Andrew Gelman

Andrew Gelman is a professor of statistics and political science and director of the Applied Statistics Center at Columbia University. He has received the Outstanding Statistical Application award from the American Statistical Association, the award for best article published in the American Political Science Review, and the Council of Presidents of Statistical Societies award for outstanding contributions by a person under the age of 40.

Recent Articles

Chris Schmid on Evidence Based Medicine

Chris Schmid is a statistician at New England Medical Center who is an expert on evidence-based medicine. I invited him to present an introductory overview lecture on the topic at last year’s Joint Statistical Meetings, and here are his slides . All 123 of them. I don’t know how he expected to go though all of these in an hour. You could teach a semester-long course based on this material. Good stuff, I recommend you all read it. I’m posting this here because (a) medicine is important in its own right, and (b) the principles of evidence-based medicine also apply to public policy.

Jacobs and Shapiro on how politicians try to mold public opinion

In my NYT column on how to think about conflicting polls, I wrote: A vast majority of Americans — including half of all self-identified Republicans — think there is “too much power in the hands of a few rich people and large corporations.” And a solid majority believes that “the country’s economic system unfairly favors the wealthy.” On the other hand, close to 60 percent of Americans do not see the country as “divided into haves and have-nots” and over 60 percent see “big government” as the biggest threat to the country in the future. What gives? I think it’s fair to say that there’s enough here to support different political themes. A supporter of higher taxes for higher incomes can focus on the “too much power in the hands of the rich” angle, whereas a supporter of cuts in low-income and middle-income entitlement programs can focus on the lack of resonance of the haves and have-nots argument. The ambiguity revealed in these polls actually makes sense: if there were a clear and...

Probability Theory 101

Gregg Easterbrook : Gingrich is a wild card. He probably would end up a flaming wreckage in electoral terms, but there’s a chance he could become seen as the man unafraid to bring sweeping change to an ossified Washington, D.C. There’s perhaps a 90 percent likelihood Obama would wipe the floor with Gingrich, versus a 10 percent likelihood Gingrich would stage an historic upset. This is the dumbest thing I’ve seen since . . . ummm, I dunno, how bout this ? It actually gets worse because Easterbrook then invokes game theory. What next? Catastrophe theory? Intelligent design? P.S. Maybe I should explain for readers without an education in probability theory. Let’s suppose “wipe the floor” means that Obama gets 55%+ of the two-party vote, and let’s suppose that “an historic upset” means that Obama gets less than 50% of the vote. Now try to draw a forecast distribution that has 90% of its probability above 0.55 and 10% of it’s probability below 0.50. It’s a pretty weird-looking...

North Korea, East Germany, . . . California

Andrew Sullivan passes on this amusing line from James Pethokoukis: We’ve had some eye-opening natural economic experiments: North and South Korea. East and West Germany. California and Texas. Enough is enough. I don’t quite see the parallel here. I’ve heard that North Korea isn’t such a fun place and neither was East Germany. But I’ve been to California and it’s pretty nice. In all seriousness, what interests me about this quote is the way that two completely opposite arguments are being juxtaposed: 1. Life sucks in North Korea. Life sucked in East Germany. Communism sucks, and one piece of evidence for this is that life sucks in communist countries. 2. They have it too easy in California. California-ism sucks because Californians spend like grasshoppers and enjoy themselves too much. Argument 1 seems pretty strong to me. (Not perfect—-after all, communism isn’t really such a natural experiment—-but the point still seems pretty clear. ) Argument 2, though, . . . it’s possible , it...

The influence of strategic retirement on the incumbency advantage in US House elections

Ben HIghton writes : Failure to take into account ‘strategic retirement’ leads to inflated estimates of the incumbent electoral advantage. The one attempt to address this issue in the context of US House elections implies that much of the supposed incumbency advantage and most of its presumed increase over time are illusory (Cox and Katz, 2002). This paper identifies possible problems with the Cox and Katz (2002) method and develops a new approach based on simulating the counterfactual condition of incumbents standing for re-election rather than retiring. The results show that when the bias induced by strategic retirement is removed, much of the apparent incumbency advantage and its increase over time remain evident. This makes me happy because it is consistent with our claims here .