Civitas Poll: McCrory Support Shifts In Key Voting Bloc | Eastern NC Now

    Publisher's note: This post, by Jim Tynen, was originally published in the Legislative Activity section of Civitas's online edition.

    A new Civitas Poll of registered, unaffiliated North Carolina voters shows Gov. Pat McCrory trailing Attorney General Roy Cooper in a hypothetical gubernatorial race.

    Asked whom they would vote for if the election were held today, 32 percent of voters would back Republican Gov. Pat McCrory, 34 percent would cast their ballots for Democrat Cooper, and 33 percent were undecided or were leaning toward one of the candidates.

    "This number is interesting as it compares with the 44 percent of unaffiliated voters who said in a June Civitas Poll that they would support McCrory. Although the June poll included a smaller sampling of unaffiliated voters, such a shift may be a concern for McCrory's advisors," Civitas President Francis De Luca said.

    The poll surveyed 400 registered North Carolina voters, 30 percent of whom were reached on cell phones. The survey was taken August 10-12 and had a margin of error of plus/minus 4.9 percent.

    Following is the exact wording of the relevant questions:

32%	Total McCrory
34%	Total Cooper
33%	Total Lean/ Undecided
21%	Definitely McCrory
12%	Probably McCrory
6%	Lean McCrory
20%	Undecided
7%	Lean Cooper
13%	Probably Cooper
20%	Definitely Cooper
1%	Refused

FAVORABLE	UNFAVORABLE	HEARD OF NO OPINION	NEVER HEARD OF
Barack Obama
46%	44%	9%	-
Pat McCrory
36%	43%	19%	2%
Richard Burr
25%	28%	33%	14%
Roy Cooper
26%	14%	29%	32%
The North Carolina Democratic Party
34%	39%	25%	2%
The North Carolina Republican Party
31%	45%	22%	2%

50%	Total Approve
48%	Total Disapprove
20%	Strongly Approve
29%	Somewhat Approve
11%	Somewhat Disapprove
37%	Strongly Disapprove
2%	Undecided/Don’t Know
-	Refused

    Totals may not equal 100 percent, because of rounding.

    Crosstabs for poll here.

    About the poll: This poll of 400 registered voters in North Carolina was conducted August 10-12, 2015 by National Research, Inc., of Holmdel, NJ. All respondents were part of a fully representative sample of registered general election voters in North Carolina. Thirty percent of the respondents were cell phone-only users. For purposes of this study, voters interviewed had to have voted in at least one of the past two general elections (2012, 2014) or be newly registered to vote since November 1, 2014. The confidence interval associated with a sample of this size is such that: 95 percent of the time, results from 600 interviews (registered voters) will be within +-4.9% of the "True Values."