If this poll is to be believed, it seems like LGBT folks are among the happiest with Obama. 67% of LGBT adults think the country is headed in the right direction; only 45% of straight adults agree.
That all said, LGBT people are having an easier time with their finances, or are at least more optimistic about their financial futures:
When it comes to their outlook on the economy and their own personal financial situation, GLBT adults also appear somewhat more confident then their heterosexual counterparts. A majority (57%) of GLBT adults said they expect the economy to improve in the coming year, compared to 45 percent of heterosexual adults. In the next 6 months just looking at gay and lesbian adults, 31 percent expect their household's financial condition to be better, compared with 24 percent of heterosexual adults. The survey also found that more than two-thirds (69%) of GLBT adults would rate the overall job President Barack Obama is doing in handling the economy as excellent or pretty good, while only 38 percent of heterosexual adults agree.
That's a pretty big difference on that last question, and I'm kinda wondering where that's coming from. If I were asked, I'd have been one of the 31% who thinks he's doing a less-than-good job on the economy, but that might just be because I'm not hardwired to experience "optimism" or "hope."
What do you all think about this poll? What would you all rate Obama's performance on the economy?
Of course, I don't know how much credibility to give this poll, since it's an online poll, and that means that the participants are, in some ways, self-selecting. While the gurus over at fivethirtyeight.com don't consider it at all because it's conducted online, they also don't trash it as much as the other big online poll, Zogby Interactive. That might be because Harris does a better job of trying to make up for the fact that its participants are self-selecting: people get paid points to take polls so they aren't entirely dependent on people's interest in a topic to get people to participate. Plus they do adjust their scores to make it representative along the lines of race and gender.
That said, bias from appearing online should, theoretically, be eliminated if the point is to compare LGBT people found online to straight/cis people found online. And, unlike Zogby, Harris Interactive doesn't make it easy to email your friends about the poll, so that should cut down on people just getting all their friends and family to freep an online poll.
Here's what Nate Silver had to say about online polls, but more specifically about Zogby Interactive:
Telephone polls have their problems, particularly if they do not include respondents with cellphones, but they are a long ways ahead of Internet-based polling. Zogby Interactive, the most prolific (if the least methodologically sound) Internet-based pollster, has missed the outcome of recent elections by an average of 7.6 points when conducting polls online. (Internet-based polling is cheaper to conduct, but as is the case with fine dining in Manhattan, "value" should not be confused for "cost". Any organization commissioning an Internet-based poll is probably wasting its money, because the poll isn't likely to be any good.)
That said, I don't really see another way to conduct a poll like this. I'd imagine that asking people if they're gay or trans over the phone will result in bias as well, considering the fact that there are a lot of people who would lie in answer to that question either instinctively or because someone else is in the room.
Also, it focused almost entirely on economic issues - it would have been nice to get a straight-up approval rating from the entire sample (i.e., LGBT and straight/cis) to see how this compares to other polls conducted this past month.