Exploring the Effect of Participation in Sports on the Risk of Overweight
Applied Research in Quality of Life, 10(3), 381-404, 2015
Posted: 2 Sep 2014 Last revised: 26 Apr 2016
Date Written: September 2, 2015
A growing number of academic studies on public health have highlighted that participation in sports can help prevent people from becoming overweight. Estimates of the effect of sports on the risk of overweight based on sample surveys have primarily controlled for a number of socio-behavioural and demographic observed covariates and neglected to control for both observable and unobservable confounders. The purpose of our study is to estimate the effects of sport activity on risk of overweight by accounting for both confounders. We use microdata collected by the Italian National Statistical Office for 2011 during the Daily Living Conditions Survey of Italian households. The sports-overweight relationship is estimated after controlling for a number of observables using univariate probit modelling. A recursive bivariate probit approach is used to account for both observed and unobserved confounders (better known as endogeneity). Both modelling strategies are performed within a semiparametric and parametric framework. The main findings of the proposed sensitivity analysis suggest that participation in sport activities has a statistically significant negative effect on the risk of overweight and that endogeneity might not be as important as initially suspected. We conclude the presentation of our results by suggesting some policy recommendations and future extensions of the analysis.
Keywords: Average treatment effect on the treated; Endogeneity; Overweight; Probit models; Recursive bivariate probit models; Sport activity
JEL Classification: C3; C36; I12; I18
Suggested Citation: Suggested Citation