regression, the estimated impact of enrollment delay on hourly wages drops to 1.7
percent and is no longer statistically significant. Thus, it appears that the wage impact of
enrollment delay arises because delayers attain less education than non-delayers. We
investigate the schooling impact of delay in more detail below.
Table 3 presents estimates of the DCR model described by equations (1) and (2)
above in the case where the coefficients of all predictor variables are restricted to be time-
constant and the unobserved heterogeneity distribution is assumed to have two mass
points. As can be seen from the estimates in Table 3, the waiting duration until college
enrollment has a significantly negative effect on the graduation hazard rate and a
significantly positive effect on the dropout hazard rate. The longer an individual delays
college enrollment, the more likely he/she is to drop out and the less likely he/she is to
graduate with a four-year degree.
To assess the magnitude of this effect, simulations were performed in which a high
school graduate’s delay until college enrollment was alternatively set to 0 (enrollment
immediately after high school) and 1 (enrollment after delaying one year ).11 The college
dropout and completion hazards from these simulations are presented in Figures 2 and 3,
respectively, when conditioning on entry into a four-year institution and Figures 4 and 5,
11 In these simulations we calculate the probability of gradation for each individual and then average
probabilities across individuals.
17