Civitas Flash Poll: After RNC, Romney ahead

Support Civitas Polling with your tax-deductible donation
Donate »
To get direct polling results to your inbox, enter your email address.

Sept. 10, 2012
CONTACT: Francis De Luca or Jim Tynen (919) 834-2099  or

RALEIGH — In the wake of the Republican National Convention, a Civitas Institute Flash Poll found that Republican candidate Mitt Romney took a 10 percentage-point lead over President Obama.

The Flash Poll of 500 registered North Carolina voters was taken Sept. 4-6 and had a margin of error of plus-minus 4.5 percent. Asked if the election for President were held today who they would vote for, 53 percent chose Romney and 43 percent chose Obama.

Fifty percent of the voters had a favorable opinion of Romney, a former governor of Massachusetts; 40 percent had an unfavorable opinion. The survey showed that 50 percent of those polled had an unfavorable opinion of Obama, while 42 percent had a favorable one.

“Any one poll is just a snapshot. What is important are the overall trends,” said Civitas President Francis X. De Luca. “Our regular polling will show whether this Flash Poll is an outlier, or a harbinger of a new trend in voter sentiment.”

Upcoming Civitas Polls will be highlighted at Poll Presentations ( Sept. 27 in Raleigh and Sept. 28 in Charlotte.

Asked if their opinion of Romney changed after the convention, 39 percent of those responding to the Flash Poll said it got more favorable, and 33 percent said it grew more unfavorable. Those voters also said they had a favorable opinion of the GOP convention, 46 percent favorable to 37 percent unfavorable, and of the Republican Party in general, 45 percent to 40 percent.

Click here for crosstabs.

About the Poll: This poll was conducted by telephone in the voice of a professional announcer. Respondent households were selected at random, using Random Digit Dialed (RDD) sample provided by Survey Sampling, of Fairfield CT. All respondents heard the questions asked identically. Where necessary, respondents were weighted using the most recent US Census estimates for age, gender, ethnic origin and region, to align the sample to the population. Research methodology, questionnaire design and fieldwork for this survey were completed by SurveyUSA of Clifton, NJ. This statement conforms to the principles of disclosure of the National Council on Public Polls.

For more information on Civitas polling, see

More information on the Civitas Institute is available at, or contact Jim Tynen at (919) 834-2099.

About Jim Tynen

Communications director at Civitas.
This article was posted in Press Releases by Jim Tynen on September 10, 2012 at 11:48 AM.

© 2011 The Civitas Institute. Visit us on the web at
This article can be found at

Comments on this article

  • 1

    Total Sep 10, 2012 at 13:19

    30% of African-Americans in NC are going to vote Romney? In other news, space aliens will descend and anoint him President.

    Both are equally plausible.

  • 2

    Rick Sep 10, 2012 at 13:46

    Wow, and i though Rasmussen made up numbers…He’s credible compared to the insane clown possee over at Civitas…This is nothing but poll cooking, PERIOD….Etch a Sketch up by 10% in NC, winning 30% of the black vote…..There is no way etch a sketch will win NC by 10%, President Obama likley will win NC, and it will be close….

    That $3 bill i found at LAX was less of a phony than this poll from Civitas….

    Try doing REAL polling, not just making it up as you go along…

    You are worst pollster

  • 3

    merko Sep 10, 2012 at 14:24

    Tip for Civitas: If you want your lying to be believed, do it less brazenly.

  • 4

    Shawn Harmon
    Shawn Harmon Sep 10, 2012 at 14:24

    Political Axiom: When all the news is bad, make up good news.

    For the truth in polling see this non partisan –

  • 5

    F Bentley
    F Bentley Sep 10, 2012 at 15:20

    I also believe that the cow jumped over the moon, there is an Easter Bunny and Santa Claus is at the Betty Ford Clinic in rehab! BWHAHAHAHA

  • 6

    michael Sep 10, 2012 at 16:08

    In addition to the questionable crosstabs, I’d just like to note that this was right after the republican convention, which could have had a temporary bounce for Romney.

  • 7

    Francis De Luca
    Francis De Luca Sep 10, 2012 at 16:21

    Not to disappoint all of the haters out there but these are the numbers as reported to us by SurveyUSA. We could have “adjusted” them but than we would be doing what accuse of doing, cooking the numbers. If we were going to cook these numbers, they would not look like this.

  • 8

    Total Sep 10, 2012 at 16:36

    Not to disappoint all of the haters out there but these are the numbers as reported to us by SurveyUSA. We could have “adjusted” them but than we would be doing what accuse of doing, cooking the numbers. If we were going to cook these numbers, they would not look like this.

    I’m not sure that it’s a good defense to argue that “we lie better than this.” But, in any case, you should certainly get your money back from Survey USA, or at least highlight the oddity of a poll that gives the GOP candidate a higher percentage of the black vote than that candidate has gotten since the _1960_ Presidential election.

  • 9

    El Cuervo
    El Cuervo Sep 10, 2012 at 16:52

    I agree with Total. I would never accuse you guys of lying, but there’s really something fishy about these numbers.

  • 10

    Chris Sep 10, 2012 at 17:15

    It’s not just the black vote that is out of sync here. Mitt Romney gets 60% of the 18-34 year old vote vs. 39% for Obama? 55% of 35-49 year olds vote for Romney vs. 38% for Obama? I think we can safely disregard this poll.

  • 11

    sweetjazz3 Sep 10, 2012 at 17:16

    Given the small sample size, you are looking at a sample of approximately 100 African-American voters in North Carolina. Given that the sample is not truly random, there is some possibility for fraud (responders claiming a different race than their actual one), and the very small sample size, it does not seem that unlikely that this is just a statistical outlier. It’s also possible that some African-Americans responded to the RNC with a feeling that Obama hasn’t done enough for them and indicated this displeasure with their response. That said, it would be shocking if Romney got over 15% of the African-American vote, let alone 30% of it.

    In all likelihood, this poll will end up being an outlier. Right now, North Carolina seems very tight, with Romney perhaps a slight favorite. If Obama’s current 4-5 point lead in the polls lasts through November (instead of more likely being mostly a post-convention bounce), then NC will probably narrowly go his way. If Romney wins the election or loses by 1-2 points, he’ll probably take the Tar Heel state’s electoral votes.

  • 12

    Parelli Sep 10, 2012 at 17:30

    “The vision of the Civitas Institute is of a North Carolina whose citizens enjoy liberty and prosperity derived from ***limited government***, personal responsibility and civic engagement. The mission of the Civitas Institute is to facilitate the implementation of ***conservative policy solutions*** to improve the lives of all North Carolinians.”

    Need I say more???

  • 13

    Darrius Sep 10, 2012 at 17:30

    @ Total

    Actually, aliens descending and annointing Romney President is MORE plausible than the possibility of Romney getting 30% of the African-American vote.

    To my favorite movie, Dumbo….

    But I’ll be dun’ seen ’bout everything when I see an elephant fly.


    Mitt Romney get more than 5% of the Black vote.

  • 14

    rokkitman Sep 10, 2012 at 18:39

    Francis, do you not understand that the rest of us are already in on your secret code? We know that “all of the haters out there” is RWNJ code for anybody who doesn’t swill the Kool-Aid. I’m sorry to tell you, my friend, that your credibility was pretty low after you released the poll, but it fell to absolute zero when you gave up the code in public.

  • 15

    Zhonni Sep 10, 2012 at 19:17

    Democrats have been getting over 90% of the Black vote for a long time now. Romney is getting no more than 10% as history would suggest.

    McCain got 9% but I don’t think Romney would get that. If he does, it certainly wouldn’t be a whole lot more.

  • 16

    parelli Sep 10, 2012 at 20:21

    Looking at the data again, this poll is even more unbelievable in that it says that 45% of the respondents were Democrats and only %30 were Republicans.

  • 17

    Alex Sep 10, 2012 at 20:45

    “This poll was conducted by telephone in the voice of a professional announcer”

    Hahaha I bet once people heard a robo-call with a professional announcer voice they just started making stuff up. 30% Black support for Romney in N.C. would be historical.

  • 18

    alex Sanchez
    alex Sanchez Sep 10, 2012 at 20:56

    I was one of the people u called and Itold u Romney, but I will vote Obama

  • 19

    Joshua Chavers
    Joshua Chavers Sep 10, 2012 at 21:03

    I was one of the people u called and Itold u Obama, but I will vote Romney.

  • 20

    Jordan Sep 10, 2012 at 21:18

    Another big problem is who they sampled. Apparently they sampled some really screwed up people to get 30% Republican / 45% Democrat at the same time as getting 42% Conservative / 15% Liberal. The apparently don’t identify with parties that align with their personal beliefs/

    Or more likely, these numbers were just made up in attempt to make Romney look REALLY good in light of the recent developments. The left-leaning poll to come out recently has Obama ahead by 1%. I would completely disregard this one and never return to this heavily biased site again.

  • 21

    Woody Sep 10, 2012 at 21:58

    Saw this at RCPs daily poll listing, which was awash in blue today with numbers from +1 to +6, except for this single blaring red +10 from Civitas. Literally LOLed.

  • 22

    Prantha Sep 10, 2012 at 23:55

    One word: Hogwash.

  • 23

    Fia Sep 11, 2012 at 0:02

    This poll is just like 2008 look it had Mccain up 20

  • 24

    Zhonni Sep 11, 2012 at 2:14


    Wow! That was insightful.

  • 25

    Chris B
    Chris B Sep 11, 2012 at 6:27

    I came from RCP too, made me think again about RCP as a source of polling data, but gave me a chuckle.

    If you factor in how wrong they were about Mccain in 08 (thanks Fia), this poll implies Romney has 10% less support than Mccain did at this point in the campaign…and he lost NC! Even though it was the only red on RCP, this poll is potentially the worst news for the GOP today – stinks of desperation.

  • 26

    Harold Sep 11, 2012 at 7:31

    I guess the Willardbots are getting nervous. This sham poll is a huge overreach.

    The author of this Civitas piece should have stayed out of this comment section. Jumping into it only confirms that he and they are bought off hacks.

  • 27

    Salemst Sep 11, 2012 at 11:47

    I thought the poll was fine. Independents favor Romney 49-40%. That’s the key. Either the CBS/Washington Post or CNN poll showed Romney up 54-40% with Independents.

    See, you guys complaining should look at all the polls showing Obama ahead (besides Rasmussen and Gallup.)

    They vastly oversample Democrats by between 6-19% (Pew poll). One oversampled women by 13%.

    These polls are going by 2008 voter turnout which was an historic Democrat year with 7% more voters than Republicans. In 2004 and 2010 the vote was even. Since 1980 Democrats typically have 3% greater numbers at the polls. Plus, it’s all registered voters. When it’s likely voters the GOP picks up 2-3%.

    So complain away. But the polls showing Obama ahead poll too many Democrats on registered voter polls. So no question you’ll enjoy the results.

    Bottom line. We have a Depressionesque economy which will end Obama’s reign at one term.

    Despite overpolling Democrats 32-26% the CBS/Wash Post poll shows only 20% are better off today than 4 years ago, 32% worse off, 47% the same.

    So when asking “are you better off today than you were 4 years ago?” 20% say “yes.”

    Good luck Obama supporters.

  • 28

    Ryan Sep 11, 2012 at 12:16

    “I don’t often poll people about presidential candidates and preferences, but when I do, I lie brazenly about it by making up ridiculous numbers and then take to the comments section to try to lick my horrible wounds by claiming I could’ve lied even better.”

    Civitas: the most interestingly irrelevant poll in the world.

  • 29

    Total Sep 11, 2012 at 14:33

    @Salemst Yes, I know the latest Republican meme is that the polls are *all* wrong, that they’re oversampling, and that Romney is really ahead. First rule of thumb: if you’re complaining about the polls, then you’re losing. It reminds of John Kerry’s campaign in 2004 talking about how the polls were wrong.

    Second: shockingly, the pollsters have actually thought about it, and mostly ended up using a turnout/sampling model that is somewhere between 2008 and 2010 (and actually quite close to 2004).

    tl;dr: Nice try, but no.

  • 30

    Justadad Sep 11, 2012 at 15:02

    Seriously, did you guys think that anyone would believe a poll in which 30% of African Americans are going to vote for Mitt Romney? I was born at night, but not last night.

  • 31

    Salemst Sep 11, 2012 at 15:10

    @Total–you have to look at the demographics either way.
    If voting shows between an even and 3% Democrat edge in elections, you have to poll accordidngly and not take the outlier 2008 model.

    When I see a poll that’s approximately 53-47% women, within 3 points of Democrat and Republican poll sampling, and likely voters……THAT is a poll I pay attention to.

    Too many polls either use poor demographic polling sampling methodology, or are like PPP (SEIU pollster) deliberately trying to influence the news their desired way skewing Democrat.

    Until Gallup polls likely voters I’m ignoring them as well.
    Rasmussen has Obama up 3%, within the margin or error. They’re credible.

    This Civitas poll may have an outlier Black result. But Blacks are only 11% of the voting population so knock off 3 points and Romney’s up 7% in North Carolina

  • 32

    Salemst Sep 11, 2012 at 15:36

    @Justadad, yet you’ll believe a typical MSM poll with a 9% greater sampling of Democrats than Republicans claiming Obama’s ahead?

    ABC/Washington Post Poll shows Obama up 49-48% among likely voters. But they poll 32% Democrat and 26% Republican.

    Why are you buying that?

  • 33

    Geoff Sep 11, 2012 at 16:04

    Funny, one poll with a R bias and the left screams. Just start with the PPP and Democracy Corps and you have a big bias in the RCP average. At least those two admit their bias. Now we have one poll from Civitis that also admits it is biased and the progressives come unglued. Forget the fact the supposed mainstream media polls are massively oversampling Dems (Gallup, Abc/Wash Post, CNN etc.).

    Oh, which pollster was most accurate in 2008 when Obama won? Rasmussen. Yes, Rick, you are so biased you think the firm with the most accurate record for national races is making their numbers up. That is how far you are down the rabbit hole with the Dems. IF you look at the polling – not just the last poll – but polling in the month ahead of the election, Rasmussen was even more ahead. Rasmussen’s state by state polls have been much more variable (bad in 2010, great in 2008, above average in 2004 etc).

  • 34

    Salemst Sep 11, 2012 at 16:10

    Today’s Washington Post/ABC News poll has Romney leading by 11 points among independents.

    Last night’s CNN poll shows Romney up 14 among independents.

    You might ask yourself how in the world Romney could be losing (according to the corrupt media and their intentionally skewed polls) when he has a commanding lead with independent voters.

    Well, that’s simple, the polls are weighted with over-generous samples of Democrats. In fact, the CNN poll linked above shows Obama ahead by six points overall but CNN’s absurd sample only includes 5% of the very same independent voters Romney’s enjoying a commanding lead with. The Washington Post poll, however, samples 36% indies and that poll only shows a 1% lead for Obama.

    But both polls assume 2012 is going to be 2008 again where Democrats enjoyed a D+7 advantage in turnout. WaPo sample is D +6 and CNN is D +5.

    If anyone thinks that in this environment, Obama’s going to turn out that kind of advantage, you must be one of Obama’s media-shills.

    But this is why the media is pushing this Inevitability Narrative — because they know this is a turnout election and that, by proclaiming the race over, they hope to suppress Republican turnout.

    See how this works?

    The media lies and lies and shills and lies and lies and shills.

    I’ll take the Civitas Poll over theirs

  • 35

    Justadad Sep 11, 2012 at 16:22

    @Salemst, first of all, you don’t know what polls I am buying. All I said was I am not buying Civitas’ nonsense. (If your physics experiment suggests that gravity does not exist, you’ve done something wrong and need to look again. Issuing a press release just makes you look stupid.) Second, folks can argue about individual polls and methodologies until they are blue in the face (and campaigns that are behind almost always do). What matters at this stage are trends and considering polls as a group. President Obama is ahead. Things can always change, but he’s ahead right now. All of the whining about polls from Romney’s supporters really does remind me of the Kerry campaign in 2004. They claimed the polls were wrong until election day.

  • 36

    Salemst Sep 11, 2012 at 17:25

    @Justadad…..I don’t think Obama’s ahead. I believe it’s tied and Romney the last week–like Reagan in 1980–pulls away winning by 10-12 votes on the popular and gainst at least 345 Electoral

  • 37

    Total Sep 11, 2012 at 17:39

    And the power of magic thinking continues.

  • 38

    Dr. Butter
    Dr. Butter Sep 11, 2012 at 19:27

    Salemst: “But Blacks are only 11% of the voting population so knock off 3 points and Romney’s up 7% in North Carolina.”

    Get your facts straight — and your math.

    Blacks comprise about 22% of registered voters in North Carolina, give a generous 10% of them to Romeny and you knock off 4.4 points from Romney’s +10*, but don’t forget to add them to Obama. Net result: Romney +1 — of registered voters, prior to the DNC.

    See how this works?

    (This poll had Romney with 30% of 22% of the electorate, 6.6 of the total electorate. Give Romeny 10% of the 22% of the electorate that is black, and that’s 2.2, 4.4 points less than the 6.6 of this poll.)

  • 39

    Matt W
    Matt W Sep 11, 2012 at 19:38

    WOW! If this poll was conducted fairly, the results would fall about 4 standard deviations (8% in this case) away from what the the aggregate of polling suggests (civitas gives MOE of +/-4% @ 95% confidence). The chances of that happening given a normal distribution are .0000335 ([1-.999937]/2). Unless, something drastic was reflected in this poll that the others missed because of timing, the chances of this are 3 in 100,000. And, I seriously doubt it was the Republican convention, as polling only suggested a 1-2 point bump for Romney. That bump would make this poll slightly more plausible at 1 in every 5,000. Add in the even more unlikely cross tabs and you get the feeling that somebody screwed up intentionally or unintentionally.

  • 40

    Salemst Sep 11, 2012 at 19:51

    @Dr Butter, Blacks comprise around 11% of the voters, they are 13% of the population.

    I have no clue where your 22% comes from.

  • 41

    Total Sep 11, 2012 at 20:02

    @Dr Butter, Blacks comprise around 11% of the voters, they are 13% of the population.

    I have no clue where your 22% comes from.

    It comes from the fact that the poll you are defending was comprised of 21% African-Americans.

    It helps to defend something if, you know, you actually have some idea of what you’re defending.

  • 42

    Salemst Sep 11, 2012 at 20:56

    @Total–I was asking where the 22% figure came from.

    If Blacks were overpolled, then Mitt’s ahead in NC by more than 10% as 70% in the poll supported Romney.

    Maybe he’s actually up 12%

  • 43

    Matt W
    Matt W Sep 11, 2012 at 22:17

    Blacks were not over-polled. 22% of North Carolina’s population is black. 21% of NC’s registered voters are black. Exit polling and enthusiasm polls show a likely high black turnout as well. Its ridiculously easy to understand. “Dr Butter” is probably closer, even though he only considers the skewed results in the black representation.

  • 44

    Dr. Butter
    Dr. Butter Sep 11, 2012 at 22:35

    Salemst: “I have no clue where your 22% comes from.”

    The North Carolina Board of elections.

    Total Registered Voters: 6,418,370
    Registered black voters: 1,418,113


    Are you surprised that you were so uninformed?

    Get your facts straight before you start telling people what you think you know, it’s embarrassing.

  • 45

    Total Sep 12, 2012 at 8:07

    l–I was asking where the 22% figure came from.

    And thus revealing you didn’t understand the poll you were defending.

  • 46

    Dr. Butter
    Dr. Butter Sep 12, 2012 at 9:47

    Salemst, I posted a comment with links to to the North Carolina Board of Elections for your edification, but it has not yet cleared moderation. Here is the comment without the links:

    Salemst: “I have no clue where your 22% comes from.”

    The North Carolina Board of elections.

    Total Registered Voters: 6,418,370
    Registered black voters: 1,418,113


    Are you surprised that you were so uninformed?

    Get your facts straight before you start telling people what you think you know, it’s embarrassing.

  • 47

    Salemst Sep 12, 2012 at 10:38

    @ Total and Dr Butter

    So Blacks are 22% of the North Carolina registered voters, but what percentage of that 22% actually vote?

    My bet is since NC is a fairly conservative state it means Blacks may not vote monolithically for Obama/Democrat as they would where I live in Boston where 105% would vote Obama.

    Perhaps some NC Blacks are peeled away from Obama over:
    Homosexual marriage
    Private sector job seeking
    Entrepreneur/business owner
    School choice

    So it’s very possible instead of getting the obligatory 90% of the Black vote in North Carolina perhaps Obama gets 70-80%?

    So if it really bothers you, knock off 2-3% of Civitas’ Romney 10 lead and Romney’s actually ahead by 7%–as I said earlier.

  • 48

    Salemst Sep 13, 2012 at 16:11

    Guess the comments are over.
    Just want to thank the Civitas Poll for having the courage to publish poll results unfavorable to Obama–regardless of what your polls reflect in the future.

    Threats of broken knee caps and lawsuits from the White House must be unnerving.

    I follow polls, in that poll sampling is critical to me. If the sample is within the normal range I closely will look at the poll.

    My gripe is all these polls showing Obama ahead by 3-6%. Yet the over sample Democrats by 6-19% (Pew Poll at 19%). So if using a methodology of even to +3 Democrat, 52-48% women then Romney’s up 2-8% in all these national polls.

    It’s tough when the referees call them for the other team.

    Best of luck to you


  • 49

    tarheel Sep 23, 2012 at 11:03

    Agree with Total. Extremely odd how this poll has 30% of Blacks likely to vote for Romney..And absolutely NO Hispanics, Asians, etc. were polled – just whites & blacks?

8 Trackbacks on this article

Leave a Reply

Sorry, due to spammers you must have Javascript enabled to make comments.

Raleigh Web Design, WordPress & Web Development