Surveys and political identification (yet again)
I’ve written before about the problem of getting actual opinions rather than social or political identifications in surveys. US late-night TV host Jimmy Kimmel had a great demonstration. He went out on the streets of LA on the afternoon before the 2nd presidential-candidate debate, and asked people who they thought won “last night’s debate”. There wasn’t any shortage of people with confident opinions (presumably these are a selected subset of the most entertaining victims, but the point still holds) (via)
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »
I agree completely, but I’d also say that this is exactly the same as what happens in the voting booth. Polls suck, but voting is just a different kind of poll.
12 years ago
But in the voting booth you’re supposed to be answering the question “Who do you want to win the election?”
12 years ago
Sure. But what incentive has an individual voter to get that question right, given the near infinitessimal likelihood of changing the outcome of the election?
12 years ago