Social information and educational investment : nudging remedial math course participation

In the context of a voluntary remedial math course for university students, we conduct randomized field experiments to study whether social information can help counteract low participation rates. Incoming students receive postal invitation and reminder letters that inform them about past sign-up rates and past evaluations on the usefulness of the course, respectively. On average, we find that neither of the two interventions increases sign-up or participation. Heterogeneity analyses provide evidence that a targeted provision of the invitation letters can be beneficial: i) By increasing the salience of the course, they raise attendance among late enrolling students, which in turn increases their academic performance in the first year of studies. ii) Students whose ex-ante signup probability falls just short of the descriptive norm increase sign-up and participation in response to the social information, while the opposite is true for students whose signup probability exceeds the norm.