Mathematical models of judgment policy are compared in evaluating 625 Cadet Performance Reports during a single rating cycle. Linear and nonlinear regressions are used to describe the cues most important in determining overall ratings. In addition, performance rating policy is compared with stated policy for each of forty cadet raters. In general, the findings support current literature in judgment policy-capturing. Cadet raters are internally consistent in applying their individual policies to cadet ratings. The relative importance of the ten cue variables varies significantly between raters while the aggregate data allow overall ratings to be described with only ten of the cues. There is no difference between the linear and nonlinear models in the amount of variance explained. The latter model does provide considerable insight into the composition of the overall ratings in terms of cue variable interactions. Finally, cadet raters' stated policies differ greatly from their actual policies. Recommendations are proposed to provide a more consistent interpretation and application of the Cadet Performance Report rating system. A logical extension of this study includes the implementation of a weighting scheme and a follow-on study to further validate the present findings. (NTIS)