Saturday, December 12, 2009

Your Climate Change Probability Score

From Andrew Gelman:
What probability do you assign to the following statement: increasing the atmospheric carbon dioxide concentration above 800 ppm will change the global average surface temperature by more than 2.5 degrees C (4.5 F)? This would imply a climate sensitivity somewhat below the extreme low end of what the Intergovernmental Panel on Climate Change says is credible.

I think I'm a 0.95 (or a 95-percent-er). What's your number?

Wednesday, December 9, 2009

Three Burning Questions

From Matt Yglesias:
1. Why are we spending a multiple of Afghanistan’s total GDP on fighting a war in the country?
2. Couldn’t more be done, for cheaper, with cash for bribes and development?
3. How is it that it doesn’t take the Taliban years to train competent soldiers?

More silly Poll numbers

So yesterday I made a little funny about how 10.2 sometimes being greater than 10.8. Just to be clear, I do know that 10.2 is actually less than 10.8, but it was a metaphor for a thing that we can jokingly say about statistics being similar to bikinis: What they reveal is interesting, but what they conceal is essential. In this case looking at the national average unemployment now versus in 1982-4 conceals some essential facts about these two recessions, namely that when we control for age and education, unemployment for each subgroup of the population is higher now than in 82-4.

But some things don't add up. Like this:
The Daily Show With Jon StewartMon - Thurs 11p / 10c
Gretchen Carlson Dumbs Down
Daily Show
Full Episodes
Political HumorHealth Care Crisis

Tuesday, December 8, 2009

Every dog has his day

Last week's economix econoquiz results.

Congrats to Jim Bang, Jeffrey, GDeFelice, Ken, Kevin McGrade, Richard Nerland, Pradeep Srivastava, Jennifer Shepardson, Matt Warner, Bob Glassberg, Hippo, Carolyn, Arjun Raguram and John Gardner, who all received perfect scores.

On tight grading distributions

High schools in Virginia were faced with the need to "raise standards," so they did 2 things: (1) they instituted "SOL's" - standardized tests that, as far as I can tell, mainly give teachers a teast to teach to; and (2) tightened grade distributions so that each letter grade is only five units wide. Well, they did the same thing with the GRE around 2001.
I always knew this was a stupid idea (in both cases) but I could never quite put my finger on why until today: As Craig puts it on his blog, it will "reduce the variance in the scores and, unfortunately, reduce the signal-noise ratio in the scores." If the test is written so that the D students are able to get an 80-85, how am I supposed to distinguish between a D student who was lucky and a B student who was just unlucky on one or two questions? I'm all in favor of keeping a wider variance and letting the distribution fall where it may.

When is 10.2 > 10.7?

Apparently during the current recession. Let me explain. According to current statistics, the average unemployment rate was 10.2% in October (now 10.0). In 1983, the unemployment rate peaked at about 10.8%. There have been a lot of bogus comparisons of the current recession to past ones designed to support the claim that the current recession is somehow "worse" than previous ones by looking at the change in jobs instead of the total number of jobs.
But, it turns out that there might be a good reason to believe that these folks are correct, even if their metric is bad. If we disaggregate by education (or age), we find that the unemployment rate now is higher at every level of education than it was at the peak of the 1982-4 recession. How? Simpson's paradox. Demographics have shifted towards an older, more educated work force. Older, more educated workers have historically enjoyed lower rates of unemployment.

Monday, December 7, 2009