The effects of high-quality student mentoring

- I examine the impact of a college mentoring program on students study performance.
- A quasi-experimental design allows to identify the causal program impact.
- The paper finds larger effects than previous studies evaluating mentoring programs.
- The program has some high-quality features which can explain the larger impact.

This paper evaluates the impact of a university mentoring program for first semester economics and management students on academic performance. For identification, I use a difference-in-differences approach with students in a similar degree program as control group. The mentoring program decreases failure rates by 15.4 percentage points. These large effects can be explained by quality features of the mentoring program, including graduate mentors, a low mentor-to-mentee ratio and compulsory, pre-scheduled, face-to-face appointments.