John R. Hills

On the Use of Student Ratings of Faculty in Determination of Pay, Promotion, and Tenure.

Student ratings of faculty have traditionally been obtained in a manner designed to be useful to the individual faculty member or to other students. It is now sometimes proposed that the resulting data be used to determine faculty pay, promotion, and tenure. Recent articles and the reviews of past literature on ratings are analyzed to determine whether student ratings are usually associated with teaching effectiveness, whether they are sometimes biased by irrelevant factors, whether faculty can effectively revise their behaviors to improve their ratings, and whether improved ratings result in improved teaching effectiveness.The results tend to indicate that student ratings of faculty as they are currently collected cannot be trusted for considerations of pay, promotion, and tenure.

Univariate selection: The effects of size of correlation, degree of skew, and degree of restriction

Pearsons formula for univariate selection was derived with the assumption of normality of variates before and after selection. This study examined the influence of skew upon estimates from Pearsons formula under certain conditions. It was found that even with essentially symmetric distributions, a large proportion of the data is necessary to obtain reasonably precise estimates of low correlations. With increasing skew, estimates become increasingly erroneous, the direction of the error depending upon which tail of the distribution is the basis of the estimates. Difficulties in applying correction for univariate selection in several studies of the predictability of college-grades for Negroes from scores on standard aptitude tests are discussed.

Olkin's New Formula for Significance of r13 vs. r23 Compared with Hotelling's Method1:

Two alternate formulas for determining the significance of the difference between a certain two correlation coefficients have been reported in the literature.2 These coefficients, r13 and r23, denote the correlation of two predictor variables (1 and 2) with a predictand (3) within a single sample drawn from a population. From these statistics the researcher can tell whether the correlation of 1 with 3 is significantly different from the correlation of 2 with 3. The older formula for testing the significance of this difference was derived by Hotelling (1940) and is commonly cited in statistics textbooks (e.g. Ferguson, 1966; McNemar, 1969; Tate, 1955; and Walker and Lev, 1953). This statistic, which has Students t distribution with N 3 degrees of freedom (where N is the sample size), is

Easier test improves prediction of black students' college grades.

Recently some predominantly Negro colleges have begun to require that all of their applicants for admission submit scores on a nationally administered test such as the College Entrance Examination Board Scholastic Aptitude Test (SAT) or the battery of the American College Testing Program. The frequency distributions of test scores in these institutions are often markedly skewed positively and have smaller standard deviations than are typically found in white colleges. For example, on the SAT, while the mean Verbal score for all public-highschool seniors would be in the neighborhood of 390,1 the mean SAT-V score for entrants to a representative predominantly Negro college in a Southern state in the fall of 1966 was 277. (A chance score would be in the neighborhood of 220230). The standard deviation for all public-high-school sen.iors would probably be much greater than 100, but for entrants to this college the standard deviation of SAT-V was 50.