Measurement Techniques, Grades and Ratings of Instructors

There have been many studies of the effect of grades on evaluations which instructors receive from students. (For references, see Costin, Greenough, and Menges [1] or McKenzie [4].) Some studies have found a positive relation between student grades and teacher ratings (e.g., Nicols and Soper [2]); some have found essentially no sucli relation (e.g., Kelley [3]). As Nicols and Soper have argued, we would generally expect a positive relationship; how can these ambiguous results be explained? The answer is that there are at least two ways in which the desired effects can be measured, and different results can be expected from these alternative methods of measurement. The first part of this paper will explain the measures and the theory; the second will test the theory.