The Timing of Student Graduation: a Joint Model Approach

The joint models for longitudinal and time-to-event data are a recentfamily of models that analyse jointly longitudinal and survival data. These modelsare composed of two sub-models, a hazard function related to the true and unob-served value of the longitudinal outcome, and a longitudinal mixed model where theoutcome is equal to the true level plus a random error term. Several authors appliedthis model to a biostatistic area, but this paper aims to analyse the timing of studentgraduation in an Italian university. The model will analyse jointly the time to grad-uation and the student’s path, focusing on the average grade and on the number ofexams that the student has already passed before a fixed time.