statistics Section
• Author(s): Shiyu Wang* and Susu Zhang and Jeff Douglas and Steven Culpepper
• Companies: University of Georgia and University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign
• Keywords: response time; latent class model; longitudinal

Abstract

Response times, which is the amount of time the test taker spends considering and answering each item, has been extensively studied and used in a testing environment as the useful source of information to reflect individual response behavior and item characteristics. In this study, we consider using response time in a learning environment to model students' learning progress. This type of information, in addition to their responses to the test questions, can be a valuable source of information to measure students learning trajectory. We propose a joint modeling framework that use responses and response times together to characterize individualized students growth pattern. The challenges and issues on model estimations and comparisons are discussed as well. The proposed models are evaluated through a computer-based learning system that is designed to improve students' spatial skills.