Prediction of Graduate Performance in Economics

Professor Hansen's paper describes an effort to predict the success of graduate students in economics at The University of Wisconsin. Undergraduate grade point average, GRE scores, and the quality of the undergraduate schools were used as independent variables in a linear multiple regression model. Graduate grade point average was the dependent variable. The value of the independent variables as predictors is analyzed both jointly and in isolation. Among the important conclusions reached by Hansen is that heavy reliance upon a single predictor such as the undergraduate grade point average is unwise, that the GRE quantitative score takes on greater significance as the graduate student advances into the second year of his program, and that school quality must be taken into account.