Bruce Wydick on the Impact of giving away TOMS Shoes: He gives kudos to TOMS for being open for evaluation and being responsive to findings, but what caught my eye was this observation: "The bad news is that there is no evidence that the shoes exhibit any kind of life-changing impact,..."
The WaPo Q&A with Brendan Nyhan on making scientific research more trustworthy. I learned a new term HARKing, which apparently stands for Hypothesizing after the results are known. Life would be so boring without harking, which would mean that there were no unexpected results, ever...
From the CGD Blog, Do Mobile Phone Surveys Work in Poor Countries? My takeaway: not yet.
The NYT Opinion Piece Where Terrorism Research Goes Wrong was trending for a second in my tiny Twitter circles last week and it looked weak. Eli Berman of UCSD and co-authors provide a good response in the Monkey Cage blog.
Finally, J-PAL's March 2015 Research Insider seems to confuse Ben Olken and me and, what the hell, forms a composite: Ben Olzer! I am sure he is psyched:
"Why is Difference-in-Difference Estimation Still Popular in Experimental Analysis?"
Writing in a recent Development Impact blog post, Ben Olzer asks why difference-in-difference estimation is still popular for analysis within RCTs that include baseline and follow up surveys. Instead, Olzer discusses David McKenzie's paper showing that researchers can maximize statistical power by using ANCOVA when analyzing baseline data with follow up surveys. See the full blog post here.