So...how come Romney lost...?

[_EAllusion]

Incidentally, in my personal life by far the bulk of asinine political views that suggest profound ignorance I am exposed to are of the liberal sort. Fortunately, I have enough sense to understand that my experience might not be representative of the nation as a whole and likely is warped by the fact that I live in Madison, WI. When I lived in Waukesha county*, which is Madison's political polar opposite, my experience was reversed.

*Waukesha is a fairly wealthy area of the state that is overwhelmingly Republican. It's a well-educated area of the state, so on a lark I looked it up and it has more college degrees per capita than the top state in the country. And it's as red as red gets. WI, meanwhile, is blue despite it, not because of it.

[_Milesius]

He lost due to low information voters like the gaggle of vapid justme clones who stood in line behind me while I waited an hour and a half or so to vote. (Although, I considered writing in Pawlenty.)

[_Analytics]

Does your claim that "the poorly educated in those states tend to vote Democratic" hold true if you control for race? I doubt it. In the South, blacks vote Democratic, and whites vote Republican. The big question is whether college-educated white southerners are more or less Republican that non-college-educated white southerners.

Why's that the big question? If you're trying to argue that there's a correlation between education level and political preference to imply some point about the relative intelligence of each political preference you can't exclude races to juice the stats in your favor.

The objective of this is to understand why people see things differently and vote differently. I'm not trying to argue that there is a simple correlation between education level and political preference--I'm trying to argue that education is a covariate in explaining the causation of why people vote the way they do. But to really understand causation, you can't just identify a set of correlations--you need to understand the relationship between them.

How do the less-educated whites in South Carolina vote compared to the more-educated whites in South Carolina? That is not only a valid and more insightful question regarding the effect education has on political preferences.

[_EAllusion]

The objective of this is to understand why people see things differently and vote differently. I'm not trying to argue that there is a simple correlation between education level and political preference--I'm trying to argue that education is a covariate in explaining the causation of why people vote the way they do. But to really understand causation, you can't just identify a set of correlations--you need to understand the relationship between them.

How do the less-educated whites in South Carolina vote compared to the more-educated whites in South Carolina? That is not only a valid and more insightful question regarding the effect education has on political preferences.

Understanding white voting behavior is a fun, informative pursuit, but it has nothing to do with the point Kevin was trying to make or my responses to it. So to interject that the "big question" is whether college educated white southerners are more or less Republican than non-college educated white southerners in the middle of that conversation is distracting and misguided. You might as well have interjected that the big question was whether Republicans do better or worse among Latinos who speak English as a first language compared to ESL and non-English-speaking Latinos. It's an interesting question that tells us something about voting behavior and education, but it's also far afield the point Kevin was making and I was disputing.

[_Analytics]

Understanding white voting behavior is a fun, informative pursuit, but it has nothing to do with the point Kevin was trying to make or my responses to it. So to interject that the "big question" is whether college educated white southerners are more or less Republican than non-college educated white southerners in the middle of that conversation is distracting and misguided. You might as well have interjected that the big question was whether Republicans do better or worse among Latinos who speak English as a first language compared to ESL and non-English-speaking Latinos. It's an interesting question that tells us something about voting behavior and education, but it's also far afield the point Kevin was making and I was disputing.

Just to clarify, you told Kevin,

You're trying to make some inane point about red states being dumb/uneducated, but those states are red despite their illiterates, not because of them.

I'm responding that this isn't necessarily true--the less educated aren't a homogenous group.

[_EAllusion]

I'm responding that this isn't necessarily true--the less educated aren't a homogenous group.

The less educated don't have to be a homogenous group for that statement to be true. Take the subgroup of literal illiterates. If 51% of them vote Democrat, no matter how demographically diverse a group that is, then the state can't be red because of its illiterate population. It can only be despite them. You might want to say, "Well yeah, but subgroup X that is illiterate is more Republican" but that's beside the point when the general illiteracy rate in red states is attempting to be leveraged to argue a general relationship between ignorance and voting Republican.

[_Jaybear]

If 51% of them vote Democrat, no matter how demographically diverse a group that is, then the state can't be red because of its illiterate population. It can only be despite them.

That is true, but it can still be red because of its illiterate white population.

I think its fair to say that Republicans have managed to alienate minorities regardless of their education levels. Accordingly if you are going to dispute the existence of causal relation between a lack of education and support for republicans you have to filter minorities out of the data pool.

I haven't seen data cross refencing race and education, but I doubt you can find a state, even a blue state, where less than 65% of white voters without any college education voted for Romney. Ignorant, uneducated, racist white voters were an important part of the Romney base.

[_Analytics]

I'm responding that this isn't necessarily true--the less educated aren't a homogenous group.

The less educated don't have to be a homogenous group for that statement to be true. Take the subgroup of literal illiterates. If 51% of them vote Democrat, no matter how demographically diverse a group that is, then the state can't be red because of its illiterate population. It can only be despite them. You might want to say, "Well yeah, but subgroup X that is illiterate is more Republican" but that's beside the point when the general illiteracy rate in red states is attempting to be leveraged to argue a general relationship between ignorance and voting Republican.

I hear what you are saying, but I find this approach at best simplistic and at worst misleading.

Let’s flesh this out with some hypothetical numbers I constructed to illustrate the point. Assume a red state that gave Romney a 20% margin of victory. In this state, the Obama won the less-educated by 2 points, 51% to 49%.

If you dig a little deeper, you find that 25% of the total population is less-educated blacks, and 18% of the total population is less-educated whites. 10% are more-educated blacks, and 47% are more educated whites.

Digging deeper, you find the following:
1- Less-educated blacks for Obama: 80%
2- More-educated blacks for Obama: 90%
3- Less-educated whites for Obama: 10%
4- More-educated whites for Obama: 20%

In this example, if you control for race, you see that there is absolutely a correlation between people with more education favoring Obama: more-educated blacks favor him more than less-educated blacks, and more-educated whites favor him more than less-educated whites.