A Study of Salary Equity at a Teaching-Oriented Undergraduate Institution Using Multiple Models.

A recent study of salary equity at an undergraduate institution using traditional regression-based methodology is presented. In a traditional salary equity study a linear model is empirically fitted for predicting salaries (the dependent variable) from a set of independent (predictor) variables using the most up-to-date data possible for some base group, usually Caucasian males. The base group model is then applied to all eligible faculty members to predict salaries that they would have earned under the model. The differences between actual and predicted salaries are measures of pay inequity. Data for this study came from the 1989-90 fiscal year for Winona (Minnesota) State University. When the regression equation based on male faculty data (n=176) without rank as a predictor, was applied to the female faculty data, it was found that, on the average, female faculty members (n=78) were underpaid by $3,113 when compared to male faculty. Predictor variables and departmental differences are discussed in detail. Five tables present study data. (SLD)