Chris Blattman

Search
Close this search box.

IPA’s weekly links

NudgingEffectSizes

Guest post by Jeff Mosenkis of Innovations for Poverty Action

  • A very cool job market paper and explanatory thread, from Ph.D. candidate Matthew Klein. He, Bradford L. Barham, and Yuexuan Wu, link women’s household bargaining power to malaria rates in Malawi. They find that a one standard deviation increase in a woman’s household bargaining power implies a 40% reduction in chances that anybody in the household contracts malaria. They caution their ability to infer why this works is limited in their data set, but it offers an intriguing route to test to decrease malaria rates, something the WHO just called for.
  • The review paper “How effective is nudging? A quantitative review on the effect sizes and limits of empirical nudging studies”, the source of the figure above, reminds us that nudge effect sizes aren’t huge (except for changing defaults). But that never should have been the claim – they’re just usually cheap and scaleable, if they involve things like rearranging information on an existing form, or another way of presenting information differently.
    • It reminds me of the paper “Do The Effects of Social Nudges Persist? Theory and Evidence from 38 Natural Field Experiments”, which reviews several studies showing that effects of habit change interventions often fade after the intervention ends. If I understand it, they then look at a series of field experiments on home energy reports (those mailings from the power company about your energy use) which is considered a very successful nudge. First, they mention that it only reduces power use by 2.4 percent (which over a big population is nothing to shrug at), and that the effects fade by over half after the program ends. The remaining effects are driven by people who’ve invested in “physical capital” (more energy-efficient appliances), which in a sense is like changing the default. Big behavior change is hard, but if you can change the setup around people to reduce the continuing challenge, you’ll have a better chance.
  • A nice read via David McKenzie’s links last week: “Humans of Policy Research” the personal perspectives of policymakers in Ethiopia who are collaborating with researchers, and what they hope to get out of it.
  • How do we liberate agriculture and development from academic preferences? By Charles Dhewa, on the FP2P blog, argues slow academic literature accumulation led by professors far away is not a good way to either capture or retain what’s actually known by the people who know their areas best. But there’s been elite academic capture of development agencies which then privilege those kinds of limited distant knowledge when it comes to directing development efforts.

Literature review cannot explain emerging issues in agriculture and health. For instance, many countries in Southern Africa are now experiencing crop and livestock diseases like Tuta AbsolutaFall Army Worm and January diseases as well as several human ailments which did not exist a few years ago. You cannot find useful literature on these diseases going back 10 years to 50 years. When literature review is prioritized ahead of real-time knowledge, countries in the Global South end up recycling old ideas at the expense of new ideas that speak to the evolving context. Countries end up doing endless policy reviews instead of developing new fluid policies.

He suggests an alternative way of thinking about knowledge management that collects what people are currently talking about. (Sorry about the formatting break)

  • Duncan’s weekly audio summary podcast of his daily FP2P blog posts (Apple) are a great quick way to get the point of that week’s posts or to find out which ones are of interest to go back and read in full.
  • IPA and J-PAL have merged their RCT data hubs on Dataverse, and now have 149 data sets for free download there. They list some of their most recent and most popular downloads in the blog announcement.
  • And, this important replication came out this week and thankfully is open access. Really, the paper is a gift that keeps on giving: