Public Policy Polling

The fact that 65% of respondents 18-29 said they plan to vote for the amendment seems to fly in the face of the evidence that younger voters are more supportive of gay rights in general and gay marriage in particular. They even have a higher rate than those over 65. That seems highly suspect to me.

NC is not NY, gaylib. According to the CNN exit poll, 51% of voters 18-29 in FL voted for the amendment there in 2008- and NC is more conservative than FL. It is probably not 65%, but it could easily be 60% or 55%.

I live in NC, Mike. I know how conservative it is, and believe it or not, NC is not nearly as conservative as Florida. That number just doesn't make sense.

You also have the fact that the margin of error on young folks in the primary sample is AT LEAST 10.8%, probably higher. So it could theoretically be less than 55% in the actual population, and probably is.

I would also question your sample as far as what you expect as the makeup of voters for a primary which will only be for a Republican candidate (and one where the candidate is almost assuredly going to be a foregone conclusion at that point). Do you really think 47% of the voters will be Democrats? or that 20% will be African Americans? maybe in a general election (and I understand you were polling for Perdue/McCrory too), but in this case, with nothing else on the ballot, I highly doubt it.

I don't live in NC, so I do not doubt that you are more familiar with the state than I. All I will say is that by most measures -CNN exit polls, Gallup polls, 2008 election results, etc.- NC is more conservative than FL. Also, a much larger portion of the Democratic vote in NC is black as compared to the Democratic vote in Florida, and there is no reason to think that black voters in NC will vote differently on this issue than black voters elsewhere.
NC is also much more religious than FL. Nate Silver (among others) has demonstrated that importance of religion is the variable most closely correlated with support for a marriage amendment.

As for the youth vote, I doubt very much that many young folks will be showing up for the primary vote, and doubt even more that the election will come down to them. Young voters just do not vote as much as other groups. This was true even in 2008 and is truer in low-turnout elections. I will also say that young voters are not going to vote monolithically (i.e. 70%+) against the measure, which would seem to be necessary to defeat it.

Finally, it is no secret that support for marriage amendments has been chronically underestimated by polls by about seven points (see the work of Patrick Egan on that). By contrast, the level of opposition measured in polls is usually about right.

Gaylib,

If you think we have too many Democrats and that our young people are too conservative, maybe those two things will just balance each other out and the poll will be just right.

@Tom--didn't expect my rather pertinent questions to be met with such dismissive, nasty sarcasm. If your profesionalism is any indication of your polls trustworthiness, than I'd say it isn't worth the pixels on the screen. I won't even bother reading next time.

Gaylib, the demographic information was for the general election numbers. But even still, at this point we do have more people saying they'll vote in the Democratic primary next May than the GOP one.

Mike P: Our polling on the Maine referendum in 2009 was deadly accurate, and we generally are finding less support for gay marriage than other pollsters, due to the lowered social desirability bias of respondents feeling their answers are more anonymous when given to a recording than a live person.

On the 18-29 cohort, the issue isn't the absolute level of support or opposition to SSM; it is the support level relative to the other age groups. If the 18-29 cohort comes in as more anti-SSM than the older groups that would be a red flag, as it would fly in the face of every other poll ever taken on this issue. Even the 2008 CNN exit poll from FL mentioned above bears this out, as the younger cohort while supporting the amendment, had support running 12% below FL voters as a whole.

So whether the poll shows the 18-29 group in NC supporting the amendment by 45, 55, or 65%, none of these results would be an indicator of inaccuracy. But if the poll shows the 18-29 group supporting it at a level greater than the 50+ group(s), that may suggest that something is screwy with the poll.

Finally, I have a hard time believing that the turnout in NC in May 2012 is going to resemble the general election. That having been said, PPP is very good at polling on this issue and I hesitate to dismiss this poll on that basis.

Has PPP ever polled on an initiative or referendum that was slated to appear on the ballot on a state primary day?

A lot of factors will have to be accounted for over time, such as if the Republicans have a presidential nominee before the NC primaries, because if that is the case then i think these numbers will be fair, but if NC is a game changer for the GOPs then opponents of the amendment have little to no chance of generating the amount of turnout needed to dismiss this awful amendment.