NEWSWEEK Poll: Obama Loses Some Ground

After an important primary win in Pennsylvania, Hillary Clinton has reduced Democratic rival Barack Obama's double-digit lead among registered Democrats and voters leaning Democratic by more than half, according to the latest NEWSWEEK poll. Plagued by controversies over Rev. Jeremiah Wright's comments and the candidate's own "bitter" remarks, Obama has seen his favorability rating slip significantly in the last week, the poll found.

The survey found that Clinton now trails Obama by seven points, down from 19 just one week ago. The previous NEWSWEEK poll, conducted on the eve of the Pennsylvania primary, found that more than half (55 percent) of registered voters believed Obama was more electable, while 33 percent gave the edge to Clinton. The current poll finds Obama leading 46 percent to 38 percent. (For complete poll data, click here.)

One of the more problematic results for Obama was that four in 10 of registered voters (including Republicans and independents) now have an unfavorable opinion of him--and the same number said there is "no chance" they will vote for Obama if he becomes the nominee. Four in 10 registered voters (41 percent) say they have a less favorable opinion of Obama based on his association with his former pastor, Rev. Wright, whose racially and politically inflammatory sermons have been circulated on the Internet and covered in the media. A similar number (42 percent) say they will not vote for Obama because of comments he made about "bitter" small-town residents clinging to guns and religion.

Even so, the NEWSWEEK poll indicates that despite the political damage inflicted over recent weeks, Obama still edges out McCain in a trial head-to-head heat for the White House, 47 to 44 percent. That margin was only one point wider a week ago. Clinton—whose own favorability rating has not improved even as Obama's has slipped—also holds a three point lead with 48 percent of the vote to the Arizona senator's 45. Among all registered voters, more than half (53 percent) still hold a favorable opinion of Obama, compared to the 47 percent who view Clinton favorably and 51 percent who have favorable views of McCain.

Last week's poll did not point to much difference in general election preferences according to class. But after the Pennsylvania primary, support for all three candidates tends to break down along class divides. When asked to choose between Clinton and McCain, working class and poor white voters appear evenly split (47 percent for Clinton, 46 for McCain). When Obama is swapped in for Clinton, McCain's support climbs to 53 percent, with Obama netting just 35 percent. Among upper- and middle-class whites, Obama runs marginally better than Clinton against McCain—the Illinois senator trails McCain 51 to 40 percent; Clinton trails 54 to 40 percent.

The new poll also suggests that charges of elitism against Obama have resonated somewhat in a campaign that has perhaps focused undue attention on his weak bowling skills and his calorie-conscious campaign trail eating habits. A third (31 percent) of working-class whites agreed with the statement that Obama "looks down on people like you." The other two candidates scored no differently—33 percent felt that statement applied to Clinton, 31 percent McCain. But asked bluntly whether they felt Obama was "elitist" or "down-to-earth," 25 percent said elitist, 53 percent went with the latter. Working class and poor whites are also less inclined to view Obama as the candidate who would "fit in well with people in your local community." Only 45 percent agreed that this fit Obama, compared to 56 percent for McCain and 53 percent Clinton. Still, they viewed Obama as the candidate least likely to favor the interests of the rich if elected (10 percent, compared to McCain's 45 percent and Clinton's 29 percent).

Americans remain somewhat schizophrenic on the question of race. Among registered Democrats nationwide, Obama owes his current lead in large margin to nonwhite voters (62 percent of whom support him, compared to 30 percent whites). Nineteen percent of American voters say that the country is not ready to elect an African-American president (though interestingly, an even larger percentage, 25, say the United States isn't ready for a woman president). Yet when asked if Obama's race makes a difference, only 3 percent of whites say Obama's race makes it less likely they would support him, while 5 percent of whites (and 16 percent of non-whites) say his race would make it more likely they would support him. More than half the voters said they think "most" or "some" white voters will "have reservations about voting for a black candidate that they are not willing to express." And yet a full three quarters (74 percent) of registered voters think America is ready to elect an African-American president (up from 59 percent last July). Still, white Democrats favor Clinton by a 47 to 41 percent margin.

Asked which candidate better "understands the problems and concerns of people like you," nonwhite Democratic voters chose Obama 56 percent to Clinton's 28. Upper- and middle-class whites also went for Obama (48 to 36 percent). Working-class or poor whites, however, felt more connected to Clinton (52 percent vs. 30 percent).

On the experience question, which Clinton's campaign is pushing hard, 47 percent of registered voters believe Obama doesn't have enough experience to be a good president (45 percent thinks he does have enough experience). In July 2007, 39 percent thought Obama had enough experience; 35 percent said he didn't.

On a final, somewhat surprising, note, 13 percent of the poll's respondents believed that Obama, whose middle name is Hussein, is a Muslim. The candidate is a Christian. A quarter (26 percent) didn't know what religion he practices—half (52 percent) got it right.

Methodology statement (4/26/08) Results are based on telephone interviews with 1,203 registered voters nationwide. Interviewing was conducted April 24-25, 2008. The overall margin of sampling error is plus or minus 3 percentage points for results based on 1,200 registered voters. Results based on smaller subgroups are subject to larger margins of sampling error. The margin error is plus or minus 5 for results based on 600 registered Democrats and Dem leaners. In addition to sampling error, the practical difficulties of conducting surveys can also introduce error or bias to poll results. This poll was conducted by Princeton Survey Research Associates International.