Selection in the Lab: A Network Approach

We study the selection problem in economic experiments by focusing on its dynamic and network aspects. We develop a dynamic network model of student participation in a subject pool, which assumes that students' participation is driven by the two channels: the direct channel of recruitment and the indirect channel of student interaction. Using rich recruitment data from a large public university, we find that the patterns of participation and biases are consistent with the model. We also find evidence of both short- and long-run selection biases between males and females, as well as between cohorts of students. Males tend to have higher participation rates than females, and participation rates tend to decrease with a cohort's age. Our empirical findings confirm that dynamic and peer effects play an important role in shaping the selection problem. Our model allows us to reconcile some of the mixed results in previous studies.