Endogeneity in water use behaviour across case studies of household water treatment adoption in developing countries

Abstract

Endogeneity or reverse causality in regression analysis results in biased estimation of the effects of independent variables on the dependent variable and leads to inaccurate interpretations. However, the biased estimation of the effects of psychological factors on water–related behaviours are rarely discussed. This study investigated the endogeneity of psychological factors in water-related behaviour using an instrument variable (IV) approach. Data from eight household water treatment (HWT) studies in Asia, Africa, and South America were utilized. A combination of several socio-economic characteristics, such as education and accessibility, was used as a control variable and three psychological factors, i.e., perception of risk, attitude towards HWT, and social norms, were used as the predictors of the adoption of HWT. Variables related to institutional quality of the countries, based on the World Governance Indicators of the World Bank were used as IVs to predict the psychological factors. The results suggest that endogeneity exists in water-related behavioural studies. Institutions were found to be valid IV for psychological factors attitude and norms, but not for the perception of risk. This suggests that the institutional quality influences households’ attitude and norms regarding behaviour. If the feedback effect of actual behaviour on the psychological factors were not considered, the effects of attitude and norms on HWT adoption were underestimated by 59% and 40%, respectively. Finally, despite the challenge of finding valid IV, the endogeneity effect of psychological factors needs to be controlled when estimating the effect of psychological factors on water-related behaviour in future water-related behavioural studies.