Course Withdrawal Dates, Tuition Refunds, and Student Persistence in University Programs

University policies, such as the last date for withdrawal from courses without academic penalty and tuition refund schedules, vary across universities and over time. Data on those policies at 38 Canadian universities, 1997-2005, are used to estimate their relation to whether students: (i) continued in their first university program, (ii) switched to another program or institution, or (iii) exited post secondary education. The Youth in Transition Survey, Cohort B, provides data on students' characteristics and education outcomes. Controls for students' characteristics and backgrounds, cohort year effects, and university characteristics are included. Students enrolled in schools with more generous tuition refund schedules are less likely to exit post secondary education between second and third year, but the result is not robust to the inclusion of individual university fixed effects. Students facing later withdrawal deadlines are more likely to switch (transfer) to other programs or institutions between their first and second year, in both the university characteristics and university fixed effects specifications.