Analyzing multiple-choice questions by model analysis and item response curves

• Main Authors: Wattanakasiwich P., Ananta S.
• 格式: Conference or Workshop Item
• 語言: English
• 出版: 2014
• 在線閱讀: http://www.scopus.com/inward/record.url?eid=2-s2.0-78649889727&partnerID=40&md5=fefff2ba834562e94b55853928d5a6b1; http://cmuir.cmu.ac.th/handle/6653943832/6134

In physics education research, the main goal is to improve physics teaching so that most students understand physics conceptually and be able to apply concepts in solving problems. Therefore many multiple-choice instruments were developed to probe students' conceptual understanding in various topics. Two techniques including model analysis and item response curves were used to analyze students' responses from Force and Motion Conceptual Evaluation (FMCE). For this study FMCE data from more than 1000 students at Chiang Mai University were collected over the past three years. With model analysis, we can obtain students' alternative knowledge and the probabilities for students to use such knowledge in a range of equivalent contexts. The model analysis consists of two algorithms - concentration factor and model estimation. This paper only presents results from using the model estimation algorithm to obtain a model plot. The plot helps to identify a class model state whether it is in the misconception region or not. Item response curve (IRC) derived from item response theory is a plot between percentages of students selecting a particular choice versus their total score. Pros and cons of both techniques are compared and discussed. © 2010 American Institute of Physics.