Designing Effective Interventions

Most interventions fail to change behavior. We argue that the reason for this failure is that the interventions do not adequately address the type of the underlying problem. We develop a systematic, parsimonious, and generalizable framework that uses a simple survey tool (anamnesis) to identify three types of fundamental problems: awareness, intention, and implementation problems. We then test in an online experiment with 7,500 subjects whether our framework can predict the effectiveness of three typical interventions (reminders, incentives, simplifications) that are designed to address a specific fundamental problem. As hypothesized, we find that the interventions' effectiveness varies substantially between the different settings, but that our framework can predict this heterogeneity remarkably well: On average, a predicted effectiveness of 10% corresponds to an actual effectiveness of 8.92%. Choosing an intervention based on our framework increases an intervention's effect size by around 58% compared to randomly choosing one of the tested interventions. We also apply our framework to predict the findings of a large-scale megastudy about booster vaccinations, providing evidence for its external validity for designing effective interventions.