Clustered Local Average Treatment Effects: Fields of Study and Academic Student Progress

Multiple unordered treatments with a binary instrument for each treatment are common in policy evaluation. Existing instrumental variable (IV) methods identify local average treatment effects (LATEs) under assumptions that may be too restrictive on individual behavior. This paper introduces an IV strategy that identifies an interpretable weighted average of LATEs under relaxed assumptions, in the presence of clusters with similar treatments. We analyze the effect of fields of study on academic student progress, and find violations of the LATE assumptions, clusters with similar fields, treatment effect heterogeneity across students, and significant differences in student progress due to fields of study.

Keywords: Multiple treatments, Instrumental variables, Treatment clusters, Field of study

JEL Classification: C36, I23

Suggested Citation: Suggested Citation

Nibbering, Didier and Oosterveen, Matthijs and Silva, Pedro Luis, Clustered Local Average Treatment Effects: Fields of Study and Academic Student Progress (October 27, 2022). Available at SSRN: https://ssrn.com/abstract=4058313 or http://dx.doi.org/10.2139/ssrn.4058313