Development and application of a rubric for evaluating students' performance on newton's laws of motion

Abstract

This study aims to develop and apply a rubric to evaluate the solutions of pre-service primary science teachers to questions about Newton's Laws of Motion. Two groups were taught the topic using the same teaching methods and administered four questions before and after teaching. Furthermore, 76 students in the experiment group were instructed about the features and use of the rubric and asked to construct a rubric, while 77 students in the control group were not. Students' solutions were evaluated with the agreed rubric by the instructor, an independent coder and the peers in the experiment class. The effectiveness of the rubric on students' achievement was examined by applying descriptive statistics and linear regression to scores obtained from both tests. T-test statistics and analysis of variance procedures were also used to analyze the reliability and validity of the assessments made. The results revealed that the developed rubric was used consistently by the instructor and peers and significant correlations (p < 0.001) were found among the scores. The inter-coder reliabilities were 0.98 and 0.93 in the pre- and post-tests with 76 peer coders. A generalizability study showed that the estimates of 16 peer coders on average matched the reliability of single-instructor assessments. It was concluded that the developed rubric was able to highlight the aspects of the problem solutions and helped increase students' achievement.