Weekly links October 19: Men are from Mars, Women are from Venus – but only if you live in a rich and equal country, updates in randomization inference, graduation programs vs cash, small clusters not such a problem?
This page in:
- At VoxDev, Sarah Baird, Berk, and I explain why the Econ 101 leisure-labor trade-off model can lead us so far astray in considering the labor impacts of cash transfers of different types (UCTs, CCTs, remittances, pensions, etc.) – and our view of what research needs to measure going forward.
- On the Education for Global Development blog, Holla, Molina and Pushparatnam ask us what The Wire can teach us about psychometrics – with examples of what things to look for in “validating” a test score.
- Women tend to have preferences that are more pro-social and are less risk-taking and less patient on average (the latter I was surprised by, but the difference is not so large). In Science this week, Falk and Hermle look at how these gender differences in preferences are correlated with economic development and gender equality using survey data on 80,000 individuals in 79 countries! They find gender differences increase with GDP and with gender equality – their explanation is “As suggested by the resource hypothesis, greater availability of material resources removes the human need of subsistence, and hence provides the scope for attending to gender-specific preferences. A more egalitarian distribution of material and social resources enables women and men to independently express gender-specific preferences.”
- At the CGD blog, Patel, Sandefur and Subramanian argue that everything you know about cross-country convergence is now wrong – “While unconditional convergence was singularly absent in the past, there has been unconditional convergence, beginning (weakly) around 1990 and emphatically for the last two decades.”
- At Vox, Dylan Matthews summarizes recent evidence on graduation programs and when a bundle of stuff might beat cash.
- This week’s DeclareDesign blog looks at different estimators for standard errors when you have a small number of clusters.
- New in the comments – Simon Heβ discusses changes in his randomization inference code that now allows for doing pairwise tests when there are more than 2 treatment groups. More generally Simon has done amazing public service in answering a number of queries in the comments thread here.
- Job openings at the Gender Innovation Lab.