HIGH POINT, N.C., Nov. 1, 2012 – Less than five days before the 2012 Presidential Election is held, the HPU Poll finds that slightly more national registered voters say they would vote for President Barack Obama than former Massachusetts Gov. Mitt Romney if the election were held today. An additional sample of North Carolina registered voters, however, narrowly favored Gov. Romney.
The poll results from registered voters across the country finds support for the candidates split as 46 percent said they would vote for President Barack Obama, while 43 percent said they would vote for Gov. Mitt Romney.
Within a special oversample of North Carolina registered voters, 46 percent of respondents said they had voted or would vote for Gov. Romney while 45 percent said they had voted or would vote for President Obama.
“The race for the White House continues to be extremely close here in North Carolina and across the country,” says Martin Kifer, director of the HPU Poll. “These results show why North Carolina continues to receive considerable attention from both sets of candidates.”
Presidential race – registered voters (who have and who have not yet voted)
Combined responses of these two questions for people who have voted early and those who have not yet voted: If the November 6th election for President of the United States were being held today would you be voting for the Democrat Barack Obama or the Republican Mitt Romney? AND If you already voted for president, for whom did you vote: Democrat Barack Obama or Republican Mitt Romney?
In the United States (50 states):
Democrat Barack Obama – 46 percent
Republican Mitt Romney – 43 percent
(Other) – 1 percent
(Don’t plan to vote) – 1 percent
(Don’t know/Refuse) – 10 percent
(For registered voters, n = 805, margin of sampling error is approximately 3.45 percent, percentages may not total 100 due to rounding)
In the North Carolina oversample:
Democrat Barack Obama – 45 percent
Republican Mitt Romney – 46 percent
(Don’t plan to vote) – 1 percent
(Don’t know/Refuse) – 8 percent
(n = 403, margin of sampling error is approximately 5 percent, percentages may not total 100 due to rounding)
Past HPU Poll Results in North Carolina have also shown a narrow divide between candidates during the last two months.
- Registered voters in NC, Sept. 29 to Oct. 10, n = 605: Obama 46 percent; Romney 45 percent; Other 2 percent; Don’t plan to vote 1 percent; Don’t know/refuse 6 percent
- Registered voters in NC, Sept. 8 to Sept. 13 and Sept. 15 to Sept. 18, 2012, n = 448: Obama 46 percent; Romney 43 percent; Other 3 percent; Don’t know 9 percent
- Registered voters in NC, Aug. 26 to Aug. 30, n = 543: Romney 46 percent; Obama 43 percent; Someone else 5 percent; Undecided 6 percent (With SurveyUSA)
- Registered voters in NC, Aug. 18 to Aug. 23, n = 540: Romney 43 percent; Obama 43 percent; Someone else 6 percent; Undecided 8 percent. (With SurveyUSA)
The survey was fielded by the Telephone Centre, Inc of Greensboro, N.C. from Oct. 22 to Oct. 30, 2012. The responses for the National Sample came from 805 registered voters with landline or cellular telephones. The North Carolina portion of the survey comprises an oversample of 403 registered voters. In total, the Telephone Centre conducted 1,208 interviews. All respondents are contacted through phone numbers attached to state Board of Elections registration data. The Survey Research Center contracted with Survey Sampling International and Aristotle International to acquire these registered voter samples. The national portion of the survey has a margin of sampling error of approximately 3.45 percentage points. The North Carolina portion of the survey has an estimated margin of sampling error of approximately 5 percentage points. The data are weighted when appropriate toward population estimates for age, gender and race. For the national sample, which was weighted toward population estimates for gender, the population estimates came from U.S. Census estimates of the registered voting population of the U.S. For North Carolina, the population estimates for race were taken from North Carolina Board of Elections data from the week of Oct. 27. The population estimates for age and gender were taken from the U.S. Census estimates of registered voters for North Carolina. In addition to sampling error, factors such as question wording and other methodological choices in conducting survey research can introduce additional error into the findings of opinion polls.
Dr. Martin Kifer, an assistant professor of political science, serves as the director of the HPU Poll, and Dr. Sadie Leder, assistant professor of psychology, serves as the associate director of the HPU Poll.
Further results and methodological details from the survey and can be found at the Survey Research Center website at http://src.highpoint.edu/.