The 2008 presidential campaign is, for all intents and purposes, underway. Pundits and political analysts have noted, at least in passing, the breadth and depth of the Democratic field, but most of the media attention (and, it would seem, the early money) has gone to just two candidates: Hillary Clinton and Barack Obama.
As evidence of that, consider the following. By my count, there are five stories about Hillary Clinton on the politics page of the New York Times website this morning (1 2 3 4 5) and one story about Barack Obama. None of the other Democratic candidates gets a headline, though one story does at least mention John Edwards. The only Democratic contender for president to get a mention on the WaPo's main politics page is Senator Clinton, though if you click through to the elections page there is at least an older story about Obama there, and another story regarding the diversity of the Democratic field. Things look much the same at MSNBC, CNN, the Chicago Tribune, and other media sites.
There is, however, some recent research that suggests America may not be quite as ready to welcome its first woman (and possibly black) president as it would like to believe itself to be:
DeKalb, Ill. — Polls consistently show that about nine of 10 Americans say they would vote for a qualified female presidential candidate. But new research indicates a significant percentage of survey respondents are hiding their true feelings to avoid the appearance of being sexist.A team of political scientists led by Matthew Streb of Northern Illinois University found that slightly more than one in four Americans would likely be unsupportive of a female candidate for president. The findings were consistent among both male and female respondents across several demographic groups.
(Emphasis in original)
Their research has not yet been published, though it has been accepted by the Public Opinion Quarterly and will appear in that journal later this spring, which indicates to me that there is sufficient weight behind it for it to be taken seriously. The linked press release doesn't go into enough detail for me to be able to evaluate the study's research design with any degree of clarity, but the phenomenon they think they're seeing in the poll results they've analyzed is one that is well known to survey designers--and one of the hardest ones to control for.
Here's what the press release has to say about the survey's design:
"Polling is usually fairly accurate, but there is a long line of research supporting the influence of social desirability, or the tendency of respondents to provide socially desirable responses to questions dealing with controversial issues," Burrell said. "Most public opinion polls don’t tease out whether people are hiding their true feelings."In order to do just that, the researchers employed a well-established public-opinion technique known as "the list experiment" to measure public sentiment on female candidates for president.
A baseline group of survey respondents was asked how many of the following four statements make them "angry or upset:"
- The way gasoline prices keep going up.
- Professional athletes getting million dollar-plus salaries.
- Requiring seat belts to be used when driving.
- Large corporations polluting the environment.
A fifth statement was added to the list given to the test group:
- A woman serving as president.
The researchers attributed an increase in the mean number of angry items in the test group to the "woman serving as president" statement. The list experiment question was added to an Opinion Research Corporation CARAVAN telephone poll, a national survey of Americans conducted in March 2006. Of the 2,056 respondents, roughly half were given the test question.
"Traditional polls find anywhere from 5 percent to 15 percent of the public say they will not vote for a female presidential candidate," Streb said. "We found that roughly 26 percent of respondents expressed anger over a female president. It is extremely likely that those who are angered by the prospect of a female president are not going to vote for a woman."
The one potential quibble I have with the conclusion voiced by Streb in that last paragraph is that of the five statements on their list, three of them make me "angry or upset"--and it isn't the one about a woman serving as president. (Nor would it be the one about a black man serving as president, if they were to repeat the survey using that statement.) With energy prices being what they are (and particularly given allegations that oil companies deliberately lowered gasoline prices in the runup to the 2006 midterm elections, and have recently begun to raise them again), with concerns about global warming and the environment on the rise, and with some of the recent news about absolutely obscene salaries being paid to professional athletes, many of whom have recently been in the news for other, less savory reasons, I have to wonder whether at least some of the increase Streb and Burrell are seeing in their list experiment data aren't due to statistical noise. I'll have to wait for the article to come out to find out whether they've accounted for that possibility. But meanwhile, given that there is at least some evidence (h/t to Devilstower) to suggest that the road ahead may not be as rosy or as smooth for either Clinton or Obama as other data seem to indicate, I suggest we treat this early polling data on the 2008 horse race with even more skepticism than usual.
Comments