Comparing the reliability of performance task scores obtained from rating scale and analytic rubric using the generalizability theory

Abstract

In the performance assessment, unbiased and accurate scorings depend not only on raters but also on accuracy of scoring keys. It could be confusing to choose the type of scoring key for educators in most situations. The study aims to find whether there is a difference between ratings with rating scale and analytic rubric and to compare reliability of scores given by the same raters with both scoring keys to the same performance tasks using generalizability theory. The results of this study would be a guide for implementers to determine the type of scoring keys for scoring performances. By the analyses, results reveal that scores obtained with the rating scale have higher reliability compared to scores obtained from the analytic rubric. Interviews related to both scoring keys reveal that for the scoring of performance tasks of teacher candidates, the rating scale is economic in terms of time.