PRINCETON, NJ -- While 80% of registered voters are aware of former Secretary of State Colin Powell's endorsement of Barack Obama for president, only 12% of this attentive group say the endorsement makes them more likely to vote for Obama, while 4% say it makes them less likely to vote for the Democratic nominee.
Powell's announcement of his endorsement of Obama on Sunday's "Meet the Press" program received a considerable amount of press coverage. This may help account for the fact that 80% of registered voters interviewed Tuesday night were able to correctly identify not only that Powell had made an endorsement, but also that the recipient of that endorsement was Obama. (Eighteen percent of registered voters were unaware of the endorsement, and another 2% either misidentified or were unable to identify the person Powell endorsed.)
But only a relatively small 12% of those aware of the Powell endorsement indicated in the survey that Powell's gesture makes them more likely to vote for Obama. (This means only 10% of the entire registered-voter population is both aware of the endorsement and report that it makes them more likely to vote for Obama.) An even smaller group of 4% of those aware of the endorsement (or 3% of all voters) say it makes them less likely to vote for Obama.
It might be expected that Obama supporters would be more aware of Powell's endorsement of their candidate, but the results show no difference in awareness between current supporters of Obama and current supporters of John McCain.
The Obama campaign may have hoped that the endorsement would move McCain supporters into the Obama camp. However, its exact impact is difficult to pinpoint, because some voters who are currently for Obama, for example, may have been McCain supporters before the endorsement. Having said that, the Tuesday night data show that 20% of current Obama supporters (a group that could in theory include voters who are recent converts to Obama as well as longtime Obama voters) say the endorsement makes them more likely to vote for Obama. And only 9% of current McCain supporters (a group that could include defectors from Obama since the endorsement) say it makes them less likely to vote for Obama.
These data suggest that the high-profile Powell endorsement of Obama has had fairly minimal direct impact. A high percentage of voters are aware of the endorsement, but only 12% of those aware and 8% of the total population of registered voters say it makes them more likely to vote for Obama. The endorsement produced a small backlash against Obama, all from current McCain voters, many of whom were not going to vote for Obama anyway.
There has been little major change in the structure of the presidential race coincident with Sunday's endorsement: Obama was ahead before the endorsement and remains ahead in the latest Gallup Poll Daily tracking results. The significant majority of both candidates' current supporters say the endorsement did not affect their vote choice. So it's difficult to prove the endorsement has had a major impact on the status of the two candidates in the presidential race -- which has, in recent weeks, of course, been moving in Obama's favor anyway.
Given that most of the "more likely to vote for Obama" sentiment comes from current Obama supporters themselves, one effect of Powell's gesture may have been to shore up support among the Democratic base -- and thus to increase Democrats' motivation and potentially their turnout on Election Day. It is also possible that the endorsement had a more subtle effect on voters' perceptions of Obama -- one that has not yet translated into a change in vote choice, but could be a building block that would have this effect before the voter casts his or her ballot.
Results are based on telephone interviews with 845 registered voters, aged 18 and older, conducted Oct. 21, 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 ±4 percentage points.
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).
Polls conducted entirely in one day, such as this one, are subject to additional error or bias not found in polls conducted over several days.
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.