The Emperor’s New Polls – What Are Words For?

Yesterday, we looked at a piece by Joe Loveland in the Twin Cities leftyblog Wry Wing Politics.  WWP rates my ultimate endorsement for a leftyblog – its author isn’t in line for a harassment restraining order and remedial logic class.

That’s all to the good.

But he cited a series of media polls about the public’s opinion on Universal Background Checks, listed off the top-line numbers…:

• CNN/ORC (89% support background checks)
• Quinnipiac (91% support background checks)
• Morning Joe/Marist (87% support background checks)
• CBS (90% support background checks)
• Fox News (85% support background checks)
• ABC/Washington Post (90% support background checks)
• Pew/USA Today (83% support background checks)
• University of Connecticut (69% support background checks)
• Gallup (91% support background checks)
• Associated Press-GfK (84% support background checks)

…as dispositive evidence that the public overwhelmingly favors universal background checks.

And unlike many leftybloggers, Loveland knows that polls aren’t, themselves, iron-clad.  He takes a whack at a pre-emptive defense of the results:

For those who quibble about question wording, these polls all asked the question a bit differently.

“A bit differently”.

Let’s take a look at the questions asked in the individual polls that Loveland cited:

• CNN/ORC (89%) – The 89% response came from asking if the respondent supports background checks “If the buyer is trying to purchase a gun from a gun store or other business that sells guns”.  Other questions came in much lower.
• Quinnipiac (91%) – “Do you support or oppose requiring background checks for all gun buyers?”
• Morning Joe/Marist (87%) – “Do you support or oppose legislation that would require background checks for private gun sales and sales at gun shows?”
• CBS (90%) – “Do you favor or oppose a federal law requiring background checks on all potential gun buyers?”
• Fox News (85%) – There were several questions; the one getting 85% was whether the respondent favored “Requiring criminal background checks on all gun buyers, including those buying at gun shows and private sales”
• ABC/Washington Post (90%) – “Would you support or oppose a law requiring background checks on people buying guns at gun shows?”
• Pew/USA Today (83%) – Asked if the respondent supported “making private gun sales and sales at gun shows subject to background checks”.
• University of Connecticut (69%) – asked if the responded supports “a law which would require background checks before people – including gun dealers – could buy guns at gun shows”, which is, by the way, borderline-incoherent.
• Gallup (91%) – “Require criminal background checks for all gun sales”
• Associated Press-GfK (84%) – Asked if US should  “Establish a federal standard requiring background checks for those trying to buy guns at a gun show”
• Strib “Minnesota Poll”/Mason Dixon (72%) – “Do you support or oppose a universal background check on all gun sales, including those sold at gun shows?”

Loveland’s right.  They’re a “bit” different, all right.

And yet they are all exactly the same.  The questions – all of them – float a high-level proposal  (“Should we have background checks at gun shows?”) with no further context.

Like the pro-gun-control push poll from the 1980s I cited yesterday, which showed 85% of the people “supported gun control”, it was a hopelessly broad question – a gauzy proposal that gave the less-informed respondent no context that would help them actually understand the issue beyond the top-level sound bites – far from enough information to give an informed answer.

Valid questions on the subject, questions that provided the context needed for informed answers on the subject, would go something like this:

• “Do you support universal background checks at gun shows knowing that the checks create a paper trail leading to every gun and gun buyer in the country – which is de-facto registration?”
• “Do you support universal background checks, even though criminals don’t subject themselves to background checks of any sort?”
• “Do you support universal background checks, even though violent gun crime has dropped by over 40% in the past 20 years, and the drop has accelerated over the past five years, after the sale of 70 million firearms in the US?”
• “Do you support universal background checks, knowing that in California similar legislation has added about $100 to the cost of every firearm, pricing poor people out of the market for guns to defend themselves, their families and their homes?”

Yeah, I know – there’s bias in the wording of my questions; I’d be happy to work with a poll writer on the actual verbiage. But sometimes you need a biased question to lead you to the truth, and sometimes an “unbiased” question, like the polling questions, are biased by omission (and the uncritical reporting on them is biased by commission).

At any rate, I’m going to hazard an informed guess here; if people know the real-life consequences of “universal background checks” (they oppress the law-abiding, hamper the poor and are useless in preventing violent crime), the results might just drop below “landslide” levels.

Because the fact is, people know this already.  Second Amendment rights have expanded over the past twenty years, and violent crime has plummeted.   Obama’s gun-control push has largely fizzled; “background checks”, useless as they are, are about all that’s left, outside the gun-grabber liberal havens on the coasts.

Anyway – back to Minnesota, and Loveland’s assertions:

For those who argue methodology, these polls all reached a different randomized sample of respondents, and relied on different methodologies.

Perhaps they did; the geographic, demographic and ethnographic details weren’t included in any of the links Loveland provided.   

For those who worry about sponsorship bias, these polls were sponsored by a wide variety of news outlets and academic institutions.

And yet the questions they uncritically asked were all nearly exactly identical.

For those who stress that polls are blunt instruments, these polls did not find slim margins that conceivably could be slightly off.

That’s correct.  They found overwhelming support for a hopelessly broad question that, by its nature, filtered all possible context from the results that were reported.  

Of course, this isn’t just about polling “science” to Loveland.  The mission is to try to undercut the Minnesota GOP, which has been gratifyingly solid on Second Amendment issues this session.  (In-line thought bubble: Where the hell was that sense of purpose last session on the freaking stadium? Or in 2011 on the budget negotiations? Hello?)

Make no mistake, on the issue of universal gun background checks, Minnesota Republicans are choosing to represent the will of NRA lobbyists over the will of the overwhelming majority of Minnesotans, including gun owners, Republicans, Independents and Greater Minnesota citizens.

“NRA Lobbyists” are the great lefty boogeymen.  The heavy lifting on gun issues this session, as between 1995-2005 in passing Concealed Carry, has been GOCRA – the most successful grass roots group in Minnesota politics, and the grass-rootsiest successful group in Minnesota politics to boot.  It has no paid lobbyists.  It has no paid anything.  It’s an email mailing list and over 20,000 Minnesotans who write letters and make phone calls.

Legislators report that phone calls and emails against the DFL’s gun grab legislation run about between 50-100:1 against the gun grab supporters.

And Minnesotans are voting with their feet; 135,000 Minnesotans now have their carry permits, and at current pace there will be 200,000 within a year.

None of those figures gives you any more context than the questions in any of the polling Loveland cites.  Or, to be fair, any less.

Frankly, Minnesotans, Republicans just aren’t just not that into you.

And given that the “you” really doesn’t exist outside of a push-poll wording card trick, they’ve got the right idea.