Application of partial model in practical performance assesment of physics learning

Abstract

The purpose of this study focus on the application of Partial Credit Model (PCM) in the assessment of students' practical performance in physics learning. The study was given to 220 students who have same cognitive abilities. there are two sets of test questions as many as 10 questions each with 20% in the form of anchor item questions. Based on the data obtained, the analysis of the estimated ability of test takers with 1 PL with PCM model. Analysis of estimated capabilities with PCM is only used on 1PL based on the difficulty level of the question item. Analysis with PCM is done with the help of the Pascale application. PCM evaluates a test based on the results of the estimated item parameters and the estimated ability of the test taker. Estimation of the ability and estimation of item parameters can be done simultaneously, the first step is to estimate the parameters of the item and the results are used for estimation of capability parameters. PCM predicts participants' ability through their ability to answer test items correctly where the higher the participant's ability, the higher the chance of answering correctly. Based on the analysis and discussion that has been done can be concluded that the elimination of physics learning using PCM provides information about the difficulty level of test items, the relationship between the chances of test takers answering correctly with the ability of test takers, and the estimated ability of test takers. The difficulty level of the item is in the difficult category, the correlation between the ability of the test taker and the chances of the test taker answering correctly by 96.8% which indicates that the higher the ability of the test taker, the greater the chances of the participant answering the correct test question. The average ability of test takers was -0.227 which showed that the ability of test takers in the category was low.