PRINCETON, NJ -- In June, voters with a high school education or less were as likely to prefer John McCain as to prefer Barack Obama for president. That represents a change from earlier in the campaign -- McCain led Obama among this group during the prior three months, but by diminishing margins.
Voters with less formal education have been a core Democratic constituency in the last four presidential elections, preferring Bill Clinton, Al Gore, and John Kerry to their respective Republican opponents. But it is a group that Obama -- who has unusually high appeal to well-educated voters -- has struggled to appeal to, both in the primaries against Hillary Clinton and in the early part of the general-election campaign versus McCain. In recent months, he has clearly begun to make gains among voters with less education, even though historically, his performance still trails that of other Democratic presidential candidates.
Since March, Obama has become increasingly competitive with McCain among men and women with less formal schooling, as well as among white and nonwhite voters who did not attend school beyond 12th grade. Among each of these subgroups of voters, the Obama-McCain gap has moved 6 or 7 points in Obama's favor over the past four months.
As a result, Obama now holds a lead over McCain among women with less formal education, and trails McCain by just 4 points among men of similar educational attainment. Obama still trails McCain by a considerable 17 points among whites with less education, but has expanded his already formidable lead among nonwhites with high school educations or less.
Overall, Obama's gains in the polls since March (from an average 1-point deficit in March to a 3-point lead in June among all registered voters) have come disproportionately from voters with less formal education. Among the other educational groups, his support has stayed relatively constant or gained only a little in recent months. But Obama was already competitive with McCain among voters with differing levels of postsecondary education when Gallup began tracking general-election preferences in March. Now, he is competitive with McCain among all educational groups, and holds a wide lead among those with the most formal education (those with postgraduate educations).
Just as they appear to be in the general-election phase, highly educated voters were a key part of Obama's winning coalition in the primaries, and much has been made of his appeal to this group throughout the campaign. But his strong support among highly educated voters is driven by his appeal to women with postgraduate educations. In June, Obama led McCain by 29 points among these voters, while running only about even with McCain among men with postgraduate educations. These patterns have been highly stable each month during the election campaign.
Obama's gains among less-educated voters are surely a positive sign for his campaign, and certainly help allay some of the concerns about his electability, specifically regarding his struggles in appealing to this core Democratic group. While he is on track to do less well than other recent Democratic presidential candidates among voters with less formal education, his great appeal to college-educated voters (in particular, those whose schooling continued after getting a college degree) helps to offset that. Women with high educational attainment appear to be one of Obama's strongest constituencies.
Results are based on telephone interviews with 23,731 registered voters, aged 18 and older, conducted in June 2008 as part of Gallup Poll Daily tracking interviewing. For results based on the total sample of national registered voters, one can say with 95% confidence that the maximum margin of sampling error is ±1 percentage point.
For results based on the sample of 5,470 registered voters with a high school education or less, the maximum margin of sampling error is ±1 percentage point.
For results based on the sample of 7,621 registered voters with some college education, the maximum margin of sampling error is ±1 percentage point.
For results based on the sample of 5,505 registered voters with a four-year college degree and no postgraduate education, the maximum margin of sampling error is ±1 percentage point.
For results based on the sample of 4,953 registered voters with postgraduate education, the maximum margin of sampling error is ±2 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).
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.
To provide feedback or suggestions about how to improve Gallup.com, please e-mail email@example.com.