We study whether living in a student room as a tertiary education student (instead of commuting between one's parental residence and college or university) affects exam results. To the best of our knowledge, we are the first to study this relationship beyond cross-sectional analysis. That is, we exploit rich longitudinal data on 1,653 Belgian freshmen students' residential status and exam scores to control for observed heterogeneity as well as for individual fixed (or random) effects. We find that after correcting for unobserved heterogeneity, the association found in earlier contributions disappears. This finding of no significant impact of living in a student room on exam results is robust for other methods used for causal inference including instrumental variable techniques.