Posted October 28, 2008

The Temple Poll: Obama leads in Pennsylvania

Will enthusiasm balance experience among voters on Election Day?

Senator Barack Obama now leads Senator John McCain by 50 percent to 41 percent among Pennsylvanians likely to vote in the Nov. 4 presidential election. Just 8 percent remain undecided, and half of that group prefers one candidate or the other. Adding in the undecided voters who lean to a candidate puts Obama’s lead at 52 percent to 43 percent.

 

These figures come from the most recent Temple Poll, sponsored by Temple’s Institute for Public Affairs, for which interviews were conducted between October 20 and 26 with 1,001 Pennsylvanians registered to vote, 761 of them likely voters.

The Poll reveals the state of the contest among key demographic groups of likely Pennsylvania voters.

  • Obama leads among moderates (62 percent to 34 percent), women (56/37), Catholics (53/41), those in middle-income families (55/42), and residents of the Southeast outside of Philadelphia (59/39).
  • McCain leads among men (51/47), Protestants (52/44), and residents of the central portion of the state (57/34).
  • The contest is effectively tied among Independents, voters in families with incomes of $75,000 or more, and residents of the Southwest and the Northeast.
 

As the race reaches the home stretch, the campaigns allocate more of their resources to making sure their supporters reach the polls on Election Day. The Poll finds that the McCain campaign enjoys a distinct advantage with regard to turnout: McCain’s supporters have had more electoral experience than Obama’s. By 65 percent to 51 percent, for example, McCain voters are more likely than Obama voters to say they follow government and public affairs “most of the time, whether there’s an election going on or not.” McCain voters are also more likely to say they always or nearly always vote and they have voted in a presidential or congressional election before. In part, these differences reflect the fact that McCain voters are older and wealthier, on average, than Obama supporters. The differences are important because engagement with politics and experience with elections facilitate voting.


“One of the questions that remains about this presidential election is whether Senator Obama’s supporters will turn out on Election Day in the numbers necessary for Obama to win,” said Michael Hagen, associate professor of political science and director of the Temple Poll.


Despite their inexperience, Obama voters are as likely as McCain voters to say they definitely will vote, the Poll showed. If they do vote in equal numbers, one reason will be the level of enthusiasm among Obama voters about their candidate. Asked to express an opinion

about the candidates on a scale from 0 to 10, 45 percent of Obama voters rate their candidate a 10; just 29 percent of McCain voters rate their candidate as high. The relative coolness of McCain voters to their candidate is not counterbalanced by hostility to their opponent; McCain voters and Obama voters feel about equally negative, on average, toward the opposing candidate.


“In terms of the motivations of individual voters,” Hagen said, “the Obama campaign has to hope that their candidate’s personal appeal will be as effective in getting voters to the polls as the habits and experience of McCain voters will be. Of course, both organizations have committed resources of their own to getting out the vote, and the Democrats will benefit from having more to spend on GOTV [Get Out the Vote] than the Republicans.”


For this Poll, a randomly selected sample of 1,001 Pennsylvanians registered to vote were interviewed by telephone between October 20 and 26, 2008. Once the interviewing was complete, the data were weighted to correct for unequal probabilities of selection and response, and to ensure that the demographic characteristics of the sample match the population of registered voters in Pennsylvania. The sample included 761 likely voters, identified on the basis of their voting history and engagement with the campaign. With samples of these sizes, the overall margin of error attributable to sampling is 3.1 percentage points for the sample of registered voters, 3.6 percentage points for the sample of likely voters. The sampling error for subgroups is larger.


Additional results from the Poll will be released over the next several days.
webcomm