Climate change represents a major global challenge. Responding to climate change, interventions have been implemented to encourage sustainable behavior. Such interventions may not only promote the target behavior but also increase (positive spillover) or reduce (negative spillover) non-targeted sustainable intentions and behaviors. This pre-registered meta-analysis integrated the experimental research on environmental spillover. Database searches in several languages supplemented by searches to

more »

... retrieve unpublished literature yielded 61 aggregated effect sizes from 37 studies and 28 articles (N = 26,749 unique participants). The three-level Bayesian meta-analysis provided moderate support for a small positive spillover on intentions, δ = 0.14, 95% CrI [-0.02, 0.29], and strong support for no spillover on behaviors, δ = 0. Moderator analyses showed that positive spillover on intentions and behaviors combined was more likely when interventions thwarted rather than preserved individual choice, and when they addressed normative rather than personal gain goals. Spillover on intentions and behaviors combined was similar in laboratory, online, and field studies, supporting generalizability across settings. Based on this meta-analysis, we developed the PRO guidelines which emphasize statistical power, reporting, and open science to set standards for robust spillover research. The Bayesian approach allows for robust conclusions about the presence and absence of spillover as well as continuous updating with new evidence. We hope that the Bayesian approach combined with the PRO guidelines paves the way toward a sustainable overview of robust spillover research that independent researchers can easily update.