Syndicate content

David McKenzie's blog

Another reason to prefer Ancova: dealing with changes in measurement between baseline and follow-up

David McKenzie's picture
A few months ago, Berk blogged about my paper on the case for more T, and in particular, on the point that Ancova estimation can deliver a lot more power than difference-in-differences when outcomes are not strongly autocorrelated.

Weekly links June 19: The anthropological argument for just giving cash, slow thinking, what those classic psych experiments really say and more…

David McKenzie's picture

Randomizing Competition: allowing CCT recipients to get more goods for their money

David McKenzie's picture
The Dominican Republic’s Solidaridad conditional cash transfer program provides its monetary transfers to poor families in the form of a debit card that can only be used at a network of grocery stores affiliated with the program (it does this in part to ensure they spend the money on food). The typical monthly transfer is about $36, which is 17% of median monthly food expenditure.

Weekly links June 5: what to read, enhancing trust in your data, what to call those skills, and more…

David McKenzie's picture

Pitch Perfect? An Update on the SME Ideas Competition

David McKenzie's picture
One of the benefits of impact evaluation has been fostering more collaboration between researchers and operational staff implementing projects. However, at present this collaboration largely happens once the project itself has been decided upon. To try and get more researcher ideas influencing what projects get done in the first place, I partnered with the World Bank’s Innovation and Entrepreneurship unit to hold a competition for new ideas for SME projects, financed by DFID under part of a Strategic Research Program (SRP) program grant (I previously blogged about the launch).
As a pilot initiative, our key questions were:
  • Is there a supply of new ideas that researchers have that are not currently being tried? Will researchers take the time to put these ideas forward?
  • Is there a demand from operational teams and governments working on SME projects for new ideas in this space?
  • Can we form matches between this supply and demand?

Is optimization just re-randomization redux? Thoughts on the recent ‘don’t randomize, optimize’ papers.

David McKenzie's picture
A couple of weeks ago, Berk blogged about a new paper by Bertsimas, Johnson and Kallus which argues that instead of randomization, it can be superior for power to choose the best of all possible allocations of subjects to treatment and control, where best is defined in terms of minimizing discrepancies in the mean and variance of the two groups.

Weekly links May 8: rainfall revisited, rethinking key results with long-term data, son preference, and more…

David McKenzie's picture