Notwithstanding the unprecedented attention devoted to reducing poverty and fostering human development via scaling up social sector spending, there is surprisingly little rigorous empirical work on the question of whether social spending is effective in achieving these goals. This paper...

Social experiments are powerful sources of information about the effectiveness of interventions. In practice, initial randomization plans are almost always compromised. Multiple hypotheses are frequently tested. "Significant" effects are often reported with p-values that do not account for...