Guest post by Jeff Mosenkis at Innovations for Poverty Action.
As we expect never to hear from Chris again, here are IPA’s weekly links:
- With the US, Japan, and North Korea appearing to be the only countries not on board the new China-led Asian Infrastructure Investment Bank, US Treasury Secretary Jack Lew seems to have softened the US stance against it.
- The new Freakonomics podcast talks with MIT & J-PAL’s Amy Finkelstein and physician & MacArthur “Genius” Jeffrey Brenner about using randomized controlled trials in policy, and an interesting new collaboration between the two to fix Camden, New Jersey’s chronic healthcare problems with data.
- One of the difficult parts for an org participating in a randomized controlled trail is not being able to offer the program to the control group, since they have targets to meet also. David McKenzie suggests adding a third “wait list” group, and if takeup is less than 100%, offering the remaining spots to people in this third group, who won’t be part of the study itself. It helps the implementer do their job without risking the control group (h/t Nathanael Goldberg).
- Justin Sandefur takes his teacher’s red pen to the U.N. Sustainable Development Goals. In the accompanying blog post, he explains what we learned from the last round about concrete measurable goals, and why gender equity language is a bad idea.
- In what must be some sort of record for most information in fewest words, David Evans has 40+ great one-sentence summaries of papers from the recent Center for the Study of African Economies conference in Oxford.
- Lessons on interpreting causation and correlations for journalists:
-J-PAL’s Rachel Glennerster has five lessons for interpreting research from a recent training for journalists in Chile (and some choice words for NPR).
-And from Buzzfeed: Dow Falls More Than 1% As Zayn Leaves One Direction
“The Dow began falling prior to the announcement of Zayn’s departure, suggesting some traders may have been tipped off early.”
Those two links together became one of our most popular recent Facebook posts, because apparently the internet is full of good data journalism fans.