Something Smells Funny

A lot has been made on the blogosphere and Twitterverse recently about a Research 2000 poll commissioned by Big Orange that polls registered, self-identified Republicans and finds that they're, well, batshit insane. Some of the poll highlights:

39% believe Obama should be impeached; 29% are not sure.

Only 42% believe Obama was born in the U.S.

21% believe ACORN stole the 2008 election; a whopping 55% are "not sure."

Only 8% - that's eight percent - believe that gay and lesbian people should be allowed to teach in the public schools. 19% were not sure; 73% said no.

That's some serious crazy... but color me skeptical. Let's take a look at the "don't ask, don't tell" question, since that's in the news nowadays. The R2K poll has Republican support for allowing gays and lesbians to serve openly in the military at a mere 26%, with 19% not sure and 55% opposed. Problem is, there's a Gallup poll from June that asks the exact same question. The result? Support for allowing openly gay soldiers is at 58% - 32 points higher than in the R2K poll. Even if we account for the fact that Gallup pushes fence-sitters, and we place all those fence-sitters in with the open-service supporters, we get 45% support - 13 points less than Gallup's number, still nothing to sneeze at. What's more, the questions asked were semantically equivalent. Something, somewhere, is horribly wrong.

There are a couple of options. One is that support for gays in the military has tanked between June and now. I find that highly unlikely - the Gallup poll had a trend among Republicans of +6 over the past four years, and even during the recent debate on the subject top military brass - who carry a lot of weight with Republicans - have been almost unanimous in supporting repeal of the DADT policy.

Another is that R2K's poll oversampled cranky old people, which it did - 37% of R2K's sample were over 65. R2K gave us crosstabs by age group, though, and the 18-29-year-olds sampled by R2K gave their support at 31% with 22% fence-sitting - that's 53% maximum support for open service. That 53% number is still below Gallup's number for all Republicans, and 18-29s are among the most likely groups to support allowing LGBT folk to serve openly.

One intriguing option, though, is that questions asked previously by R2K could have triggered a more conservative frame of mind in many respondents. R2K asked their question after they asked several other questions about radical conservative beliefs, including the ACORN question, whether Obama is a socialist or a U.S. citizen, etc. After being asked all these questions, it's easy to imagine that someone would start automatically giving the "conservative" position whether or not they actually agree with that opinion independently. They're not thinking "what do I think" anymore - on a subconscious level, they're thinking, "I'm a conservative, what do conservatives think?" It's a cognitive shortcut that's easy to trigger, and it might explain the discrepancy between the two polls.