· The Chicago IGM forum asks a panel of top economists their opinions on RCTs: whether “Randomized control trials are a valuable tool for answering some long unsettled questions in development economics research” – 100% agree or strongly agree; and whether “Randomized control trials are a valuable tool for making significant progress in poverty reduction” – 79% agree or strongly agree, 12% are uncertain, and Angus Deaton strong disagrees.
· Michael Clemens and Jimmy Graham provide three facts you aren’t hearing much about regarding emigration from Central America, over at the CGD blog.
· Alex Coppock offers a useful demonstration using Declaredesign of the problem with the all too common practice of looking at treatment heterogeneity by examining if effects are significant in one subgroup and not another – he shows when effects are moderate and actually constant across groups, this can lead us to falsely infer heterogeneity almost 50% of the time.
· Are RCTs ethical? Two new pieces to read:
o On Project Syndicate, ethics guru Peter Singer joins Arthur Baker and Johannes Haushofer in discussing ethical issues with randomized poverty alleviation experiments – they discuss issues of asymmetry of power, the risk of harm to non-participants, and testing interventions which are clearly better than nothing.
o The Data Colada blog discusses issues with that PNAS paper that found people are averse to experiments – their take is that instead what people are objecting to is either one of the policies that the experiment is choosing between, or to additional information revealed in the description of an experiment (e.g. that a Dr doesn’t know what care to provide his patients).
· Reminder: our blog your job market paper series is currently taking submissions.
Join the Conversation