Analysis: Impact of Personal Characteristics on Candidate Support

GALLUP NEWS SERVICE

PRINCETON, NJ -- A great deal of attention has been paid to the personal characteristics of potential presidential candidates and the impact of these traits on the candidates' chances of winning the White House. These considerations are particularly interesting this year because the various potential candidates have a wide variety of traits heretofore not seen in a U.S. president: a woman (Hillary Clinton), a Mormon (Mitt Romney), a black person (Barack Obama), someone 72 years old (John McCain), and thrice-married (Rudy Giuliani, Newt Gingrich). Three of these five characteristics describe potential Republican candidates (Mormon, aged 72, thrice-married), while two describe potential Democratic candidates (woman, black).

For many years, Gallup polls have tracked the public's views on electing a female or a black president. This year, Gallup has updated the public's reactions to race and gender, and added in other of the traits listed above. (See Related Items for more details).

The basic pattern of reactions to the five key traits are as followed, based on an aggregated sample involving a total of over 2,000 interviews conducted in two surveys (Feb. 9-11, 2007 and March 2-4, 2007).

Of significant interest is the relationship between specific subgroups of the U.S. population, based on demographic and political variables, and reactions to these presidential candidate characteristics. There is a natural interest, for example, in the question of whether or not women are more or less likely than men to support a female president. Are older Americans more accepting of an older president? And what are Republicans' reactions to the traits specific to Republican candidates?

The following tables -- based on the combined samples as described above -- illustrate how different groups of Americans feel about voting for presidential candidates of different backgrounds. In all instances, the question wording described the candidate as one who was an "otherwise well-qualified candidate" nominated by the person's party but who happened to be [characteristic]. The results show not only the percentages of who would vote for the described candidate and who would not, but also the results of a follow-up question that separated those who said they would vote for a candidate into those who would feel "completely comfortable" and those who would vote, but "with reservations." Each table is followed by a listing of key points.

 

Key Points About Voting for a Female Presidential Candidate

There is no gender gap in these perceptions. Men and women are equally likely to say they would be willing to vote for a woman for president.

Senior citizens are less likely than those who are younger to support a female presidential candidate. Only 79% of adults aged 65 and older say they would vote for a woman for president, significantly below the sample average.

Republicans are less likely than are independents or Democrats to say they would vote for a woman. This partisan variation could reflect Republicans' specific thoughts about a Hillary Clinton candidacy, even though the question was only asked in a generic sense, mentioning no candidate names and specifying the candidate was of the person's own party.

Religious preference plays a role in views of supporting a woman for president, with white Protestants and other Christians less likely than those in other religions to support a woman for president. These groups are disproportionately likely to be Republicans.

 

Key Points About Voting for a Mormon Presidential Candidate

There are only slight variations in willingness to vote for a Mormon for president by partisanship, with 71% of Republicans, 78% of independents, and 74% of Democrats saying they would be completely comfortable voting for a Mormon.

White Protestants and other Christians are less likely than Catholics, those who prefer other religions, and those with no religious affiliation to vote for a Mormon for president, though a majority of all groups say they would vote for a Mormon. Also, Americans who attend religious services weekly or almost weekly are less likely than those who go to church less frequently to support a Mormon candidate.

 

Key Points About Voting for a 72-Year-Old Presidential Candidate

There are not, as might be expected, significant variations by age in these perceptions. Senior citizens are no more likely to say they would vote for a 72-year-old candidate than those who are younger. In fact, those 65 and older are slightly more likely to say that they would not vote for a 72-year-old candidate than is the case for those who are under 50.

Perhaps due to the fact that Republican candidate John McCain would be 72 years old at the time of the next election, Republicans are more likely than Democrats to say they would vote for a presidential candidate that is 72.

Just 44% of blacks say they would vote for a 72-year-old candidate, which is lower than the 59% among whites.

 

Key Points About Voting for a Thrice-Married Presidential Candidate

Conservatives of either party are slightly less likely to support a thrice-married candidate than moderate or liberal members of the same party. However, a majority of all political and ideological groups say they would vote for a thrice-married candidate.

There is a relationship between church-going frequency and willingness to support a candidate who has been married three times. The more frequently Americans report going to church, the less likely they say they are to vote for a candidate who has been married three times.

Men are more likely than women to support a thrice-married presidential candidate.

Americans 65 and older are less likely than those under 65 to say they would support a candidate for president who has been married three times.

 

Key Points About Voting for a Black Presidential Candidate

Although the sample size of blacks in this aggregate sample is low (N=125), these data give no indication of a major difference between blacks and whites in their expressed willingness to vote for a black candidate.

More generally, there are few meaningful differences across any of the subgroups included in this analysis in expressed willingness to vote for a black candidate. The only exception is among Americans 65 and older, who are slightly less likely to indicate willingness to vote for a black candidate at all, and more likely to express reservations about voting for a black candidate.

Bottom Line

As has been widely discussed, these initial reactions to the characteristics of generic presidential candidates are not necessarily predictive of actual voting behavior. A voter may object in principle to voting for a Mormon, or a 72-year-old candidate, but the specific positives of an actual candidate who has these traits may override initial objections. However, what is interesting in the data reviewed here are the fascinating interrelationships between the characteristics of the voters and views on the impact of the characteristics of the candidates. One might expect that voters who share the characteristic under discussion would be more likely to vote for that type of candidate. But in several instances reviewed above, that's not the case. Blacks are no more likely than whites to say they would vote for a black candidate, women are no more likely than men to say they would vote for a female candidate, and older Americans are no more likely than those who are younger to say they would vote for a 72-year-old candidate.

There are also important political implications. Some observers have argued that a candidate like Rudy Giuliani or Newt Gingrich -- each married three times -- would have difficulty gaining the vote of conservative Republicans. These data, however, indicate that a majority of all subgroups would be willing to vote for such a candidate.

Survey Methods

Results are based on telephone interviews with 2,016 national adults, aged 18 and older, conducted Feb. 9-11, 2007 and March 2-4, 2007. For results based on the total sample of national adults, one can say with 95% confidence that the maximum margin of sampling error is ±2 percentage points. The maximum margin of sampling error for each individual subgroup presented in the tables above varies depending on the sample size involved, but is always higher than the sampling error for the total sample. 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.