Introduction and Summary

With voters focusing more on the issues than on personal evaluations of the candidates and continuity factors increasingly favoring the Democrats, Al Gore holds a small but significant lead over George W. Bush. A survey of nearly 2,000 registered voters, conducted Aug. 24 – Sept. 10, finds Gore leading Bush by a margin of 47% to 41%. Gore’s lead remained fairly stable over the duration of the survey, as the impact of the party conventions faded and the general campaign began in earnest after Labor Day. However, the vice president’s margin dips to 48% to 43% when the sample is narrowed to the 1,495 registered voters most likely to cast ballots.

Gore’s September resurgence is bearing some resemblance to Vice President George Bush’s comeback 12 years ago. After trailing Michael Dukakis in pre-convention surveys, Bush took a 50%-44% lead after Labor Day and went on to defeat the Massachusetts governor in November. Like Bush Sr., Gore has rallied his base, while his opponent’s backing has faltered somewhat among certain key support groups.

In addition, as in the fall of 1988, there are signs that a renewed desire for continuity is favoring the incumbent vice president. Satisfaction with the state of the nation has increased since April and June, and it is more positively correlated with support for Gore than it had been earlier in the campaign. Voters also now have more confidence in the vice president than the governor to handle the economy, another expression of support for continuity over change that also occurred in the fall of 1988. And while Clinton fatigue is still evident, and may actually be growing as the president prepares to leave office, it is having a less negative effect on Gore’s campaign.

Despite the distinct signs of progress for the vice president, two important factors could cause the race to yet take another turn. First, many independents and other swing voters are still on the fence. This is unlike September 1988 when Bush Sr. held a 48%-42% lead over Dukakis among independents. Second, the percentage of voters who say they might change their minds is about as large as it was before the conventions.

These are the principal findings of the Pew Research Center survey, which finds Gore gaining among Democrats, senior citizens, African-Americans and lower-income voters. The poll, which employs the Pew’s voter typology, also shows Gore enjoying strong backing from Socially Conservative Democrats, who one year ago seemed inclined to defect to the GOP over the Clinton scandal.1 They now express as much support for Gore as do the other Democratic groups in the Center’s typology: New Democrats, Liberals and the Partisan Poor.

In contrast, while core Republican groups are backing Bush, they are doing so unevenly. As a consequence, the GOP’s wide partisan enthusiasm advantage over Gore that was apparent before the conventions has disappeared. Although the Texas governor now garners strong support from Staunch Conservatives, he gets less backing from Populist Republicans — who are not as affluent as other GOP groups — and somewhat less from Moderate Republicans. Demographically, Bush has lost his big lead among affluent voters, whites, men and college graduates.

Although Gore has come out of the summer with more momentum and increased backing from his core constituencies than Bush, independents are almost evenly divided between the two major candidates. Adding another note of uncertainty, nearly one-in-three independents who have committed to one of the candidates say they still might change their minds. Among the independents without an ideological leaning in the typology, Bush continues to get stronger backing from younger, affluent New Prosperity Independents and holds a smaller advantage among the less well-off Disaffecteds.

Younger women, older men, white Catholics, mainline Protestants and voters from non-union households continue to divide their support about equally between Gore and Bush, as they have for the most part since the end of the primaries.

Clearly, issues have fueled Gore’s gains. An increasing percentage of voters say that they have decided to vote for the candidates based on their positions on issues. Fully 45% of voters say their choice for president is based on his stand for issues compared to 36% who expressed this opinion in June. Since the summer Gore has shored up his dominant position on two of the top three issues — protecting Social Security and improving health care. Bush has made some progress since the conventions on the other top-tier issue, education, but a thin plurality still has more confidence in Gore on this issue. Moreover, in spite of the rising sense of satisfaction with the state of the nation, individuals’ personal financial anxiety is also growing, and Gore has an edge among voters with these concerns.

In part, Gore has caught the voters’ attention because many of the specific issues he has been addressing for the past several weeks, such as the addition of a prescription drug benefit to Medicare, have proven very popular with the electorate. Nine-in-ten voters (and 84% of Republicans) favor adding the drug benefit to Medicare. In addition, voters strongly prefer targeted tax breaks, like the ones Gore is promoting, to an across-the-board tax cut, which is the centerpiece of Bush’s economic plan. However, two issues being championed by Bush and the Republicans — eliminating the inheritance tax and allowing younger workers to invest some payroll taxes into private retirement accounts — also win wide backing.

Despite his deficit on issues, Bush is still regarded more highly than Gore on two crucial personal dimensions. More see the Texas governor as a strong leader, which is a key personal judgment made by voters, according to Pew’s analysis. Bush also continues to be seen as having more political courage than Gore. But the vice president’s personal standing has improved, along with his issue ratings, since the summer. A majority of voters (56%) say they like him more personally than they did earlier in the year; just 46% say the same about Bush, although the vice president’s fairly dismal ratings in this department earlier in the campaign left him more room for improvement.

The push and pull between Gore’s success with the issues and Bush’s advantage on the leadership dimension is seen in the way key groups evaluate the two candidates in these regards. For example, honesty and personal appeal are real advantages for the governor among older men, who strongly favor Gore on such issues as health care. Parents favor Gore on most issues, but see Bush as more honest and possessing better judgment in a crisis. Only half of Populist Republicans think the governor cares about people like them, and as many as one-in-four favor Gore for health care and Social Security and Medicare.

Other key findings of the survey include:

Unlike in June, voters are now paying as much attention to the campaign as they were four years ago, suggesting that turnout may at least match the 1996 level.

Fewer Republicans (49%) than Democrats (63%) think their party is doing a good job of standing up for its traditional positions. Populist Republicans are much less satisfied with their party than Social Conservatives on the Democratic side.

Most voters (58%) think Bush is more conservative than he admits, and an equal number think Gore is more liberal.

The vice presidential candidates are rated fairly evenly on a personal level, but Joe Lieberman has more bipartisan appeal than Dick Cheney.

A plurality of voters (42%) credit Gore with running the more positive campaign, compared to 31% who choose Bush.

These are the results of a Pew Research Center survey conducted among a nationwide sample of 2,799 adults (1,999 registered voters; 1,495 likely voters), 18 years of age or older, during the period August 24 ­ September 10, 2000. For results based on the total sample, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus 2 percentage points. For results based on registered voters, the sampling error is plus or minus 2.5 percentage points. For results based on likely voters, the sampling error is plus or minus 3 percentage points.