This is interesting. West Texas is above average for racism. Is anyone surprised?
In their study, the authors looked for DMAs (designated market areas, well-accepted media market geographical areas) that exhibited TWO independent measures of racism: (1) Google searches using the N-word, a metric that had already been shown to correlate with the difference between 2008 votes for Barack Obama and 2004 votes for John Kerry, and (2) measures of high black mortality and measures of greater differential between black and white mortality (valid due to the well-known correlation of higher mortality rate of humans from heart disease, stroke, cancer, etc. who live under frequent or constant stress, in this case the stress due to racism). As the authors note, the results are statistically significant.
The regions of greatest racism were exactly what you would expect–the rural South and the Appalachians. But there were also surprising outliers of moderately higher racism, one of which is West Texas. Take a look at the maps.
I don’t know if lots of the behavior described here is really racism, or if it’s just situation specific behavior. You might cross the street if you are walking in Baltimore at night and a black guy approaches. However, you probably don’t have any problem whatsoever with black coworkers or friends.
While I’m sure that the math is correct, I’m not sure that I agree with any of the above.
If I’m reading the regressions correctly, the R-squared has a P-value of
The author claims there’s a correlation between racist google searches and presidential elections. First, that sounds like a stretch, second, he offers no evidence of it, and third, we have no idea who is doing the Google search. It could be blacks googling the word for all we know. (AFAIK, blacks use the world “nigger” or its derivatives far more than whites do.)
The authors state that high black mortality is evidence of racism. Is it possible (however unlikely) that this could be due to strictly biological factors? Or that this stress is due to other factors, other than racism? Is it possible that blacks killing other blacks is a cause of frequent stress? Or racism the only possible cause of black stress?
Actually, my locker at the YMCA is right between the Federal Judge (nominated by W) and the chairman of the Midland County Democrat association. Makes for some interesting conversations in the locker room.
Meh. They seem to have adjusted the data ad infinitum to arrive at a predestined conclusion.
It is a very large stretch to support the conclusion that “As a source of psychosocial stress, racial discrimination may lead to premature physiologic deterioration or “wear and tear,” ultimately compromising the ability of the body to respond to such challenges and increasing susceptibility to and acceleration of chronic diseases.”
I agree with your points #2 and #3. For all they know many non-racist people were searching lyrics to popular rap songs.
This was fairly easily debunked. Great example of correlation =/= causation. Basically, the places where the n-word was used most frequently are also places where rap music is most popular. People be searching for lyrics yo.
^Actually, the guy who sent the e-mail has a PhD in Geology and his wife has a PhD in Biology. So they’re certainly smart and educated. Whether they understand statistics, I do not know.
I would assume a PhD in Geology has a solid grasp on multiple regression, but I don’t know. And he’s probably not completely unbiased, either.