Why six points? Or 10 points? Or 15 points? Or two points? We're not told
Gloucester, MA (PRWEB) October 28, 2008
Kevin Clancy, author of "Your Gut Is Still Not Smarter Than Your Head," has a problem with the polls that put Barack Obama and his running mate Joe Biden ahead of John McCain and Sarah Palin. Recent news stories in the Associated Press, The Boston Globe, and more have reported that political polls overestimate Obama's appeal by about six points and attributed the discrepancy to racism.
"Why six points? Or 10 points? Or 15 points? Or two points? We're not told," explains Clancy, a noted marketing strategy consultant and recent inductee into the Marketing Research Hall of Fame. "We're just asked to buy into the guesstimate that the polls are off by six points without any empirical support for the racism theory." Clancy believes the number may be right, but for reasons other than racism.
There have been many explanations for polling failure throughout history. In 1948, for example, the Readers Digest poll, at the time the most prestigious in the nation, predicted that Dewey, the Republican, would be a big winner over Harry Truman, the Democrat, in the presidential election. As we all know, Truman won big time and the perception of polling as a reliable resources has never been the same. The reason for "polling failure," as it turned out, was the unrepresentative nature of the Readers Digest sample. The magazine's readers and poll samples were more upscale and more Republican than the rest of America.
Clancy believes this year pollsters and the pundits will find themselves in trouble again. Although he admits it's still too early to tell the outcome of this election, he is certain that pollsters are overstating the appeal of Barack Obama by at least five points. If the Gallup poll says that Obama is leading 47 to 42 (a five point spread) then they are really dead even: If the polls (almost all of them) show Obama ahead by 52% to 44%, then Obama is really ahead by about three points, a statistically insignificant advantage.
The discrepancy in the case of Obama is no mystery, says Clancy. Most pollsters use a relatively straightforward polling technology: they ask people a single question about who they plan to vote for in the next presidential election. They present the options and let respondents decide. They do this on the phone, sometimes in person and increasingly over the internet. Each respondent is then weighted by his/her self-reported probability of voting or estimated probability of voting based on past returns for the region/state/city/zip code and/or demographic profile they exhibit. The more sophisticated the weighting, the more accurate the poll.
Sampling and weighting, however, are not the reasons why contemporary polls are routinely overstating support for Obama. The reason is social desirability response set which might be described as the generalized tendency for respondents in an interview to say nice (aka desirable or politically correct) things about themselves--to look good--and to avoid saying negative, politically incorrect things about themselves.
Since the 1930's psychologists and sociologists have studied the effects of social desirability response set--its magnitude and implications--but political pollsters have ignored such effects because in a typical poll or election they're not that important. If one is studying a simple phenomenon: brand choice in a low involvement category; do you prefer Coke or Pepsi; white, rye or pumpernickel; chocolate vs. vanilla, there is no clear right or wrong answer. There is no pressure on a respondent to offer one answer as opposed to another. There is no socially desirable or politically correct response.
But sometimes there is a clear answer, a socially desirable answer.
"Of course we lie about some things," Clancy admits. "We understate alcohol and cigarette consumption. We overstate sexual activity, our kid's grade point averages and the frequency of attendance at cultural events. Anytime we have an opportunity to say something nice about ourselves, we tend to take it. Conversely, if we have the opportunity to underreport negative or socially undesirable attitudes and behaviors, we'll take that too."
The percentage of Americans, for instance, who self-report cheating on their income tax returns, ever being arrested, even receiving a speeding ticket is 5 - 20 points lower than it really is, while the percentage who report earning a college degree, achieving A grades in high school and college and remaining faithful to our spouses is 5 - 20 points higher than the actual facts.
In a presidential election people go into a voting booth--close the curtains--and in complete privacy pull a lever or check a box for the candidate of their choice. In stark contrast, in polls an interviewer asks the question, and records the response. There's no privacy. People are watching. They're measuring. Therefore if there's one answer category more favorable or desirable than another, some people will feel the social pressure to give that answer.
This year, with Barak Obama, the first African American candidate running for the presidency, Clancy believes there is pressure on respondents in polls to give the socially desirable response, the politically correct answer. As a result more people will favor Obama in "public" than will vote for him in "private." This is not racism at work; it's social desirability response bias.
The real mystery, says Clancy, is why pollsters aren't checking for the existence of response bias this year. By asking respondents who their best friend is going to vote for, he says, they could gauge the level of bias present. Pollsters could also do their interviews in person in locations around the country where people can actually go into a simulated voting booth. In these cases, the prospective voters are convinced of their anonymity and are much more likely to express the truth than they are when responding to a stranger on the telephone. The latter method is more expensive to execute than conventional polling, but much more likely to yield valid answers in situations where social desirability response bias is likely to occur.
Clancy predicts that the Obama bias is about five points on average but varying from place to place and demographic group to demographic group. In the more liberal suburbs of Boston, Massachusetts where Clancy lives, for instance, "it's politically incorrect and socially undesirable to admit being a McCain supporter," the Obama bias is 10 points or more.
Interestingly, pollsters even if they know about bias, don't possess the technology to correct for it, Clancy explains. They can't reliably adjust their numbers to take it into account to produce a more accurate estimate of the potential outcome of the election.
Kevin Clancy has been consulting to many of the world's leading brands for over forty years. He has pioneered many of the applications of social science research to marketing decision making that are industry standards today. A prolific writer, he is the author of seven business books and numerous articles. For more information about him, visit http://www.thekevinclancy.com.