The hazard of success: A longitudinal study of time-to-degree among doctoral students using discrete-time survival analysis

This thesis considers variation in time-to-degree among doctoral students at the Harvard Graduate School of Education using longitudinal analysis of twenty-years of student data (n = 1,640) to create an estimate of the conditional probability of graduation in each year.;A subject of study for over 40 years, the duration of doctoral study has been demonstrated to vary systematically with field of study and institution, and these factors may mask the effect of other factors (such as ethnicity, age or registration status).;Chapter 1 presents a paper describing the pattern of probability of graduation and the competing risk of non-completion for all students and for demographic sub-groups. Principal findings suggest that: (1) controlling for other factors, men are more likely than women to graduate during the first five years, but this advantage dissipates over time; (2) controlling for gender and age, foreign students are more likely than whites to graduate; (3) the comparison of minority and white students is more complex, with younger minority and white students having similar graduation probabilities, while older minority students are less likely to graduate than older white students; (4) age-at-entry was not found to have a strong effect on the probability that white or foreign students will graduate during the first seven years of the program; (5) male and minority students have an elevated risk of non-completion, but the overall probability of non-completion is low until after year seven.;Chapter 2 introduces variables on admissions characteristics, time-status and residency. Principal findings suggest that: (1) admission scores have a moderate negative relationship to graduation, but cannot be differentiated from zero once the effect of prior degrees are introduced; (2) part-time-status has a negative effect, but results in a larger decrement for men than women; (3) registering in absentia has a positive relationship to the probability of graduation, but the positive relationship for women is stronger and continues for a longer period of time.;Chapter 3 examines rival hypotheses to the main findings on the probability of graduation: that differences are due to changes among cohorts; or that variation is related to academic area.