Global Evidence on the Motives for Sustainable Behaviors

Behavioral change is an integral part of climate change mitigation andadaptation. However, many interventions that encourage individuals to make moresustainable choices do not lead to the desired behavioral change. To tackle thischallenge, we take a step back and focus on understanding the complexmotivational structure behind sustainable behaviors. We (i) identifyindividually relevant motives for sustainable behaviors, (ii) show that thesemotives can be predicted from demographic variables and individuals’ economicpreferences, and (iii) discuss how these insights can be used for designing moreefficient behavioral interventions.We base our analysis on rich multinational survey data collected from 1,954respondents across the United States, China, and the Netherlands. Our approachis based on a novel large language model. We identify 14 motives in 14,262individual responses to open-ended questions that elicit reasons for sevensustainable behaviors. For example, the motive Health Comfort is the mostrelevant and most general motive for the behaviors we studied, ranging fromshowering time to frequency of flying. Individually relevant motives areassociated with demographic variables such as gender and age, but also economicpreferences. For respondents with a lower discount rate, for example, the motiveElectricity Energy is very important, whereas Want Feel is less relevant. Ouranalysis also reveals striking similarities regarding the relevance andbehavioral specificity of motives across the cultures.When comparing the relevant motives across behaviors, we uncover that behaviorswith more overlapping motives are also more positively correlated. This findingoffers an explanation of spillover effects caused by implicitly or explicitlyappealing to behavioral motives that are relevant for multiple behaviors.By investigating the motives for multiple behaviors and cultures simultaneously,we are able to generate generalizable and novel insights of cross-culturalimportance. These results offer promising pathways towards designing moreefficient, targeted interventions focusing on individual motives.