What makes a productive Ph.D. student?

This paper investigates how the social environment to which a Ph.D. student is exposed during her training relates to her scientific productivity. We investigate how supervisor and peers' characteristics are associated with the student's publication quantity, quality, and co-authorship network size. Unique to our study, we cover the entire Ph.D. student population of a European country for all the STEM fields analyzing 77,143 students who graduated in France between 2000 and 2014. We find that having a productive, mid-career, low-experienced, female supervisor who benefits from a national grant is positively associated with the student's productivity. Furthermore, we find that having few productive freshman peers and at least one female peer is positively associated with the student's productivity. Interestingly, we find heterogeneity in our results when breaking down the student population by field of research.