FORECASTING CREDIT HOURS

This paper employs Tobit to estimate retentionprobabilities and credit hours at two universities. Theinnovation is that this technique examines credit-hourchoice with the decision to depart the university treated as the choice of zero credit hours.Tobit is appropriate for this problem because itrecognizes the lower bound of zero on credit hours andincorporates this bound into the parameter estimates and forecasts. Models are estimated for credithours in a single year and cumulative hours over asix-year horizon.