The value of multi-proxy experiments to study pro-environmental behavior

In the study of pro-environmental behavior, experimental studies are essential for identifying causal relationships and underlying psychological mechanisms. However, despite the multi-dimensional nature of the construct “pro-environmental behavior”, experimental studies often adopt a single, fixed, proxy measure (e.g. one specific pro- environmental decision) to capture this construct. The problem herewith is that this approach ignores the idiosyncratic nature of the specific proxy used, undermining the reliability of study conclusions on pro-environmental behavior in general. In contrast herewith, the proposed multi-proxy experimental approach makes use of a wide variety of proxy measures to operationalize pro-environmental behavior: participants are randomly assigned to one out of a large set of different proxy measures (i.e. a variety of different pro-environmental decisions). This approach preserves the strengths of classic experimental design while increasing the reliability of study conclusions. Moreover, it allows for the collection of additional information about specific characteristics of the proxy measures used (also post hoc), which can then in turn be employed as moderating variables. This may not alone help identify the specificity of the experimental effects, but can also be used to test different conceptual models against each other. We provide a roadmap to implement the multi-proxy experimental approach.