Revisiting Catholic School Effects: An Application of a Mechanism-Based Approach

Yourim Chai, University of California, Los Angeles

This study examines school sector effects at the elementary school level. Revisiting this classical issue is warranted for two reasons. First, school reform in recent period implies institutional changes in public and Catholic schools, which may yield changes in sector effects. Second, previous studies provided insufficient explanations why the sector effects exist. By applying a front-door criterion (Pearl 2009), I examine the important school characteristics for student outcomes and identify which school characteristics bring about schooling effects at the elementary school level. I compare this approach with more frequently used approaches such as a back-door criterion and mediation analysis, which illustrates the strengths of the new approach. For this purpose, I use the Early Childhood Longitudinal Study, Kindergarten Class of 1988-89 (ECLS-K). This study enriches our understanding of school sector differences by specifying causal mechanisms and provides an alternative way of estimating causal effects beyond average treatment effects.