Obama Maintains Eight Point Lead over McCain Among Registered Voters

ROCHESTER, N.Y. – August 12, 2008 – Senator Barack Obama retains an eight point lead over Senator John McCain among registered voters in the latest Harris Poll, virtually unchanged since the nine point lead he enjoyed in early July.

Some polls report the voting intentions of all adults, while others report on registered voters (or those who claim to be registered) and of so-called "likely voters." In this Harris Poll, it makes very little difference which analytic method is used. Obama enjoys an eight-point lead among all adults without using the leaner questions and a nine point lead when the leaner question is used. Among registered voters he has an eight point lead, with and without using the leaner question.

Typically the Harris Poll does not focus on likely voters until after the party conventions because we believe it is too early to make sound judgments about who is, or is not, likely to vote before Labor Day. Most polls also ask people who are not sure how they would vote a so-called "leaning" question, asking people which way they lean.

In many previous elections Republican candidates have done slightly better among registered voters than among all adults, reflecting the fact that they are somewhat more likely than Democrats to register to vote. The fact that there is no such difference in this poll may reflect the fact that many new voters registered to vote in the Democratic primary elections.

Demographic analysis of these voting intentions shows that this is not a typical election. Indeed, some demographic groups are behaving quite atypically. This demographic analysis also raises several warning signs for both candidates. These are some of the results from the latest Harris Poll of 2,834 U.S. adults surveyed online between August 1 and 7, 2008 by Harris Interactive^®:

• Age: Obama’s lead is critically dependent on his 33 point lead among Echo Boomers, people under 32 – but they are usually are the least likely generation to actually vote. If they do not vote very heavily in November this would hurt Obama’s chances. McCain leads among "Matures," those over 62, by nine points, the generation that is usually the most likely to vote;
• Gender: There is currently no gender gap in this election. Obama’s lead is almost the same among men (+8 points) and women (+10 points);
• Race: Obama currently wins virtually all the African-American vote (+91) and most of the crucial Hispanic vote (+25) but loses the white vote by eight points. In the past whites were more likely to vote than blacks or Hispanics;
• Education: In many elections, those with only a high school education or less vote heavily Democratic; in this poll, McCain actually leads by four points. This is the group where Obama’s race is most likely to hurt him. Obama leads among those with some college (+16), college graduates (+14) and has his biggest lead among those with post-graduate education (+30). However because more educated people are more likely to vote these findings are not good news for McCain;
• Income: There is only a modest correlation between income and voting intentions. Obama leads among all income groups, with a 17 point lead among those with less than $35,000 and six points among those earning $75,000 or more;
• Party Identification: Unsurprisingly, most Democrats prefer Obama (+72) and most Republicans favor McCain (+71). The crucial Independent vote currently tilts toward Obama by 11 points;
• Political Philosophy: Conservatives favor McCain by 52 points and Liberals favor Obama by 73 points. The very important Moderate vote, a larger group than either Conservatives or Liberals, currently prefers Obama over McCain by 19 points.

"We’d like to get your opinion about how you might vote in the election for president on November 4^th, 2008. If the presidential election were held today, for whom would you most likely vote?"

Base: All adults

*Those who said "not sure" were asked "Well, if you had to say, toward whom would you lean for president?"

Note: Percentages do not add up to exactly 100% due to rounding

"We’d like to get your opinion about how you might vote in the election for president on November 4^th, 2008. If the presidential election were held today, for whom would you most likely vote?"

Base: Registered Voters

*Those who said "not sure" were asked "Well, if you had to say, toward whom would you lean for president?"

Note: Percentages do not add up to exactly 100% due to rounding.

"We’d like to get your opinion about how you might vote in the election for president on November 4^th, 2008. If the presidential election were held today, for whom would you most likely vote?"

"Well, if you had to say, toward whom would you lean for president?"

Base: All Registered Voters

Note: Percentages do not add up to exactly 100% due to rounding.

*Less than 0.5%.

"-" indicates no response

This Harris Poll^® was conducted online within the United States between August 1 and 7, 2008 among 2,834 adults (aged 18 and over). Figures for age, sex, race/ethnicity, education, region and household income were weighted where necessary to bring them into line with their actual proportions in the population. Propensity score weighting was also used to adjust for respondents’ propensity to be online.

All sample surveys and polls, whether or not they use probability sampling, are subject to multiple sources of error which are most often not possible to quantify or estimate, including sampling error, coverage error, error associated with nonresponse, error associated with question wording and response options, and post-survey weighting and adjustments. Therefore, Harris Interactive avoids the words "margin of error" as they are misleading. All that can be calculated are different possible sampling errors with different probabilities for pure, unweighted, random samples with 100% response rates. These are only theoretical because no published polls come close to this ideal.

Respondents for this survey were selected from among those who have agreed to participate in Harris Interactive surveys. The data have been weighted to reflect the composition of the adult population. Because the sample is based on those who agreed to participate in the Harris Interactive panel, no estimates of theoretical sampling error can be calculated.

J30279c August

Q2776, 2785