A positive relationship between pro-environmental behavior and subjective well-being has been used to argue for a “double dividend”, i.e. the narrative that pro-environmental behavior is beneficial for both environment and individual, when measured in the metric of subjective well-being. Our paper argues that the (causal) evidence base for such a narrative is far too weak. We suggest methodological improvements to strengthen the credibility of multivariate regression analyses with observational data. Directed acyclic graphs help with a theoretically grounded selection of control variables and equivalence tests (and associated power considerations) help interpreting null results. Illustrating both for a novel data set from a medium sized German municipality ( ), we find no evidence for a positive relationship between pro-environmental behavior and life satisfaction. Equivalence tests robustly reject the null hypothesis of a true effect size larger than even half that from a recent meta-analysis ( ). We discuss the implications of these findings and conclude that this dampens enthusiasm for the narrative of the double dividend.     «

A positive relationship between pro-environmental behavior and subjective well-being has been used to argue for a “double dividend”, i.e. the narrative that pro-environmental behavior is beneficial for both environment and individual, when measured in the metric of subjective well-being. Our paper argues that the (causal) evidence base for such a narrative is far too weak. We suggest methodological improvements to strengthen the credibility of multivariate regression analyses with observational...     »