Estimating the Causal Effects of Organized Activities: Testing the Efficiency of Propensity

Ryan D. Heath (SWK) PI, Bhavneet Walia (PH) co-I
2019-2020 Falk Tenure-Track Assistant Professor Seed Grant Award

This study tests the utility of propensity score matching to study the effect of organized activity participation in adolescence on long-term outcomes.

The potential of organized activities has taken on new importance in recent years as content that can help support the healthy development of youth. Organized activities – including after-school programs, extracurricular activities, and summer programs – receive substantial federal, state and private investments, and rigorous methods are needed to assess their impact on young people’s educational and labor market outcomes. Research on organized activities for youth has been limited by mixed findings, self-selection, and the inability to draw causal inferences. However, as it is difficult to conduct randomized studies on these programs, there is a need to utilize other methods of modeling causality.

Thus, this study aims to test the feasibility and efficiency of propensity score matching in research on organized activities. While a few studies on organized activities have incorporated propensity score matching, fewer have assessed whether or not this method produces more efficient and less biased estimates of the effects of organized activity participation. To do so, this study will utilize data from the Educational Longitudinal Study of 2002 to estimate the effect of organized activity participation in high school with educational and labor market outcomes in young adulthood, and whether propensity score matching produces more efficient and less biased estimates as compared to traditional regression methods.