What leads government officials to use impact evidence?

Although the amount of policy-relevant academic research has grown in recent years, studies still find that policy practitioners seldom employ such research in their decisionmaking. This study considers potential methods for increasing government officials’ use of academic studies (impact evidence). We investigate how administrative accountability mechanisms as suggested by principal-agent approaches – screening, monitoring, autonomy and sanctions – correlate with practitioner engagement with impact evidence. Original survey data from 300 government officials in two developing countries, Peru and India, suggest that all four mechanisms are correlated with self-reported interest in or use of impact evidence. When we measured the actual use of such evidence on a website we created to facilitate that outcome; however, we found that only sanctions (income) correlate with actual use. These findings highlight the potential of administrative accountability to increase bureaucrats’ use of impact evidence but also warn of possible limitations.