The Star Tribune’s Minnesota Poll has come out with a poll on marriage equality today that in order to be right, well, let’s say every other poll, and the last election results, need to be wrong. They claim 38% support legalizing same-sex marriage, 53% oppose, and 9% don’t know. This is despite a majority rejecting the marriage ban amendment in the last election, and other polls consistently showing support rising with pluralities or majorities ready to make the law the same for everyone.
The same poll found 70% support background checks on gun purchases even though national polls show support around 90%, and a small majority supports upper income tax increases despite other polls showing much higher support.
The poll was conducted by Mason-Dixon, they same pollster the Star Tribune used for their pre-election polls. That’s the same outfit that said at the end of October that Romney trailed Obama by just three, while other polls had him behind by high single digits. Obama won Minnesota by 8. Mason-Dixon’s record for the rest of the country was, well, similar.
Is this poll denialism from the left, like conservatives engaged in before the presidential election? One small difference. Before the election, conservatives refused to believe the bad news from the consensus of polls, relying instead on “unskewing” or their guts or Romney’s internal pollster with its outlying result. We, however, are refusing to believe the outlier, and accepting the consensus of polls. That’s easier, granted, when you like the results, but still, one side wants to believe the outlier and one accepts the consensus. So no, not the same.
Let me put it this way: there are several possibilities. Maybe there has been some significant shift on all three issues in a very short time. Granted that’s true with gun issues, since the Sandy Hook massacre has indeed caused a sudden shift, but that’s been in favor of sane gun regulations.
Or maybe Minnesota is significantly more conservative than the national average. If so, then this run of election wins by Democrats running on liberal platforms in a high turnout state gets hard to explain. OK, I phrased it that way just so the voter fraud believers get a chance to scream at the their computers.
Or maybe Mason-Dixon has a distinct rightward house effect. Given their record, that seems pretty likely.
Someone I mentioned today’s poll to called it bad journalism. It’s not bad journalism, just bad polling, though there is one criticism to be made. The articles didn’t say the poll results contradicted other polls. They had a public figure mention this, Speaker Paul Thissen in the case of the marriage poll, but they didn’t say that the public figure was factually correct. This is that false objectivity that comes from treating each side as if it’s equally likely to be right. Reporting on your poll’s results is just fine, but whether it comes in with similar results as other polls is just a fact, and a relevant one. Especially given their pollster’s record, the Star Tribune needs to say this.
Just like is suggested by that study posted yesterday about how elected officials assume voters are more conservative than they actually are, elected officials who accept these results are in danger of putting themselves to the right of their constituents. That’s fine for Republicans worried about winning the party nomination from a base that skews far right, but that’s a bad risk for Democrats. They risk not just voting out of step with their constituents, but also demotivating their voters and alienating the people who show up at their phonebanks.