WVWV Action Fund Marriage Gap not Evident in Ratings of all Democratic Candidates

Gallup Poll

In each of the last three presidential elections, there have been sharp differences in the voting preferences of married and unmarried voters. Married voters have supported the Republican presidential candidate over the Democratic candidate by an average of 19 percentage points, while unmarried voters show a similarly large preference (17 points) in favor of the Democrat. Such marriage gaps have also been evident in the more basic attitudes toward candidates in recent years, including their favorable ratings.

An analysis of favorable ratings for the leading 2008 presidential candidates shows the prevailing marriage gaps evident in ratings of all Republican candidates, with married Americans evaluating them much more positively than unmarried Americans. While New York Sen. Hillary Clinton and former vice president Al Gore exhibit the typical Democratic pattern of being rated much less favorably by those who are currently married than by those who are not, Illinois Sen. Barack Obama and former North Carolina Sen. John Edwards are rated similarly by both married and unmarried Americans.

The analysis is based on data from the last four Gallup polls, conducted from mid-April through early June 2007. Each poll included favorable ratings on the leading presidential contenders. The combined data set consists of interviews with more than 4,000 Americans, including just more than 2,200 who are currently married and slightly less than 1,700 who are not married. The group of unmarried Americans includes those who have never been married (comprising 37% of this group), those living with a partner (13%), and those who are divorced (24%), separated (5%), and widowed (21%).

The Marriage Gap and Democratic Candidates

All of the leading Democratic presidential contenders have favorable ratings around 50% among the general public. But the underlying ratings by subgroup are not similar.

Obama's overall 53% favorable rating is the same among married and unmarried Americans. Edwards is rated only slightly more positively by those who are not married (52%) than by those who are (49%). But Gore and Clinton show large differences in their favorable ratings by marital status, consistent with what is normally the case for Democratic political figures. There is a 14-point gap in the public's ratings of Clinton by those who are married (42%) and unmarried (56%), and a similar 14-point gap in ratings of Gore (45% and 59%, respectively).

Favorable Ratings of Leading Democratic Presidential Candidates
By Marital Status

	All Americans	Married	Not married	Marriage Gap (Married- Not married)
	%	%	%
Barack Obama	53	53	53	0
John Edwards	51	49	52	-3
Al Gore	51	45	59	-14
Hillary Clinton	48	42	56	-14

These marriage gaps expand when factoring the gender of respondents. For example, Clinton's favorable rating among married men is 37% while among unmarried women it is 61%, a difference of 24 points. Gore's rating shows a similar divide among those two groups. On the other hand, there is very little difference in ratings of Obama by any of the martial status gender groups, with each giving him a favorable rating between 50% and 55%. Edwards is rated somewhat less favorably by married men than by the other groups.

Favorable Ratings of Leading Democratic Presidential Candidates
By Marital Status and Gender

	Married men	Married women	Unmarried men	Unmarried women
	%	%	%	%
Barack Obama	53	52	50	50
John Edwards	45	53	51	54
Al Gore	42	48	54	62
Hillary Clinton	37	47	49	61

All of the Democrats are rated most positively by unmarried women, though in the cases of Edwards and Obama ratings among this group are not significantly better than those from all the other groups.

The Marriage Gap and Republican Candidates

In the field of Republican candidates, former New York City mayor Rudy Giuliani (59%) and Arizona Sen. John McCain (50%) have much higher favorable ratings than former Tennessee Sen. Fred Thompson (26%) and former Massachusetts Gov. Mitt Romney (25%), mostly because a majority of Americans still have no opinion of Thompson and Romney. Even so, all of the candidates are rated more positively by married Americans than by unmarried Americans. The marriage gap is in the double digits for Giuliani, McCain, and Romney, and is 8 points for Thompson.

Favorable Ratings of Leading Republican Presidential Candidates
By Marital Status

	All Americans	Married	Not married	Marriage Gap (Married - Not married)
	%	%	%
Rudy Giuliani	59	64	53	+11
John McCain	50	56	43	+13
Fred Thompson	26	30	22	+8
Mitt Romney	25	30	19	+11

Each Republican is also rated most positively (or at least as positively as any other group) by married men, and least positively by unmarried women. The resulting differences in their ratings between these groups range from a low of 12 points for Giuliani to a high of 21 points for Romney.

Favorable Ratings of Leading Republican Presidential Candidates
By Marital Status and Gender

	Married men	Married women	Unmarried men	Unmarried Women
	%	%	%	%
Rudy Giuliani	64	64	54	52
John McCain	59	53	47	41
Fred Thompson	37	22	26	18
Mitt Romney	36	23	24	15

Giuliani and Obama are the only candidates of either party who have favorable ratings of 50% or higher from those in all marriage by gender groups.

Why Is There a Marriage Gap?

The marriage gap in ratings of political figures is due mostly to the fact that married and unmarried Americans are aligned with different political parties. In these data, 49% of married Americans identify with the Republican Party or say they are independents who lean Republican, while 43% identify as Democrats or lean to the Democratic side. Among unmarried Americans, there is a nearly two-to-one Democratic preference, 61% to 31%.

Much of the difference comes from the party preferences of married men and unmarried women. While both unmarried men and women show a decided Democratic leaning, it is much more pronounced among unmarried women. Married men are solidly Republican (52% to 39%), while married women are in fact evenly divided in their partisan leanings, making them a potentially crucial swing voter group in 2008.

Party Affiliation
By Marital Status and Gender

	Married men	Married women	Unmarried men	Unmarried Women
	%	%	%	%
Democrat/Democratic leaner	39	46	56	64
Republican/Repbulican leaner	52	46	34	29

Americans may divide politically on the basis of marital status in part because of religious influences -- Gallup has found very different attitudes toward marriage among those who are more religious (and likely to align with the Republican Party) and those who are not that religious (and likely to align with the Democratic Party). Those who see a religious significance in marriage are probably more likely to want to get married and to stay married than those who are not. There may also be an economic basis, as those who settle down and are married may prefer the Republican platform of lower taxes and more limited government services.

But the different patterns in ratings of Obama and Edwards versus Clinton and Gore are not merely due to the party affiliations of the married and unmarried. Obama is rated more positively by married Democrats (71%) than by unmarried Democrats (65%), as is Edwards (74% to 63%). On the other hand, Gore and Clinton are rated slightly more positively by unmarried Democrats than married Democrats, both by about 5 percentage points.

It is unclear why these different patterns exist among the Democrats. One possibility is that Gore and Clinton may be perceived by Americans as the more liberal candidates than Edwards and Obama, something that may appeal more to the unmarried.

Survey Methods

These results are based on 4,027 combined telephone interviews with randomly selected national sample of approximately 1,000 adults each, aged 18 and older, conducted Apr. 13-15, May 4-6, May 10-13, and June 1-3, 2007.

For results based on the total sample, one can say with 95% confidence that the maximum error attributable to sampling and other random effects is Â±2 percentage points.

For results based on the sample of 2,286 married adults, the maximum margin of sampling error is Â±2 percentage points.

For results based on the sample of 1,696 unmarried adults, the maximum margin of sampling error is Â±3 percentage points.

For results based on the sample of 1,264 married men, the maximum margin of sampling error is Â±3 percentage points.

For results based on the sample of 1,022 married women, the maximum margin of sampling error is Â±3 percentage points.

For results based on the sample of 654 unmarried men, the maximum margin of sampling error is Â±4 percentage points.

For results based on the sample of 1,042 unmarried women, the maximum margin of sampling error is Â±3 percentage points.

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.