Perceived Grading and Student Evaluation of Instruction

We analyzed student evaluations for 3,585 classes collected over 20 years to determine stability and evaluate the relationship of perceived grading to global evaluations, perceived fairness, and appropriateness of assignments. Using class as the unit of analysis, we found small evaluation reliability when professors taught the same course in the same semester, with much weaker correlations for differing courses. Expected grade and grading related questions correlated with overall evaluations of courses. Differences in course evaluations on expected grades, grading questions, and overall grades were found between full-time faculty and other types of instructors. These findings are expanded to a model of grading type questions mediating the relationship between expected grade and overall course evaluations with a moderating effect of type of instructor.