Choice of Variables and Gender Differentiated Prediction Within Selected Academic Programs

The existence of differential prediction of 2‐year grade point average is reported for gender groups within programs of study at a large state university. Johnson‐Neyman analyses indicate regions on the predictor score scales where differential prediction has practical impact within each program. A significant portion of students is generally affected. It is shown that careful selection of a subset of variables reduces differential prediction while maintaining predictive potency.