This paper provides a survey of six widely used non-experimental methods for estimating the impact of programmes in the context of developing economies (instrumental variables, regression discontinuity, direct matching, propensity score matching, linear regression and non-parametric methods, and difference-indifferences), and assesses their internal and external validity relative both to each other and to randomized controlled trials. While randomized controlled trials can achieve the highest

more »

... gree of internal validity when cleanly implemented in the field, the availability of large, nationally representative datasets offers the opportunity for a high degree of external validity using non-experimental methods. Whereas these methods are often presented as competing alternatives, we argue that each method has merits in some context and that experimental and non-experimental methods are complements rather than substitutes.