Syndicate content

Experimental design

Guest Post by Ken Leonard: Gender and Biological Differences between the Sexes

Women are less likely to occupy the top paying jobs in developed economies, in part because they are less competitive than men. A whole series of laboratory experiments has detailed the gap in competitiveness between the average woman and the average man, even when women are just as good, if not better than men. Is this result due to the fact that women are biologically female, or the fact that they are socialized as female? Although we often alternate between gender and sex in describing males and females, they are not strictly the same.

Designing experiments to measure spillover effects

Berk Ozler's picture

Many programs affect those who were not directly targeted by the intervention. We know this for medical interventions (e.g. deworming: Kremer and Miguel 2004); cash transfer programs (e.g. PROGRESA: Angelucci and de Giorgi 2009); and now voter awareness programs (Giné and Mansuri 2011).

Sampling weights matter for RCT design?

Berk Ozler's picture

One of the most important things while designing an intervention is to try to ensure that your study will have enough statistical power to test the hypotheses you're interested in. Picking a large enough sample is one of a variety of things to increase power. Another is block stratified randomization, of which paired randomization is the extreme.

If you want her business to grow, don’t just give her cash

David McKenzie's picture

That’s one blunt message from my new working paper with Marcel Fafchamps, Simon Quinn and Chris Woodruff, which replicates in Ghana a study that Chris and I had previously done in Sri Lanka with Suresh de Mel. In the new experiment, we take almost 800 microenterprises in urban Ghana, and randomly divide them into treatment and control groups.