Survey Results Predicted Narrow Tillis Victory
Nov. 12, 2014
FOR IMMEDIATE RELEASE
CONTACT: Francis De Luca (919) 834-2099 francis.deluca@nccivitas.org
“If turnout is high, it will be because Hagan and her allies turned out less-frequent voters. If turnout is low, it will resemble a typical midterm turnout and will benefit Tillis. Right now, evidence points to an election that more typically resembles a midterm, though this can change over the next two weeks. But if the race were held today, Tillis would eke out a win against Hagan, 49%-47%, with the balance going to Haugh.”
So read the concluding paragraph of the polling memo that the Civitas Institute received from our pollster, Adam Geller, of National Research Inc., on October 22, a full 13 days prior to Election Day.
A review of our monthly surveys over much of the year shows that we always expected the race to be close, and each of the surveys’ top-lines contained a spread between the two candidates that were well within the margin of error of the actual results.
But the real value in analyzing polling results goes far beyond merely reading top-line polling results, according to Civitas President Francis De Luca. “To simply read the top-lines and draw conclusions based on that misses the point entirely,” he said.
“For example, survey results contain undecided voters. But these voters are either going to decide on one candidate or they are simply going to stay home. The value of a good poll, and a good pollster, is to help figure out how those undecided voters are going to break, if they decide to vote at all.
“We challenged Adam and his team at National Research to tell us what they thought would happen, based on the crosstabs, and they once again came through for us,” De Luca said. “While they always poll ‘likely voters’, they created different turnout models, some of which were based on Democrat surge voters coming out, and some of which were not. But beyond that, they told us which one they expected on Election Day. And they were right.”
The 2014 survey continues a streak of accuracy for the Institute. In 2012, the Civitas poll was one of the few polls that correctly predicted both the presidential and the gubernatorial results. “The Civitas Institute Poll is the authority on public opinion in North Carolina because we go the extra step in reading the details of the poll and they we apply our in-state acumen for added value,” De Luca said.
Civitas conducts the only regular live-caller polling of North Carolina voters. For more information on Civitas polling, see http://www.nccivitas.org/category/poll/.
The Civitas Institute – “North Carolina’s Conservative Voice” – is a policy institute based in Raleigh, N.C. More information is available at www.nccivitas.org, or contact Jim Tynen at (919) 834-2099 or james.tynen@nccivitas.org.
Leave a Comment