Group Differentiated Prediction

Publisher Website
Google Scholar
Abstract
Studies of predictive bias have frequently shown that a prediction equation based on majority group members tends to overpredict the criterion perfor mance of minority group members. Two statistical ar tifacts that may cause the overprediction finding are reviewed and evaluated using data for black and white students at 30 law schools. It is shown that (1) the de gree of overprediction decreases as the predictive ac curacy for white students increases, and (2) that ov erprediction can be caused by the effects of selection on variables not included in the regression model. Use of Heckman's (1979) procedure to adjust the estimates of the regression parameters was found to essentially eliminate overprediction.

This publication has 4 references indexed in Scilit: