This critique by Ted Goertzl, Myths of Murder and Multiple Regression, is making the rounds. It made me think of this old apocryphal story:
There is a famous anecdote inspired by Euler's arguments with secular philosophers over religion, which is set during Euler's second stint at the St. Petersburg academy. The French philosopher Denis Diderot was visiting Russia on Catherine the Great's invitation. However, the Empress was alarmed that the philosopher's arguments for atheism were influencing members of her court, and so Euler was asked to confront the Frenchman. Diderot was later informed that a learned mathematician had produced a proof of the existence of God: he agreed to view the proof as it was presented in court. Euler appeared, advanced toward Diderot, and in a tone of perfect conviction announced, "Sir, (a + bn)/z = x, hence God exists--reply!". Diderot, to whom (says the story) all mathematics was gibberish, stood dumbstruck as peals of laughter erupted from the court. Embarrassed, he asked to leave Russia, a request that was graciously granted by the Empress.
The second law of thermodynamics talking point is an unsophisticated descendant of Euler's gambit, while William Dembski's whole career is a homage to the trick. In regards to the shenanigans that Goertzl refers to, you can go to UC Berkeley's GSS interface, and after reading the documents, "massage" the outcomes yourself. I do think that econometric methods are still useful tools, their frequent misuse shouldn't lead us down the path to epistemological nihilism. Many use the law to further their own selfish ends and pervert its intent, but that doesn't mean that the general idea of an objective legal framework has to be discarded. Specifically, I think the Cowles Foundation has done some worthwhile work (e.g., showing that stock newsletters add no value).
- Log in to post comments
I was dumbfounded by the following passage:
Lott and Mustard were comparing trends in Idaho and West Virginia and Mississippi with trends in Washington, D.C. and New York City. What actually happened was that there was an explosion of crack-related homicides in major eastern cities in the 1980s and early 1990s. Lott's whole argument came down to a claim that the largely rural and western "shall issue" states were spared the crack-related homicide epidemic because of their "shall issue" laws.
I mean, seriously, what's wrong with trying to get a little intuitive understanding of your methods before you apply them?
a delicate flower you are sir. anyway, they understood the methods i presume, they just knew the answer they wanted and so if it can predict one must do the fit! such behavior knows no ideology. anyway, of course we come here to bury econometrics and not praise it, but at least these sorts of tricks are transparent. michael bellesiles simply lied and got away with it because it takes a lot more footwork to dig through purported archival records than to reanalyze data sets. R is free.
Frankly, this captures my whole suspicion about quantitative social sciences: it seems so easy to get the answer you want. As a natural scientist of delicate sensibility, if my results differ from another labs, I expect there to be differences that can be explained by something other than preference for a particular result. Depending on the lab.
Now that we have given these people access to fMRI machines, the bullshit flows like a river and science journalists have never been so gee-whiz golly dazzled.
Ha! West Virginia and Idaho are different than Washington DC and NYC because of their shall issue laws?!? That really made my day. How far in the sand can a head be sunken? Political correctness is the worst thing to ever happen to academia. Why indeed aren't rural white teens capping each other over dime bags? Such a mystery.
Hilarity, Lott's regressions control for such demographic factors. They are not deluded by political correctness into thinking those are non-factors. The problem is that there just isn't enough data on big cities with concealed carry to say much about the effect and know whether such statistical controls completely account for other factors properly.
I thought Powerpoint was the worst thing to ever happen to academia. That or Thompson/ISI.
One thing I'm wondering out of curiosity after reading many of your posts Razib...
I don't know much about peer review for social research but with lots of online data available from home, like the stuff you do with the GSS, do you think you could get publishable results into academic papers?
I mean some papers I've seen in soc. research are data mining and then an interpretation, without needing "fieldwork", actually designing the survey, setting up an behavioural experiment etc. anyways.
How hard d'ya reckon it is for someone without formal training in social sciences but who's an audodidact to do his own research because he has tools at his disposal that let's say someone interested in biochemistry but without a lab can't do?
that's how tom rees publishes. can't be that hard. a few of my posts were dismissed on a message board when someone said, "i'm a political science grad student, and any political science grad student could do this in a week." thing is, i rarely spend more that 2 hours on a data analysis posts, max :-)