Brown Bag Seminar: Chia-Wen Chen

Speaker: Chia-Wen Chen

Title: Structural Equation Mixture Model for identification of Misconception with Certainty of Response Indices

Abstract: A student having a misconception is identified when she/he solves scientific problems by using an intuitive understanding that does not agree with the scientific concept. Misconceptions often interfere with learning new knowledge and should be unlearned before the correct conception can be taught. A Certainty Response Index (CRI), an additional question complementing an item on a cognitive test that requests respondents to provide their confidence levels in answering that item, is a tool for detecting misconceptions and has been applied in physics and mathematics education for years. There is a gap in the literature regarding having a parametric measurement model for cognitive items and CRIs to identify misconceptions. To address this issue, this study developed a structural equation mixture model incorporating cognitive items and CRIs by which practitioners can obtain the classification of students’ misconceptions and their latent ability concurrently. Simulation studies showed that the parameters of the proposed model can be recovered well and that the proportion of classifying students’ misconception correctly was high. Data collection of item responses together with CRIs is planned.

Published Sep. 3, 2019 6:43 PM - Last modified Sep. 12, 2019 5:32 AM