Using framed field experiments to evaluate real-world policy interventions : a case study on changing environmental preferences

Many policy interventions in the environmental sector aim at changing environmental preferences, because these provide the basis for adopting environmental conservation behavior or technologies. To evaluate these changes, standard measurement tools in survey-based impact evaluations have their limitations. We discuss the potential of framed field experiments as an alternative to attain unbiased outcome measures, and present a case study from an impact evaluation in the Colombian coffee sector. While clear advantages exist, we show that indicators from framed field experiments require substantial sample sizes to provide well-powered results. Moreover, preference indicators are highly context specific, which calls for an elaborated framing to attain the intended kind of preferences.