The relative merits of different types of overall evaluations of teaching effectiveness

An unresolved issue in student evaluations of teaching effectiveness (SETE) literature is what type of overall evaluation of teaching effectiveness should be used in personnel decisions. The objective of this study is to compare the merits of: (a) an overall evaluation made by students, (b) a weighted average overall evaluation with the weights determined by students, (c) a weighted average overall evaluation with the weights being determined by the individual instructors teaching their respective classes, (d) an unweighted average overall evaluation, and (e) a second-order factor that proxies for an overall evaluation. Our results indicate that: (a) all of these overall evaluations were very highly intercorrelated, (b) the unweighted and weighted average overall evaluations measured virtually the same thing, and (c) the second-order factor that served as an overall evaluation was most highly correlated with the other overall evaluations and had the advantage of being most understandable to the faculty.