February 9, 2010
Category:
Tyler Cowen quotes Barbara Demick as writing, "North Koreans have multiple words for prison in much the same way that the Inuit do for snow." So do we, no? But in our case, they seem to come from 1930s B-movies
I wonder if there are almost as many words for prison in Russia, Turkmenistan, and the other leaders on the list. Apparently North Korea is off the charts, so perhaps they have ten times as many words for prison/jail as we do.
P.S. America includes a bunch of Inuits, so I guess we have multiple words for snow also!
Posted by Andrew Gelman at 9:07 AM • 0 Comments • 0 TrackBacks
February 8, 2010
Category:
Ouch.
Here's the story (which Kaiser forwarded to me). The English medical journal The Lancet (according to its publisher, "the world's leading independent general medical journal") published an article in 1998 in support of the much-derided fringe theory that MMR vaccination causes autism. From the BBC report:
The Lancet said it now accepted claims made by the researchers were "false".
It comes after Dr Andrew Wakefield, the lead researcher in the 1998 paper, was ruled last week to have broken research rules by the General Medical Council. . . . Dr Wakefield was in the pay of solicitors who were acting for parents who believed their children had been harmed by MMR. . . .
[The Lancet is now] accepting the research was fundamentally flawed because of a lack of ethical approval and the way the children's illnesses were presented.
The statement added: "We fully retract this paper from the published record." Last week, the GMC ruled that Dr Wakefield had shown a "callous disregard" for children and acted "dishonestly" while he carried out his research. It will decide later whether to strike him off the medical register.
The regulator only looked at how he acted during the research, not whether the findings were right or wrong - although they have been widely discredited by medical experts across the world in the years since publication.
They also write:
The publication caused vaccination rates to plummet, resulting in a rise in measles.
An interesting question, no? What's the causal effect of a single published article?
P.S. I love it how they refer to the vaccine as a "three-in-one jab." So English! They would never call it a "jab" in America. So much more evocative than "shot," in my opinion.
Posted by Andrew Gelman at 9:06 AM • 5 Comments • 0 TrackBacks
Category:
Eric Bettinger, Bridget Terry Long, Philip Oreopoulos, and Lisa Sanbonmatsu write:
Growing concerns about low awareness and take-up rates for government support programs like college financial aid have spurred calls to simplify the application process and enhance visibility.
Here's the study:
H&R Block tax professionals helped low- to moderate-income families complete the FAFSA, the federal application for financial aid. Families were then given an estimate of their eligibility for government aid as well as information about local postsecondary options. A second randomly-chosen group of individuals received only personalized aid eligibility information but did not receive help completing the FAFSA.
And the results:
Comparing the outcomes of participants in the treatment groups to a control group . . . individuals who received assistance with the FAFSA and information about aid were substantially more likely to submit the aid application, enroll in college the following fall, and receive more financial aid. . . . However, only providing aid eligibility information without also giving assistance with the form had no significant effect on FAFSA submission rates.
The treatment raised the proportion of applicants in this group who attended college from 27% (or, as they quaintly put it, "26.8%") to 35%. Pretty impressive. Overall, it appears to be a clean study. And they estimate interactions (that is, varying treatment effects), which is always, always, always a good idea.
Here are my recommendations for improving the article (and this, I hope, increasing the influence of this study):
Read on »
Posted by Andrew Gelman at 9:05 AM • 0 Comments • 0 TrackBacks
January 27, 2010
Category:
Following up on our recent discussion (see also here) about estimates of war deaths, Megan Price pointed me to this report, where she, Anita Gohdes, Megan Price, and Patrick Ball write:
Several media organizations including Reuters, Foreign Policy and New Scientist covered the January 21 release of the 2009 Human Security Report (HSR) entitled, "The Shrinking Cost of War." The main thesis of the HRS authors, Andrew Mack et al, is that "nationwide mortality rates actually fall during most wars" and that "today's wars rarely kill enough people to reverse the decline in peacetime mortality that has been underway in the developing world for more than 30 years." . . . We are deeply skeptical of the methods and data that the authors use to conclude that conflict-related deaths are decreasing. We are equally concerned about the implications of the authors' conclusions and recommendations with respect to the current academic discussion on how to count deaths in conflict situations. . . .
The central evidence that the authors provide for "The Shrinking Cost of War" is delivered as a series of graphs. There are two problems with the authors' reasoning.
Read on »
Posted by Andrew Gelman at 3:45 AM • 0 Comments • 0 TrackBacks
January 26, 2010
Category:
You've all heard about how you can predict all sorts of things, from movie grosses to flu trends, using search results. I earlier blogged about the research of Yahoo's Sharad Goel, Jake Hofman, Sebastien Lahaie, David Pennock, and Duncan Watts in this area. Since then, they've written a research article.
Here's a picture:
And here's their story:
We [Goel et al.] investigate the degree to which search behavior predicts the commercial success of cultural products, namely movies, video games, and songs. In contrast with previous work that has focused on realtime reporting of current trends, we emphasize that here our objective is to predict future activity, typically days to weeks in advance. Specifically, we use query volume to forecast opening weekend box-office revenue for feature films, first month sales of video games, and the rank of songs on the Billboard Hot 100 chart. In all cases that we consider, we find that search counts are indicative of future outcomes, but when compared with baseline models trained on publicly available data, the performance boost associated with search counts is generally modest--a pattern that, as we show, also applies to previous work on tracking flu trends.
The punchline:
We [Goel et al.] conclude that in the absence of other data sources, or where small improvements in predictive performance are material, search queries may provide a useful guide to the near future.
I like how they put this. My first reaction upon seeing the paper (having flipped through the graphs and not read the abstract in detail) was that it was somewhat of a debunking exercise: Search volume has been hyped as the greatest thing since sliced bread, but really it's no big whoop, it adds almost no information beyond a simple forecast. But then my thought was that, no, this is a big whoop, because, in an automatic computing environment, it could be a lot easier to gather/analyze search volume than to build those baseline models.
Sharad's paper is cool. My only suggestion is that, in addition to fitting the separate models and comparing, they do the comparison on a case-by-case basis. That is, what percentage of the individual cases are predicted better by model 1, model 2, or model 3, and what is the distribution of the difference in performance. I think they're losing something by only doing the comparisons in aggregate.
It also might be good if they could set up some sort of dynamic tracker that could perform the analysis in this paper automatically, for thousands of outcomes. Then in a year or so they'd have tons and tons of data. That would take this from an interesting project to something really cool.
Posted by Andrew Gelman at 10:49 AM • 0 Comments • 0 TrackBacks
January 23, 2010
Category:
First the scientific story, then the journalist, then my thoughts.
Read on »
Posted by Andrew Gelman at 3:44 PM • 1 Comments • 0 TrackBacks
Category:
Alan Turing is said to have invented a game that combines chess and middle-distance running. It goes like this: You make your move, then you run around the house, and the other player has to make his or her move before you return to your seat. I've never played the game but it sounds like fun. I've always thought, though, that the chess part has got to be much more important than the running part: the difference in time between a sprint and a slow jog is small enough that I'd think it would always make sense just to do the jog and save one's energy for the chess game.
But when I was speaking last week at the University of London, Turing's chess/running game came up somehow in conversation, and somebody made a point which I'd never thought of before, that I think completely destroys the game. I'd always assumed that it makes sense to run as fast as possible, but what if you want the time to think about a move? Then you can just run halfway around the house and sit for as long as you want.
It goes like this. You're in a tough spot and want some time to think. So you make a move where the opponent's move is pretty much obvious, then you go outside and sit on the stoop for an hour or two to ponder. Your opponent makes the obvious move and then has to sit and wait for you to come back in. Sure, he or she can plan ahead, but with less effectiveness than you because of not knowing what you're going to do when you come back in.
So . . . I don't know if anyone has actually played Turing's running chess game, but I think it would need another rule or two to really work.
Posted by Andrew Gelman at 2:47 PM • 3 Comments • 0 TrackBacks
January 22, 2010
Category:
Stephen Dubner reports on an observational study of bike helmet laws, a study by Christopher. Carpenter and Mark Stehr that compares bicycling and accident rates among children among states that did and did not have helmet laws. In reading the data analysis, I'm reminded of the many discussions Bob Erikson and I have had about the importance, when fitting time-series cross-sectional models, of figuring out where your identification is coming from (this is an issue that's come up several times on this blog)--but I have no particular reason to doubt the estimates, which seem plausible enough. The analysis is clear enough, so I guess it would be easy enough to get the data, fit a hierarchical model, and, most importantly, make some graphs of what's happening before and after the laws, to see what's going on in the data.
Beyond this, I had one more comment, which is that I'm surprised that Dubner found it surprising that helmet laws seem to lead to a decrease in actual bike riding. My impression is that when helmet laws are proposed, this always comes up: the concern that if people are required to wear helmets, they'll just bike less. Hats off to Carpenter and Stehr for estimating this effect in this clever way, but it's certainly an idea that's been discussed before. In this context, I think it wouldb useful to think in terms of sociology-style models of default behaviors as well as economics-style models of incentives.
Posted by Andrew Gelman at 3:43 PM • 5 Comments • 0 TrackBacks
Category:
I posted a note the other day about the difference between internal and external coherence of political ideology. The basic idea is that, a particular person or small group can have an ideology (supporting positions A, B, C, and D, for example) that is perfectly internally coherent--that is, all these positions make sense given the underlying ideology--while being incoherent with other ideologies (for example, those people who support positions A, B, not-C, and not-D). What's striking to me is how strongly people can feel that their beliefs on a particular issue flow from their being a liberal, or a conservative, or whatever, even though others with similar opinions will completely disagree with them on that issue.
Read on »
Posted by Andrew Gelman at 3:42 PM • 0 Comments • 0 TrackBacks
