Paying teachers for performance may have very different effects on the effort that current teachers put in and on the kinds of people that choose to become teachers. In a novel experiment by Leaver and others in Rwanda, advertisements in some randomly selected communities offered candidates pay-for-...
Rwanda’s performance-based incentives were effective for some indicators, but unconditional financing also induced improvements. The incentive effects persisted in the medium-run and as the program was scaled-up.
We examine alternative strategies for targeted sampling of health clinics for independent verification. Our results indicate that machine learning methods, particularly Random Forest, outperform other approaches and can increase the cost-effectiveness of verification activities.