Abstract
On merit rating, it is ideal that all raters give the same scores to a given ratee. However, in fact, raters usually give him various scores, which may be caused by the rating tendencies. So we proposed a rating model for the quantitative analysis of the raters' tendencies in the last paper. In this paper, we make a model relating to the effect of the raters' tendencies, which are obtained in the last paper, and the categories of the rating tendencies (lenient tendency, central tendency and severe tendency). Then we analyze them quantitatively.
