A meta -analysis of structural differences in framing problems

A meta-analysis was conducted to determine the effects of conceptual and operational differences in complexly diverse framing problems. Many researchers have explored the effects of framing, at first finding clear and strong results consistent with Tversky and Kahneman\u27s framing effect. Recently, and with increasing frequency, framing studies report apparently inconsistent or null results. Other researchers have acknowledged this general trend toward inconsistency and smaller effect sizes in the framing literature. The current research, based on over 50,000 participants from 128 published reports included 286 effect sizes indicate that the size of the framing effect overall is small to moderate in size. The problems used in framing studies have become less homogeneous and more complex over the years, resulting in great diversity among framing definitions, operationalizations, and dependent variables to such an extent that reported framing effects may or may not reflect the same phenomenon. The present study explored the magnitude of the framing effect based on a typology of structural differences in framing problems. Results indicate that important determinants of the size of the framing effect include the location of the frame manipulation in the decision, the length of the problem, problem domain, decision task, equivalency of frames, methodological variations, and gender