Higher Education and Cultural Liberalism: Regression Discontinuity Evidence from Romania

Abstract

Some studies suggest that university attendance exerts a liberalizing effect on attitudes toward cultural issues such as sexuality and sexual identity, prostitution, drug addiction, abortion, capital punishment, divorce, parenting, gender, race, religion, science, and technology. Other studies find only a weak effect or no effect. Divergent findings may stem from research designs that rest largely on observational data when assignment to treatment is nonrandom and there are many threats to inference. In this study, we enlist a unique research setting in Romania that allows for a fuzzy regression discontinuity design separating those qualified to matriculate to university from those unqualified to do so. We find that university attendance contributes to a more liberal outlook as measured by our composite index, corroborating the main (preregistered) hypothesis. Evidence for subsidiary hypotheses is mixed.