PRINCETON, NJ -- Independent women who are Catholic, middle aged, not college graduates, of average religiosity, and of mid-range incomes are most evenly split in their presidential candidate choices, and thus may be most "up for grabs" in the remaining weeks of the campaign.
The importance of the female vote has been underscored by the selection of Sarah Palin as John McCain's vice-presidential running mate. The reaction of women to Thursday night's debate between Palin and Joe Biden -- which will not be fully evident until the days to come -- will thus be carefully watched.
Taken as a whole, women who are registered voters tilt in their candidate preferences toward the Democrat Barack Obama, reflecting the usual gender gap that affects national politics today. But since over half of all voters in the United States are women, this broad-brush look at their preferences is only the beginning point.
Most importantly, about two-thirds of women today identify themselves as either Democrats or Republicans, and are highly likely to be voting for their party's candidate. A special Gallup analysis of more than 26,000 interviews conducted in September confirms that Democratic women overwhelmingly support Obama, and Republican women overwhelmingly prefer McCain. This leaves independent women, who are much more even in their vote-choice distribution.
It's a reasonable assumption that changing minds among partisan Democratic and Republican women will be difficult, although these groups remain highly important in terms of turnout. Indeed, motivating the "base" is a key dictum of modern-day politics. It has been argued that one motivation behind McCain's choice of Palin as his vice-presidential running mate may have been to increase enthusiasm among Republican women.
But it is the group of independent women who -- based on the fact that they are currently about evenly divided between the two candidates -- were perhaps the most important viewers of Thursday's debate, at least in terms of the possibility that their voting preferences will change.
Assuming that independent female registered voters are a key "swing" group in the campaign, it is useful to delve further into the ways in which they subdivide in terms of candidate preferences. Gallup's large sample of September interviews allows for a careful analysis of the vote patterns of demographic subgroups within the larger group of independent women. The accompanying table displays candidate choices among a number of these different subgroups of independent women.
Most of these patterns are microcosms of the divisions evident in the overall voting population this year -- e.g., younger Americans, regardless of their other characteristics, are more likely to support Obama than are those who are older.
All in all, the segments of independent women who are strongest in their support of Obama -- and thus who were presumably most sympathetic to Biden in Thursday's debate -- include:
those with no religious identification
those aged 18 to 34
those with college educations
those who seldom or never attend church
The segments of independent women who are most likely to support McCain and thus may have been sympathetic to Palin Thursday night include:
those who attend church weekly
those who are married
those aged 55 or older
Finally, the following groups of independent women are most closely divided in their candidate preferences, and -- it could be assumed -- were the most likely to be swayed one way or the other by the debate (and indeed the events of the final few weeks of the campaign):
those who do not have a college degree
those with no children under 18
the middle aged, 35 to 54
those making between $12,000 and $60,000 a year in household income
those mid-range in religiosity, attending church almost every week or monthly
In theory, because of a lack of a strong partisan attachment, independent women are more susceptible to changing their vote preference than are women who identify as Republicans or Democrats. Within the group of independent women, subgroups that are currently most evenly divided in their existing vote between the two major-party candidates may be in turn the most amenable to changing their vote preferences as a result of the campaign. For example, the finding that independent women who are Catholic were exactly split between Obama and McCain (in September interviewing) suggests that there is no strong tendency among this group taken as a whole to go for one candidate or the other. The individuals who make up the group may be less rigid in their support patterns, and thus more open to shifting.
Thus, if the McCain-Palin ticket is to pick up steam either as a result of Thursday night's debate or forthcoming campaigning, one logical place to look for that process to occur is among independent women who are Catholic, middle aged, not college graduates, of average religiosity, and of mid-range incomes.
Results are based on telephone interviews with 26,620 registered voters, aged 18 and older, conducted Sept. 1-29, 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. The maximum margin of sampling error for results based on interviews conducted with 13,362 female registered voters is also ±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.