Friday, June 01, 2012

The futility of surveys

On the Political Gabfest they cited a phone survey conducted last year where Americans were asked whether they would vote for a theoretical presidential candidate who had a specific non white heterosexual Christian male aspect. If the person were qualified but were African American, or Latino, or gay, or atheist, etc., would that prevent voting for that candidate. And the results apparently were such that only 94% responded they would vote for a qualified black candidate—that there were 6% who would admit in a non-anonymous survey to another person on the phone that no matter what they'd never vote for someone on that racial basis. As an aside, the host pondered aloud who, in 2012, says that openly?

It's not that we don't all grasp that there's still plenty of people who have bigoted views—heck, in each of us there's probably at least one opinion that could be considered bigoted (perhaps not racial but some judgment against a group that dismisses that group in an unfair way), but it seems like even hardcore racists grasp that it's better not to seem racist in public. But before it starts to appear that I'm touting racists who are open about that (and the awkward ethical ground that puts us in) let's get back to the larger point: When it comes to surveys, conceivably decades of political correctness has essentially trained people what they're supposed to say in such circumstances. Heck, even among what we'll call the non-bigoted, who perceive themselves to genuinely believe their responses, can we truly dismiss the possibility that they have merely convinced themselves of that from the same "training"? (I'm not saying everyone is secretly deluding themselves that they don't have latent bigotry; I'm positing the human mind is a complicated thing where our conscious thoughts and our unconscious motivations may not be entirely aligned as we like to think they are.)

But I digress again. Given that we all know what we're supposed to say, the poll results may be of specious worth at best, yet polls continue to not only be conducted but cited by campaigns and the media as having worth, and regarded by the public as something where the campaigns and media should consider worth citing. I think people generally regard the survey results as worthy to the extent those results reinforce what they individually already believe, and don't necessarily find them all that persuasive, as we grasp the answers are the "right" ones which may or may not be the honest ones, but polls haven't been abandoned.

And we won't even get into how the questions in polls have to avoid any level of nuance or intricacy, despite the way our minds are (as I noted) ridiculously complex.

What's even more fascinating is the notion that how people, when they stop and think about a theoretical question in a survey, use that same sort of behavior when they get into the voting booth, as though the one predicts the other accurately. These theoretically qualified candidates that are cited in the surveys never seem to be the ones who are actually on the ballot, and then their race or religion or sexual orientation is probably far down on the list of reasons not to vote for them.

But what do you think? Please rate on a scale of 1 = strongly agree to 5 = strongly disagree…

1 comment:

  1. I think polls are purely advertising by groups, designed to sway people via the "herd" mentality. Either you'll want to agree with the majority of the poll's respondents, or you'll give up opposing the majority, is the thinking. I ignore polls. In the United States, political polls traditionally over-sample Democrats. It used to be that they exclusively sampled landlines, which are overwhelmingly Democrat (unemployed, welfare, elderly). Now I'm not sure that they do that anymore, but they find other ways to overstate liberal responses.


So, what do you think?