Comments on: January 2012 Poll Results

By: Poll Watch: Civitas (R) North Carolina 2012 Presidential Survey | Race 4 2012

Sat, 27 Oct 2012 11:00:57 +0000

[...] the poll conducted February 27-28, 2012 are in square brackets. Results from the poll conducted January 9-11, 2012 are in parentheses.Results from the poll conducted September 22-25, 2011 are in [...]

By: Poll Watch: 2012 General Election State Polling Recap | Race 4 2012

Poll Watch: 2012 General Election State Polling Recap | Race 4 2012 — Fri, 25 May 2012 03:37:04 +0000

[...] the poll conducted February, 2012 are in square brackets. Results from the poll conducted January 9-11, 2012 are in parentheses. Results from the poll conducted September 22-25, 2011 are in [...]

By: Teacher Tenure and NC Voters - Civitas Review Online

Teacher Tenure and NC Voters - Civitas Review Online — Mon, 30 Apr 2012 15:45:59 +0000

[...] January 2012 and March 2011, the Civitas Institute Monthly Poll questions on teacher tenure. Results seem to [...]

By: Analyzing the analysis on marriage amendment polls | The Locker Room

Analyzing the analysis on marriage amendment polls | The Locker Room — Thu, 29 Mar 2012 18:19:32 +0000

[...] Other polling has found support right around the 60 percent mark — WRAL-TV at 58 percent and Civitas at 62 percent. [...]

By: Mike P.

Mike P. — Mon, 20 Feb 2012 21:45:46 +0000

Lex, you can disagree with the poll all you like- and every poll PPP has done on the amendment since it was approved last year. If you reject polls that do not push your position but instead use the actual ballot language, that’s your decision. According to this poll -and, as I said, every poll PPP has done- even if all undecided voters opposed the amendment (which will not happen) it would still pass, comfortably. In any case, we can see the actual results in May.

By: Lex

Lex — Mon, 20 Feb 2012 18:07:59 +0000

Correction: A poll that DOESN’T CONTAIN BOTH is misleading to the etc.

By: Lex

Lex — Mon, 20 Feb 2012 18:07:11 +0000

Mike, it’s a matter of literal vs. contextual accuracy. A poll that is misleading to the respondent and a waste of time for the pollster and money for the organization commissioning the poll.

Maybe you didn’t know that. If so, now you do.

Maybe you did know that, and your goal here wasn’t so much to gauge public opinion as to gin up a “poll number” that might be used to bring the bandwagon effect to bear on leaning or undecided voters.

I neither know nor care which is the case in this particular instance. I just know that a significant portion of this poll, as written, is junk. And now you know that other people know it and are talking about it. I leave them to their own conclusions.

Cheers.

By: Mike P.

Mike P. — Fri, 17 Feb 2012 21:02:55 +0000

Lex,

I was talking specifically about the marriage amendment in terms of wording. Giving background information is useful for policy issues but not for measures that will appear to voters on the ballot with specific wording. What does it mean to ‘validly framed’? How it should be framed is a matter for debate. Why should pollsters take a side? Voters are not going see ‘by the way, this bans civil unions’ on the ballot. If that is what opponents of the amendment want to discuss, they can do that. Presumably, pollsters are not seeking to influence public opinion, but only to measure it.

I mentioned PPP’s affiliation because some other people have suggested that Civitas’ commissioned polls cannot be trusted because it is a conservative organization.

By: Lex

Lex — Fri, 17 Feb 2012 19:00:26 +0000

And I don’t care about PPP’s political affiliation. I’ve been a Republican for 34 years, which also isn’t relevant here.

By: Lex

Lex — Fri, 17 Feb 2012 18:59:39 +0000

Mike, legislators can live with their own consciences and those of their respective electorates. And the people who put ballots together must use the specific language approved by the legislators. But pollsters have an affirmative duty to refrain from asking misleading questions.

By withholding the effects of the action in this case (and in the gas-tax question), the pollsters are not framing the question in a way that is going to yield valid, internally consistent results. If the question were validly framed, then the percentages would more closely match the results you get when you ask people directly whether they favor or oppose civil unions.