Students' scores and subsequent grades for clinical courses should reflect their performance. The scores are dependent in part on the reliability of faculty evaluations. Large variance in instructor scoring could unduly affect students' final grades. This study investigated the effect on students' scores of instructor variance in clinical evaluation. The results indicated that instructor variance was responsible for 11 percent of the variance in students' final scores. The remainder of the differences resulted from differences in student performance. More than one half of the students' grades would have been either higher or lower if the scores assigned by one or more individual instructors had been excluded. For all but one student the net change in total scores was less than one standard deviation. The authors conclude that the students' grades were reliable and that individual differences among instructors did not unduly affect the students' final scores.