This came up on another thread, but I liked results so much I want to blog it separately.

Here's a plot of state Smart Scores against the percentage of people in each state who voted for Bush in 2004. On average, the smarter the state, the less likely they voted for George, and a one point increase in smarts reduced the percentage voting for Bush by about 0.3%.

Dumber States Voted for Bush 

Of course, one cannot conclude from the graph that only dummies voted for Bush. We are merely observing that on average, states with higher Smart Scores were less likely to vote for Bush than states with lower Smart Scores. The average Smart Score of states where the majority voted for Bush was -3.0. The average Smart Score for the remaining states was +3.1.

One caveat. This analysis only controls for differences in Smart Scores, which account for about 10% of the variation in the percentage voting for Bush. Other key factors that may influence voting patterns are not accounted for. For instance, states receiving a net inflow of federal dollars (they receive more dollars then they pay in taxes) were also more likely to vote for Bush. Including other variables might change the observed relationship between Smart Scores and voting patterns...but I doubt it would reverse.

Here's another more powerful relationship: The percentage with advanced degrees and the percentage voting for Bush.  The next plot shows that for every one percent increase in advanced degrees, the percentage in that state voting for Bush decreased 2.5%.  The prevalence of advanced degrees (or the underlieing factors that determine that percentage) explains over 40% of the state-by-state variation in voting patterns.

Less Educated States Voted for Bush

 Perhaps finally, perhaps not. Here's a plot of divorce rates against the percentage voting for Bush in 2004. A one unit increase in the number of divorces per 1,000 marriages is associated with a 3.8% increase in the percentage voting for Bush.

"Family Values" States have higher divorce rates and voted for Bush

Check out this one. People were more likely to vote for Bush in states that received more in federal expenditures than they paid in federal taxes.

States that take more from the federal government than they put in...voted for Bush

I just can't stop. Some mentioned that they noticed the price of gas depended on if you were in a red state or a blue state. Here's the plot of that relationship. Yes, on average, states that went for Bush have lower gas prices, but the association is pretty weak.

You even pay a little less for gas if you voted for Bush