Predicting national direct student loan defaults: The role of personality data

The purpose of this study was to determine if National Direct Student Loan defaulters could be differentiated from payers by using information available at the time the loan is made. Data from 686 repayers and 192 nonpayers were analyzed. It was found that a combination of five student characteristics was all that was needed to accurately predict which students would not repay their loans. The combination of factors which proved to have predictive value in order of importance were: grade point average at the time of the loan, ACT composite score, OPI complexity scale, the OPI anxiety level scale, and the size of the loan. The most striking outcome was that personality measures have strong predictive value in identifying future defaulters. Further validation of these findings is underway.