What Birtherism Teaches Us About Polling

Though in the past few days, given the death of Osama bin Laden, the news cycle has dramatically changed, I still wanted to draw your attention to a recent article in The Atlantic concerning the “birther” debate. Speaking for myself, and not Research Access, I feel the need to point out that I would normally never want to draw your attention to the birther debate; I think it’s political silliness. But from my perspective as editor of Research Access, this article has a real market research slant worth considering.

The article is entitled, “What Birtherism Teaches Us About Polling: An Expert’s Take.” The article discusses a theory put forth by Gary Langer, who heads up Langer Research Associates and directs polling for ABC News.
A wave of recent surveys have shown that doubts about Obama’s birthplace are stunningly prevalent. In aCBS/New York Times poll, 25 percent of all respondents and 45 percent of Republicans said they do not think Obama was born in the United States. A total of 18 percent said they weren’t sure. According to Fox News, 24 percent do not think Obama was born in America.

Maybe, just maybe, those poll respondents don’t actually think what they say they think, Langer suggests. Maybe they say all this for some other reason — such as that they just don’t like the president.

Mr. Langer’s theory tackles the concept of “expressed belief” vs “affirmed belief.” In other words, what people say in response to a survey question, versus what they actually believe. According to the article, Langer tackled this topic further, along with Patrick Moynihan of Harvard and Peyton Craighill of The Washington Post, in a paper presented at the May 2010 American Association for Public Opinion Research Conference. The paper discussed how political and policy opinions might be influencing the numbers on global warming. They summarized their conclusion as follows:
We suggest further that there can be a message-sending element to the way respondents answer survey questions – not always to answer the question in the way we imagine, but in the way they desire. Respondents who oppose or are skeptical about proposed policy solutions on global warming, yet who see such policies as increasingly likely given the change in administration, may be more apt to express opposition to such policies by any means available – including by withdrawing their expressed belief that global warming is occurring. They use such questions as a vehicle to express antipathy toward the solution, not to voice a firm disbelief in the existence of the problem.

I think we’ve all known this for quite some time – responses to surveys do not always contain an affirmed belief. Often they express what the respondent thinks the questioner wants to hear, or even worse, they respond based on the way the question was worded, which lead them to an emotional conclusion. This is rocky terrain for a market researcher, to be sure. Polls such as these have long been used as the concrete measure of public opinion, and not just in politics. The market research industry as a whole runs the risk of eroding the value of its products and services if we don’t get this right.

I’d love to hear your thoughts. Have you experienced this issues of expressed vs. affirmed belief in your own studies? How have you tried to counter it (and/or account for it) in the past? How can we as market researchers continue to ensure that our studies accurately reflect the affirmed belief of our respondents?

About Joshua Hoffman - Joshua Hoffman is Editor-in-Chief of Research Access.