In this paper, I study the timing of high school dropout decisions using data from High School and Beyond. I propose a Bayesian proportional hazard analysis framework that takes into account the specification of piecewise constant baseline hazard, the time-varying covariate of dropout...