Addressing endogeneity in estimating the effect of social capital on psychological health

Abstract

This study examines the relationship between social capital and psychological health. Endogeneity concerns are addressed through the development of a two-stage residual inclusion (2SRI) model. I provide robust evidence that levels of social trust and support from friends significantly affect self-reported psychological health. Estimates obtained relating to the effect of these variables using the 2SRI method as well as two-stage least squares greatly exceed those obtained through standard ordered logit and OLS models. This illustrates the importance of controlling for endogeneity bias in any future studies of the relationship between social capital and psychological health.

Keywords:

• social capital
• psychological health
• instrumental variables

JEL Classification:

• I10
• C26