Now we are hearing hyperbolic headlines about the doubling of the gay GOP vote and other such nonsense. First lets look at what the numbers actually say.
In 2006, gays made up 3% of the US House electorate and split 75 to 24 in favor of Democrats vs a 52 to 46 split amoung non gay voters. In 2008, we made up 4% of the Presidential electorate and split 70 to 27 in favor of Obama vs a 53 to 45 split amoung non gay voters. In 2008, we made up 3% of the House electorate and split 80 to 19 in favor of Democrats vs a 53 to 44 split amoung non gay voters. In 2010, we made up 3% of the House electorate and voted 68 to 31 in favor of Democrats vs a 46 to 52 split amoung non gay voters.
Now, first let's look at differential which helps explain some of the change. We were plus 23 Democratic in 06, plus 27 in 08, and plus 22 in 10. Yep, that's right, we were a grand total of one point off our differential from 06 to 10. So the entire switch away from Democrats, save one single solitary point, is explained by the swing of the electorate. So maybe the increase in GOP percentage is out of whack. Well we were -22 in 06, -25 in 08, and -21 in 10. Yet again, all but one point of this supposed massive movement was echoing the movement of the electorate.
Now another thing about these polls. We were 3% of all of the House electorates and 4% of the Presidential electorate. So supposedly there was a drop off of 1 in 4 gay voters in 08 who voted for President but not for US House. But no drop at all from 08 to 10 in House votes by gays. That is a bit hard to believe. So what happened here? Did 1 out of every 4 gay people who voted in the Presidential race skip their house race? No what more likely happened is rounding. The percentage of the electorate was rounded to the nearest whole number, meaning that the 3% of the House electorate could have been anywhere from 3.49% down to 2.51%. Given the fact gays made up 4% of the Presidential electorate in 08, I think in 08 we likely made up closer to the 3.49% of the House electorate in 08. I also think, like most other groups, we are more likely to have voted in 08 than in either 06 or 10. Thus I think there was a drop off in gay voting that got rounded way. that is what happens when we try to use exit polls to analyse the voting behavior of such a small slice of the electorate. Rounding error, which can be up to a whole point, is between a third and a fourth of the size of the group we are examining.
Thus it becomes very hard to state what, if anything, these numbers show. Did a massive percent of gays stay home, thus inflating our standard number of GOP voters into a larger percentage? That seems to be the case. Our shift nearly coincides with that of the electorate as a whole and rounding error would have masked any drop off. In any case, we surely didn't double GOP support as one lying headline said.
One final comparision. Union households are loyal Democrats so let's compare the gay vote to union households.
In 06, Union households voted 64 to 30 in favor of Dems vs 49 to 49 for non union households. They were 21% of the vote. In 08, they were 64 to 30 again vs a 50 to 47 split for non unions still 21% of the vote. In 2010 they were down to 17% of the vote they voted 60 to 38 in favor of Dems vs 43 to 55 split in non union households. Differentials +15, +14, and +17 in favor of Dems. But close to one in five union households sat out the election. Was what appeared to be a tilt toward the GOP by gay voters instead gay voters staying home? Thanks to rounding error, we won't know. It should be noted that unions went from 30 to 38 percent in GOP support, the equivalent for gays was going from 24 to 31. Unions went up 8 and gays went up 7. Since gays had a lower base of GOP voters to begin with the percent increase is bigger, but our raw numbers were pretty close to the same which is why the differentials are so close.
In short, and I know I really wasn't short here, we performed about the same as unions did and most of our supposed huge shift can be explained by the overall shift of the electorate. Hardly the news the breathless headlines suggest.
Source for data
06
http://www.cnn.com/ELECTION/2006/pages/results/states/US/H/00/epolls.0.html08
http://www.cnn.com/ELECTION/2008/results/polls/#val=USH00p310
http://www.cnn.com/ELECTION/2010/results/polls/#val=USH00p2
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
