Do we know what % of voters switched party support in each election?

[Mr. Morden]

Is there any way besides exit polls alone to estimate how many people voted for one party's presidential nominee in one election, and then the opposing party's nominee in the following election?  I realize that the exit polls purportedly tell us who people voted for in the previous election, but people often lie both about whether they voted last time and who they voted for.  So I was wondering if anyone has ever tried to use turnout data and precinct level results, or something like that, to give some supporting evidence re: this question?

Reason I'm wondering is because I'll be curious to see if we get a larger than normal number of crossover votes this time because Trump is such an unusual candidate, and I'm wondering if the only thing we'll be able to use for evidence is the exit poll question.  Maybe some combination of the "Who did you vote for last time?" question and other demographic info in the exit poll might shed some light on it?  I mean, if there's a huge swing in the educational subsamples, then presumably some of that is because of people switching from one party to another rather than because of a different group of people showing up at the polls.

Of course, if we're getting more than 5% of voters going 3rd party this time, that skews it further.  There could be plenty of people who don't switch from one major party to the other but instead switch from one of the major parties to 3rd party.

[Ebowed]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

[bagelman]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

[Ebowed]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

Presumably more fiscally conservative voters who saw that Obama was actually very moderate and didn't appreciate the direction that the Tea Party was taking the GOP.

In 2008, it was still seen as acceptable for well-educated people to support the Republicans.  A great deal of damage was done to the party's credibility with the rise of birtherism and the rhetoric that the GOP used in the lead up to 2012.

[RINO Tom]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

Presumably more fiscally conservative voters who saw that Obama was actually very moderate and didn't appreciate the direction that the Tea Party was taking the GOP.

In 2008, it was still seen as acceptable for well-educated people to support the Republicans.  A great deal of damage was done to the party's credibility with the rise of birtherism and the rhetoric that the GOP used in the lead up to 2012.

You guys really do have an amusing view of your coalition.  Educated people voted GOP in 2014, did they not?

[FEMA Camp Administrator]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

Presumably more fiscally conservative voters who saw that Obama was actually very moderate and didn't appreciate the direction that the Tea Party was taking the GOP.

In 2008, it was still seen as acceptable for well-educated people to support the Republicans.  A great deal of damage was done to the party's credibility with the rise of birtherism and the rhetoric that the GOP used in the lead up to 2012.

I'd say the opposite. I recall someone doing a story on an Ohio voter, a veteran, who felt McCain was more compassionate than Romney and had voted GOP for years. Recall Romney's persona through the campaign.

[Mr. Morden]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Well, in most years they ask on the exit poll "Who did you vote for in the last election?", so supposedly we know the answer.  But as I said, not everyone tells the truth about whether they voted last time and who they voted for.

However, I just looked up the 2012 exit poll, and didn't see the "Who did you vote for last time?" question in there at all.  Is it there, but I missed it?  Because it's something they've asked in every other GE exit poll since at least the 90s, so it would be weird if they dropped it in 2012.

[hopper]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

In 2008, it was still seen as acceptable for well-educated people to support the Republicans.  A great deal of damage was done to the party's credibility with the rise of birtherism and the rhetoric that the GOP used in the lead up to 2012.

Romney won College Grad(but not Post Grad) in 2012 51-47%. Birtherism didn't cost the GOP votes in 2012.

[hopper]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

Presumably more fiscally conservative voters who saw that Obama was actually very moderate and didn't appreciate the direction that the Tea Party was taking the GOP.

In 2008, it was still seen as acceptable for well-educated people to support the Republicans.  A great deal of damage was done to the party's credibility with the rise of birtherism and the rhetoric that the GOP used in the lead up to 2012.

I'd say the opposite. I recall someone doing a story on an Ohio voter, a veteran, who felt McCain was more compassionate than Romney and had voted GOP for years. Recall Romney's persona through the campaign.

True Romney talked like a businessman in the 2012 campaign because that was his main background was in the business sector. McCain was in the Navy and was a Captain in the Navy as well as being a POW in Vietnam. In conclusion, Romney and McCain had 2 different life or work backgrounds if you will. McCain probably sounded like he had more compassion than Romney because of his experience of being a POW and being in the Navy.

[Mr. Morden]

I looked at yesterday's exit poll, and didn't even see a question about who people voted for in 2012.  Is it in there, but I missed it?  Very disappointing if they didn't ask this, as I was quite curious to have some indication of how many Obama and Romney voters swapped parties this time.

[SingingAnalyst]

I looked at yesterday's exit poll, and didn't even see a question about who people voted for in 2012.  Is it in there, but I missed it?  Very disappointing if they didn't ask this, as I was quite curious to have some indication of how many Obama and Romney voters swapped parties this time.

Me too, I think that would have been nice.

[All Along The Watchtower]

Take this with a grain of salt, since I don't remember which poll measured this, but I recall reading in 2012 that for every two Obama 08 voters that he lost, Obama gained one McCain voter.

Who would those have been? They couldn't have all been PUMA's.

Presumably more fiscally conservative voters who saw that Obama was actually very moderate and didn't appreciate the direction that the Tea Party was taking the GOP.

In 2008, it was still seen as acceptable for well-educated people to support the Republicans.  A great deal of damage was done to the party's credibility with the rise of birtherism and the rhetoric that the GOP used in the lead up to 2012.

You guys really do have an amusing view of your coalition.  Educated people voted GOP in 2014, did they not?

Postgrads voted for Dems 53-44, according to this: http://www.cnn.com/election/2014/results/race/house/#exit-polls

Republicans won (non-postgrad) college grads 54-44, some college voters 54-44, and high school grads 53-44,  but Dems won voters without a high school diploma 54-44 (a category which I strongly suspect is disproportionately non-white).

But remember that while the Dems continue to win the lowest-income voters - regardless of education - by substantial margins, their performance among lower and middle income white voters (broadly speaking) has had a considerable decline in recent decades, especially in the last few presidential and congressional elections.  Meanwhile,  Democrats' advantage among postgrads has only grown (though obviously much of this can be attributed to the rapid growth of women and minorities as a share of this demographic).

The income and education correlations with voting and ideology (both of which have also become more correlated in recent years) can be summed up with the observation that high-income voters with only (or not even!) a high school education or "some college" are the most conservative/Republican, while low-income postgrads have become one of the most reliably liberal demographics (and thus, reliably Democratic - though this group also does very well with candidates like Ralph Nader in 2000 and this year, Bernie Sanders in the Democratic primaries). Furthermore, less-educated (white) middle and upper income voters have moved far to the Right in their voting habits, while conversely, more-educated lower and middle income voters have moved far to the Left.

All of that being said, the Democrats continue to draw much of their strength among lower and middle income voters (particularly minorities) while the Republicans continue to draw much of their strength among (white) middle and upper income voters. The fact that lower income voters are considerably less reliable in elections (particularly in the midterms) than middle and especially upper income voters  - when combined with the fact that the segment of college-educated/postgrad voters who are the most Dem are highly concentrated in the major metro areas of safely Democratic states - goes a long way toward explaining the Democrats' geographic and structural problems these days.