Guest post by Jeff Mosenkis of Innovations for Poverty Action.
- Max Roser reminds us that the data sets above (known as Abscombe’s quartet) all have the same mean and variance, so you should always plot your data. Over at the IFPRI blog Alan de Brauw says don’t worry about R-squared in randomized controlled trials, they’re not designed for that, just use the treatment effect .
- After a week of pressure largely from the people of Burkina Faso, the recent coup there was undone (h/t Nicolo T).
- Worm Wars, the video. Apparently doctor, health explainer, and Incidental Economist blogger Aaron Carroll was getting a lot of requests to explain the deworming controversy, replications, and reanalysis, and did a really good job.
- A new paper (ungated here) by Cilliersa, Dubeb, & Siddiqi find in 60 Sierra Leone villages, the presence of a white foreigner in a classic “dictator” behavioral game increased allocations 19 percent, and this effect was related (negatively) to how much outside aid the village was exposed to. More discussion from New York Magazine’s Science of Us.
- The Busara Center for Behavioral Economics (an amazing resource for behavioral scientists) in Nairobi and IPA-Kenya had a volleyball match last week resulting in some serious online Econ Trash Talk (click to expand the conversations).
And last week in New York, a competition of worshipers intersected literally when the huge lines to see the Pope and at the Apple store for the new iPhone crossed on 5th Avenue. There’s been at least one time when choosing an apple over God didn’t work out so well…
Dueling lines 5th Ave NYC. For the Pope and for iPhone. pic.twitter.com/p8vCpVMISr
— Ed Baig (@edbaig) September 25, 2015