PRINCETON, NJ -- Among American registered voters who are married and whom Gallup interviewed Aug. 1-19, John McCain is leading Barack Obama by 13 points; among unmarried American voters, Obama has a 22-point margin.
A large part of the explanation for this marriage gap resides in the basic fact that the two major political parties are fundamentally divided by marital status. Almost two-thirds of Americans who identify with the Republican Party are married, while a majority of Democrats are unmarried.
This marriage gap is one of several such demographic gaps in support for the two major-party candidates. There is, for example, a significant gender gap (Obama does better among women; McCain, among men) and a significant age gap (Obama does better among younger Americans).
Since men are significantly more likely than women to be married in America today, and since those under age 35 are significantly more likely to be unmarried than are those who are older, could it be that the marriage gap is merely a reflection of these other two basic demographic characteristics?
An analysis of Gallup's August data would answer "no" to that question. The differences in support by marital status persist across gender and age groups.
Although women are in general somewhat stronger in their support for Obama, married women tilt more toward McCain (by 46% to 42%) while unmarried women favor Obama. The same pattern holds for men.
Younger Americans in general are among Obama's strongest supporters. Yet among those 18 to 34 who are married, McCain manages to best Obama by 4 points (47% to 43%), while among 18- to 34- year-olds who are unmarried, Obama wins by an overwhelming 36-point margin (63% to 27%). The same pattern holds among those who are older. Among both 35- to 54-year-olds and those 55 and older, those who are married skew toward McCain while those who are not married skew toward Obama.
Results are based on telephone interviews with 16,941 registered voters, aged 18 and older, conducted Aug. 1-19, 2008. For results based on the total sample of national adults, one can say with 95% confidence that the maximum margin of sampling error is ±1 percentage point.
Interviews are conducted with respondents on land-line telephones (for respondents with a land-line telephone) and cellular phones (for respondents who are cell-phone only).
In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.
To provide feedback or suggestions about how to improve Gallup.com, please e-mail firstname.lastname@example.org.