· Andrew Gelman (correction Dean Eckles posting on Gelman's blog) on t-stats, the width of confidence intervals, and what this means if you care about the magnitude of an effect and not just statistical significance: “Perhaps most simply, with a t-statistic of 2, your 95% confidence intervals will nearly touch 0. That is, they’re just about 100% wide in each direction. So they cover everything from nothing (0%) to around double your estimate (200%)….even at t=5, the halved point estimate is still inside the 99% CI. Perhaps this helpfully highlights how much more precision you need to confidently state the size of an effect than just to reject the null….It is only when the statistical evidence against the null is overwhelming — “six sigma” overwhelming or more —that you’re also getting tight confidence intervals in relative terms. Among other things, this highlights that if you need to use your estimates quantitatively, rather than just to reject the null, default power analysis is going to be overoptimistic.”
· “Housing Manhattan in two-story buildings” – Saarthak Gupta in Asterisk mag on how land use restrictions in Mumbai restricts space per person and is an outlier for a city of its size, while stringent rent control policies disincentivize landlords from investing in upgrades. “And much more than cities in the state of California, Mumbai’s growth is damaged by extreme restrictions on building and rentals — restrictions that only grow more binding every year. By limiting the effects of agglomeration, the city’s building and rental restrictions exert an immense drag on the city’s ability to drive growth on the national level.”
· In the Economist’s 1843 magazine, “how poor Kenyans became economists’ guinea pigs”: “To date, at least 270 rcts have been conducted in Kenya, though this is likely to undercount the real number (27 are currently under way in Busia alone)… Each county in the region has its own specialty: textbooks, mosquito nets and deworming in Busia; cash transfers in Siaya. There have been so many in Busia in recent years that it can feel as though there are few aspects of everyday life left to examine. Latterly, the region has become so saturated with research that economists have had to look farther afield for participants.” It discusses some of the lessons learned from RCTs, some criticism of what has been learned and of ethics, the lives of field workers, and more. And this on spillovers from a cash transfer study: “The distinction between the treatment and control groups sometimes became porous: in one village, the treatment group agreed to share their cash with the control group behind the researchers’ backs. The RCT also created obvious and abiding disparities between and within communities. One of the treatment villages lay just over the road from a control village. Every day, those in the control village caught a glimpse of what their lives might have been like had the randomisation algorithm smiled on them.” Yet despite these anecdotes, there is really nothing new in the article.
· Funding call:
The Agency Fund has an open window with submissions due March 15
Join the Conversation