Abstract
The purpose of this study is to empirically examine the relationship between rating accuracy and distributional rater errors through applying an analysis model of merit rating errors. The result of an experiment involving a total of 190 undergraduate students shows a high correlation between elevation, one of the rating accuracy components, and the estimated value of distributional rater errors when applying an analysis model of merit rating errors. This suggests that for distributional rater errors, using an analysis model of merit rating errors is an effective measure for rating accuracy in comparison with traditional rater error measures (e.g., average, kurtosis, skewness, standard deviation).
