This summer, the DailyKos/SEIU/PPP polls started including a question about educational attainment. So how does education relate to political views? With a whole summer of data at hand (10,000 respondents, 6/14/12-8/23/12 - the September polls are not included, as they are Likely Voter polls, not Registered Voter polls), we can aggregate the data and narrow down the variables to try to get a handle on this question.
First we'll remove race as a variable, and only look at respondents who are white (I would have also looked at data for racial minorities but these data are compromised). Second, only respondents older than 30 will be included (they've more than likely completed their education). All demographic subgroups shown in this post will have N>100 and an incorrect geography response rate of <10%.
There is a strong and obvious relationship between educational attainment and voter preference. Maybe that's why Republicans hate education so much. (And may I just insert a gentle reminder here that educational attainment is
not the same thing as intelligence.)
But perhaps the data are skewed by unequal representation in additional demographic categories? What about income, gender, and region? More below...
What about income? When people earn more money, they start to hate taxes and start voting Republican, right?
Wrong.
Once you've taken race and education out of the equation, there is little if any relationship between income and voting preferences. Respondents from the highest income levels tend to be a little less likely to support Obama, but the error bars are generally overlapping each other. In any case, it is clear that any relationship with income is far outweighed by the relationship to educational attainment. However, it is the higher educational categories that are more heavily weighted towards higher income - so if even if we did see a clear trend with income, and accounted for that in the education graph, we would see an
even stronger relationship between education and political preferences.
What about gender? More women are graduating from college, so wouldn't that skew the higher education levels towards the preferred candidate of women, Obama?
It turns out that the trend with education is evident for both men and women. Not only that, but for respondents over 30, men still are the majority of the postgraduate (55%) and college (52%) demographics. So
again, if we account for the gender mix at each education level, this would actually slightly
enhance the relationship between education and voter preference, not diminish it.
Finally, what about geography? Here's the education graph split by geography (as defined by area codes dialed in the polling).
Again, we see the same pattern for all regions, but the level of support is lower for the South. But as it turns out, there is very little difference in the distribution of educational attainment among whites between the South and the nation as a whole: the percent of white voters who reside in the South is 32% of high school graduates, 33% of some college, 32% of college graduates, and 30% of postgraduates. Although this small discrepancy in distribution does work against the trend shown in the first graph, it is very slight.
Basically, though, we see the same pattern over and over, even when split by gender, income, or region: more educated white voters are more Democratic voters. There's three possible mechanisms I can think of here:
1. More educated white voters are more Democratic because Republicans keep bashing education. For evidence, we see that since 1980, college educated voters have gradually shifted their presidential preferences from Reagan +17 to Obama +8 (although this includes an increase in non-white voters as well).
2. Education itself at most institutions tends to broadens an individual's experiences and teaches that life is not full of the black-and-white certainties associated with the authoritarian Republican worldview. This is the most basic correlation-equals-causation interpretation of the data, and one conservatives have been blathering on about since William F. Buckley, Jr.
3. White individuals who are likely to be Democrats or Republicans self-select by choosing to continue their education or not as the case may be. We actually have some evidence against this idea in a survey from 2008 showing that 18-24 year olds in college had nearly identical political preferences as those who were not in college. The demographics of both groups were fairly similar (college:64% white, 53% female; non-college: 57% white, 50% female). However, this may not have been true in the past.
While you contemplate these possibilities, I will leave you with the opinions of one of the great thinkers on the other side of the aisle, who proves single-handedly that you can get all the way through law school without being the least bit intelligent.
“President Obama said he wants everybody in America to go to college. What a snob! There are good, decent men and women who go out and work hard every day and put their skills to test that aren’t taught by some liberal college professor trying to indoctrinate them. Oh I understand why he wants you to go to college. He wants to remake you in his image. I want to create jobs so people can remake their children into their image, not his.” - Rick Santorum, 2/25/12