Skip to Content

Support MinnPost

MinnPost poll: Minnesotans blame GOP for shutdown

Forty-two percent of poll respondents blame Republicans in the Legislature for the shutdown.
REUTERS/Eric Miller
Forty-two percent of poll respondents blame Republicans in the Legislature for the shutdown.

By a whopping 2-1 margin, Minnesotans blame the Republicans who control both houses of the Legislature for the recent government shutdown more than they blame Gov. Mark Dayton, according to a poll taken this week for MinnPost.

Predictably, most Republicans blamed Dayton more (by 56 to 10 percent, with the rest saying both sides were to blame or holding no opinion). DFLers blamed the Republicans by an even more overwhelming majority (68 percent to just 2 percent of DFLers who blamed DFLer Dayton).


But the key swing group of self-identified independents was also much more likely to blame Republicans than to blame Dayton. Among independents, 46 percent "blamed" the Republicans, 18 percent blamed Dayton and 25 percent both.

Based on other questions in the poll, it was difficult to say whether the fallout from the shutdown will give DFLers a significant advantage heading into the 2012 elections, as Republicans seek to retain their majorities. Projecting current attitudes onto an election 16 months in the future would be folly.

Also, this poll, conducted for MinnPost by Daves & Associates Research, was designed to take the pulse of the state in the aftermath of the shutdown, not to predict the next election. No likely voter screen was used and sample surely includes non-voters.

But if Republicans stick to their ironclad no-new-taxes stance, they may find themselves out of sync with most Minnesotans other than the hard-core Republican base.

Source: Daves & Associates Research

Unless there is an economic miracle, and right quick, Minnesota will head into the next budget session needing to decide again whether to make big, painful spending cuts, impose a significant tax hike or some combination thereof to balance the 2013-14 budget. The MinnPost poll asked which of those approaches respondents would favor.

Only 23 percent of respondents said that they favor a cuts-only approach. A strong majority — 66 percent — say a combination of spending cuts and tax increases is the way to go. Another 5 percent would prefer raising taxes enough that no spending cuts would be necessary.

Even among Republicans in the sample, 50 percent preferred a combination of spending cuts and tax increases, compared with 42 percent of Republicans who agreed with the Republican no-new-taxes position.

Among DFLers, 11 percent endorsed an all-cuts approach, another 11 percent favored an all-tax-hikes approach and 74 percent favored a combination.

And the key swing group — Minnesotans who do not identify strongly with either major party — were closer to the DFL views. Among independents, it was 22 percent for no new taxes, 2 percent for no spending cuts and 72 percent for a combination of cuts and tax increases.

Source: Daves & Associates Research

The overall sample, by the way, consisted of 28 percent self-identified Democrats, 21 percent Republicans, with by far the largest portion (a combined 51 percent) calling themselves independents, identifying with a smaller party, or expressing no description of their partisan orientation. Somewhat similarly, the full sample was composed of 32 percent who considered themselves conservatives, just 18 percent who self-identified as liberals, with the biggest group (50 percent) calling themselves moderates or offering no opinion of their ideology.

Projecting to 2012?
As far as the business above on the difficulty of inferring partisan advantage from these results, here's what I meant: The general attitude toward the shutdown, whose fault it was and how the budget impasse should have been resolved seems to favor the DFL. If those views were to hold up and the 2012 turned into a referendum on who was to blame for the shutdown or who has the best approach to avoiding another shutdown, it would seem to be promising for DFLers hopes of capturing control of one or both houses of the Legislature. (All members of both houses will be up for election in 2012, and Dayton will not.)

The MinnPost poll asked Minnesotans whether the shutdown will make them more likely to vote for Republicans or for Democrats in 2012, or whether the shutdown will not affect their vote either way. The results:

More likely to vote for the Repubs: 17 percent

More likely to vote for the Dems: 30 percent.

Won't affect my vote: 42 percent.

No opinion: 11 percent.

Source: Daves & Associates Research

I guess that also looks promising for the Dems. But many of the respondents who said they were "more likely" to vote for the Dems voted for them in 2010. (Same for the Repubs.)

To explore more deeply whether feelings about the shutdown would actually cause Minnesotans who recently voted for one party to switch to the other party, pollster Rob Daves cross-tabulated the responses against respondents' recollection of whom they had voted for in 2010. Among those who recalled voting for a Republican in 2010, only 6 percent said that the feelings about the shutdown made them more likely to vote for a DFLer in 2010. And of those who voted for a Democrat, only 4 percent said they were more likely to vote for a Republican. There are reasons to view these number cautiously, since the sample included many non-voters and probably some who didn't reliably recall their previous vote — it's been nearly a year. But at face value, the cross-tabulation suggests that a relatively small number of Minnesotans expect to switch from red to blue or from blue to red in 2012.

Take our poll on the government shutdown

Do you have opinions on the recent state government shutdown? If so, MinnPost would like to measure them.

On the other hand, among those who described themselves as independents, 13 percent said the shutdown made them more likely to vote Republican in 2012, while 29 percent said it made them more likely to vote Democratic. (Half of independents said the shutdown would have no effect on their likely 2012 vote.) If looking at the views of independents is the best method of inferring future political momentum from current feelings about the shutdown, that analysis looks more promising for DFL legislative candidates.

Larry Jacobs' analysis
Humphrey Center Political Scientist Larry Jacobs said the results of the new poll were "basically bad news for the Republicans."

"They have to think about this fact," said Jacobs."The principles that they ran on in 2010 — that they would advocate for cuts only and would refuse to go along with any tax increase — may still be the principles that appeal to the most enthusiastic base of support they have. But that position seems to be pretty unpopular not only with two-thirds of Minnesotans, but with half of their own party, all of whom prefer a mix of significant spending cuts and at least some tax increases."

Still, for all the hype about the shutdown, almost half of the state's adults (48 percent) said the shutdown affect their own lives "not much at all" and another third (34 percent) said it affected them "only a little." That leaves just 18 percent who said the shutdown affected them "a lot." Republicans (58 percent) were significantly more likely than Democrats (39 percent) to say that the shutdown had not affected them.

Source: Daves & Associates Research

House Speaker Kurt Zellers this morning explained the poll results about the Legislature this way: "It's a little like what you see on the national level. Everybody hates Congress but loves their congressman. It's just easier to blame that amoeba-mass than a single person. I think if you go to the individual members in their districts, the results would be different. The individual members would be more popular than the mass."

He says the governor, by contrast, has "the power of the bully pulpit. Every time he speaks, he blames us. He talks about 'the Republican plan,' even though he signed it."

The House speaker also sees the public's view seeking a combination of spending cuts and tax increases: "I understand why people who do polls ask the question. But it's so easy to say, 'Sure, increase taxes and make cuts.' But the first thing I learned when I was on the tax committee was that old mantra: 'Don't tax you, don't tax me, tax the guy behind the tree.' A lot of people don't even know a millionaire, so why not tax them? I used to believe that, but then you keep talking to people — people who worked for small businesses that have been closed, people who run small businesses. They can't take more taxes."

And Zellers notes: "When you get specific about other taxes, you get a different answer. Talk to people about sales taxes. "Ask somebody, 'Should we expand the sales tax to clothes?' They'll say, 'Oh no, no, no.' So you say, 'Maybe we should expand the tax to include services, like your accountant or the mechanic. 'Oh, God no, not that.' "

In terms of the lasting impact of the shutdown, Zellers says: "I think attitudes change pretty quickly. People want to get back to summer. Nobody likes to see a car crash on the highway for very long."

Gov. Mark Dayton, reacting this morning to the poll's "blame" findings, said: "It's a positive sign to me that people overwhelmingly, 3 to 1, are supporting my position and, 2 to 1, saying it was the Legislature responsible for the shutdown. I knew going in that this would be damaging both to the legislative leaders and to me. But the $1.4 [billion] in revenue for essential services [that he won during the shutdown] are needed, and they're worth any of the [political] damage."

In terms of the long-term impact of the shutdown, the governor said, "People are changing their minds as events unfold. It will take awhile not to look at the shutdown with great dismay."

He expressed surprise at the high number (48 percent) who saw little impact from the shutdown. "I would guess that the 18 percent [affected a lot] are public employees, or private-sector employees who were laid off because of the shutdown," Dayton said.

The MinnPost-sponsored survey of 598 Minnesota adults was taken July 24-26 by Daves & Associates Research. The margin of sampling error was plus or minus 4.8 percentage points.

Doug Grow contributed to this report.

FRIDAY: How is Gov. Mark Dayton handling his job as governor?

Related: Take our poll on the government shutdown
Inside MinnPost: MinnPost's first poll and how you can help

Get MinnPost's top stories in your inbox

Related Tags:

Comments (58)

Wow. 74% of independents prefer a solution to the inevitable 2013 budget deficit that includes taxes. The GOP is waaaay outside the mainstream.

We'll see if voters have any more of a memory on this than they typically have.

SQUIRREL!

Wait, what was I saying? All tax cuts!? Yeah, that must have been what I was saying...can't really recall...

Actually, this poll is encouraging. This means that those polled were able to look at a somewhat complex situation and come up with the correct answer.

It's nice to have some actual data, but this really just confirms what a lot of people have been saying all along. I think the Democrats have an edge. Remember, the Republicans have another year and a half to dig their hole even deeper with their whacky social agenda, that's all they have left. What remains to be seen is whether or not the Democrats can exploit this advantage or blow it like they typically do.

The state poll reflects national polls--people want a balance of cuts and tax increases. Even a recent national Rasmussen poll (Rasmussen polls are the most often-cited polls by Tea Partys) shows that a majority (56%) would support a congressional candidate with a balanced approach and a minority (34%) would support a cuts-only candidate.

TP reflects "the will of the people"? Or does TP stand for "Tools of the Plutocrats"?

It should be clear to all that the Republicans won by a relatively small majority and took their election as a mandate to enforce the will of the 20 or 30% who want a cuts-only budget. That position is clearly counter to "the will of the people".

This isn't actual data. This is just a poll someone purchased, which would not exist if it hadn't been commissioned.

What people think now is of no concern. What matters is what people think a year from now.

Tony Sutton: " ...we're not going to get rolled this time." Well, Tony, you did it to yourselves and you earned and deserve all of it.

@#6
What's wrong with publishing this data? Unlike most other news sources, they openly stated that there are big caveats to the data--likely voters were not identified, a large percentage identify as 'independent,' but weren't asked what they were *registered* as, those who said they were more likely to vote one way or another voted the same way in the last election, etc.

Yes, this poll was paid for by MinnPost. That does not make it worthless, it just means that we have to consider what the data *means*, not what it says. The data is actually not surprising when national polls say almost the same thing regarding how the budget should be taken care of.

By the way, ALL polls are flawed, yet we all use them as a keystone for debate. It's only a legitimate keystone if we keep in mind the limitations of the data and don't over-interpret.

this poll was paid for by MinnPost. That does not make it worthless,

It helps make it worthless. The pollsters have a natural interest in telling those who pay for the data what they want to hear. This is true across the board. That's why news media polls are never reliable, and never worth the money that news media pay for them.

The other problem is that what people think now simply doesn't matter. What matters is what people will be thinking next fall, during the election season.

@Ralf at the #1 spot:

Perfect summation!! So sad, but true--- and I'm stealing it for future use!

@#11
The data shown in this article is NOT based on the online poll, which is probably mostly regular MinnPost readers. I am not surprised at the slant. The data for this article IS available. You simply have to look (and not very hard).

@#12
I'm as cynical or more than most, but you've topped me. Pollsters are paid to get data, not to make it up. If, in fact, the polling company made it up, that would make the data worthless. However, media-sponsored polls are not inherently bad (or worthless). Of course, if you ask the right questions in the right way, you get the answers you want. But, that's why the data, including all of the questions, should be thoroughly scrutinized. All statistics CAN be flawed, but the numbers are the numbers. We can glean information from them as long as we analyze all the variables.

While what people think won't matter for the next election, it SHOULD matter now. If you insist on it NOT mattering now, you just as well give up to the tide of underachievement and meanspiritedness that this state and country is heading toward. WE are still in control as long as WE don't roll over.

Hiram,

This is data. It may or may not be reliable data, but it's data. All polls are purchased so this is no different in that regard, and that fact has no bearing on reliability. Reliability is a function design, sample size and selection etc. As to whether or not pollsters simply provided desired data, that's a different story. They can, but many do not because reliability is ultimately the most bankable quality. Most clients want reliable data to base their decisions on, they don't pay the big bucks to be told what they want to hear... you hire hacks for that.

As to whether people will think the same thing a year from now, there's no reason to think that will change. We'll be facing the exact financial crises next time for exactly the same reasons, with the exact same solutions. The question is whether Democrats will exploit the situation effectively between now and then. I think they can if they keep is simple and focused. One basic theme should an end to these perpetual crises the Republican keep creating.

Everybody!! You're missing the most important point: "data" is plural and takes a plural verb. OK, continue on.

SHEEPLE,Please take the poll and look at the results.If you go to the bottom,you will see that 84% of the people polled voted democrat last year,HARDLY a balanced poll.
Please dont believe all headlines seek truth...

Alan #15,

Sorry but you're wrong I don't know where you get that 84%. The sample had 28% self identified Democrats. Here's what the article actually says about it's sample:

"The overall sample, by the way, consisted of 28 percent self-identified Democrats, 21 percent Republicans, with by far the largest portion (a combined 51 percent) calling themselves independents, identifying with a smaller party, or expressing no description of their partisan orientation. Somewhat similarly, the full sample was composed of 32 percent who considered themselves conservatives, just 18 percent who self-identified as liberals, with the biggest group (50 percent) calling themselves moderates or offering no opinion of their ideology."

This article doesn't surprise me one bit. MN is the land of limousine liberals. I just have one thought. The governor signed the largest budget presented in state history, that the GOP presented before the shutdown, and the GOP gets the blame? That like dropping a 100 pound rock on your foot and blaming Mother Earth? The only problem is the media picks up and runs with it because it fits their agendas. If you only take in maybe $32 Billion in projected revenues, why should you spend 35-39 Billion?? Isn't that what has gotten us into the problems we face?? Minnesotans are like sheep. Follow the leader.

This is checkbook journalism on the cheap. We are not provided with the complete poll report. We are not given access to other polls this firm has conducted which would give us a sense of sampling bias. We aren't told how much this company was paid. What we do know is that the information which is the subject of this "news story" would not exist unless someone at Minnpost had directed that it be created.

As it happens, I like this poll. It supports my opinions, it verifies my prejudices. But I know for a fact that before very long it will be contradicted by another poll, paid for by another news organization, which somehow will deliver results which are in total contradiction to the results here which I find so pleasing.

Really, it's time to stop giving in to this game.

"As to whether people will think the same thing a year from now, there's no reason to think that will change."

Of course it will. Memories of the shutdown will fade, to be replaced by a relentless marketing blitz conducted both on the local and national level, the like of which we have never seen before. In terms of what will happen a year from now, this poll is less than worthless.

#15 and #16, I think the point of confusion is that if you follow the link to take the poll online, 84% of those taking it here were Democrats. That's not the same as the phone poll. That's actually a neat way of letting readers see the questions, which brings me to my one quibble.

That question about whether you saw significant impact from the shutdown is problematic. I took it as did you individually experience severe disruption, and I answered just a little because I didn't see much effect. I might not have known much was going on, from my own experience, but I knew the impact was big. I didn't extrapolate from my own experience, but did respondents know what the intent of the question was? Maybe if it had included the phrase, "regardless of the effect on the whole state" were you personally affected.

Hiram,

From a MinnPost email: "We invite you to chip in to help cover the nearly $10,000 spent on this work, so we don’t have to take all the money away from other news spending."

Granted, that doesn't necessarily refer to the amount paid to the research firm.

And I imagine it wouldn't be too difficult to find other work that this company (Daves and Associates) has done.

//Of course it will. Memories of the shutdown will fade,

Yes but it's not about memory, we'll be facing another shutdown at that time, why will people feel differently about the next shutdown than they did this one? We're gonna have a $2-$4 billion deficit all over again... that's gonna remind people. People are going to want an end to this endless cycle of crises. People want the budget stabilized, and they want it done with revenue and cuts. That's not gonna change.

This won't be forgotten as though it were some verbal gaffe that affected almost no one but did demonstrate something about the candidate. The shutdown affected thousands; families won't forget that they couldn't go to a state park for which they had reservations on the 4th. I won't forget that I was shut out of the Minnesota Historical Society research center, where I needed to work. No state worker will forget the layoff.

//The governor signed the largest budget presented in state history, that the GOP presented before the shutdown, and the GOP gets the blame?

Thomas #17, The size of the budget was not an issue for the majority of MN's, and signing it not what most people are worried about.

The issue here was the government shutdown, who caused it, and why? Very few people were worried about the size of the budget, that was a Tea Party issue. Most folks just want the budget balanced, no matter how big it is, and they wanted it done with a combination of cuts and new revenue. The Republicans refused to compromise on revenue and loaded the budget bill all kinds of social and political stuff that was never a priority for most MN's. This is why the Republicans get the blame. They don't credit for ending the shutdown precisely because Dayton compromised and they didn't, and even then they had a heck of a time passing their own budget.

Patrick,

It would, indeed, not be difficult to find prior work by "Daves and Associates" - because for 21 years, Rob Daves ran the Strib's "Minnesota Poll", which was a national laughingstock.

For whatever reason, during the 21 years Daves ran the MNPoll, his election-eve polls on Gubernatorial, Senate and Presidential races *always* showed the GOP doing worse - usually much worse - than it ended up doing:

http://www.shotinthedark.info/wp/?p=15172

Dr. Jacobs and his Hubert H. Humphrey Institute Poll is even worse:

http://www.shotinthedark.info/wp/?p=15159

Frank Newport of Gallup savaged Jacobs' methodology:

http://www.shotinthedark.info/wp/?p=16480

And here's the funny part: if the final result of an election ended up being really close, like the '08 Senate and '10 Governor's race (as opposed to blowouts, like the '06 Senate race), the Minnesota and HHH Polls *both* shorted Republicans *even more*:

http://www.shotinthedark.info/wp/?p=15162

The reason? Well, it's a known fact that voters are prone to the "Bandwagon Effect"; they do tend to go along with what polls tell them, positively or negatively:

http://www.shotinthedark.info/wp/?p=16522

Is it conceivable that the Strib, Rob Daves, the Minnpost, the HHH Institute and Larry Jacobs are unaware of the bandwagon effect? What the heck, sure.

But I'd be a lot more convinced if Daves didn't have a 24 year record of shorting the GOP on controversial, loaded polls when the chips were down (and Jacobs' polls even worse for seven years).

"And I imagine it wouldn't be too difficult to find other work that this company (Daves and Associates) has done."

I don't see that Minnpost has provided a link either to the firm, to the report they provided, or to any previous polling they have done. Without context, these numbers might have just as well have been pulled out of the air. As will KSTP's when their polling shows different results.

"we'll be facing another shutdown at that time, why will people feel differently about the next shutdown than they did this one? We're gonna have a $2-$4 billion deficit all over again"

That's next session with a different legislature. What matters now is what kind of legislature people will elect 14 months from now. And the set of contradictory polls we might have because of checkbook journalism will tell us nothing about that.

This poll has a 4.8 percent margin of error. I love the second decimal place, by the way, which gives us the illusion of precision. Measured against what? Reality? Of course not. That's the variation in their arbitrarily chosen sample. The pollsters are measuring themselves, not anyone else.

Thomas,

"This budget is the biggest in history" is one of the silliest examples of baffling people with BS. It has no meaning. Until our population starts decreasing and a dollar today is the same as a dollar in 1911, then each budget will always be bigger than the last.

The only exception was with the stimulus aided budget, but with that stimulus, it was still "The largest budget ever".

Every budget we have ever had has been the largest budget ever.

Seriously, what is the point? It's stupid.

Question asked by Daves and Associates

In last year's election for the state Legislature, did you vote for:

These numbers need to be reported in the story.

Thank you MinnPost for making the issue muddy. In the next go around could you refer to the "online" MinnPost poll and the I presume the "telephone" MinnPost poll conducted buy Daves.

I agree that there have been some bad polls and I would suggest using a firm that has more of a history of non involvement in politics and will be happy to suggest a few. Not mine but I use a marketing pollster.

The questions looked valid the sample was a bit small and that shows up in the "sampling error" seems reasonable give the size of the sample and the cost seems about right.

The one thing I can say this was a lot better than the recent polls done by the St. Paul Pioneer Press who can't even get valid questions written.

@26 - "That's next session with a different legislature. What matters now is what kind of legislature people will elect 14 months from now. "

The legislature they will elect 14 months from now is the one that will convene 16 months from now to deal with the new deficit. By then, we'll have seen several budget forecasts and will have a pretty good idea what the new deficit will be. If it's in the multi-billion range (it will be), and the party currently being blamed for the shutdown by most Minnesotans is fielding candidates who once again say they're going to dig in and only look at cuts, do you seriously think the voters, remembering the 2011 session and shutdown (they will, if for no other reason than the dems will remind them), are going to vote for more of the same? Dayton's not up for re-election, so we know what the governor's position will be. Do you really think that the prospect of a 2013 session playing out like that of 2011 is not going to influence their vote in 2012?

If so, they're dumber than a box of rocks. Sure, it's complicated by 2012 being a presidential election year, and a whole host of other factors, most of which can't even be anticipated, but voters would have to have a brain like Homer Simpson to vote in a complete vacuum after what we've just been through in Minn., and what we're still going through in Washington.

Polling aside, Paul's point is that most people have a serious case of crisis fatigue and political gridlock fatigue. They're sick to death of it, and they want long term solutions. They've had, what, 5 budget sessions, a full decade plus, of deficits, gridlock, kick-the-can b.s. instead of real solutions, and real or threatened shutdowns. People want this mess cleaned up, and to most, that means a reasonable, balanced solution.

We can argue about the minutiae of polling from now until the election - personally, I consider polling to be mostly a distraction that's not worth the trouble, even close to an election - but to suggest that current events and budget projections are not going to influence the decision of voters facing a probable replay of the session just past - well, that's a baffling thought process to me.

Ths poll seems to have stirred up a lot of varying comments. I will express my comments this way I have designed, and supported design of polls in Mn and nationally. There are two approaches to polls 1) design the poll to get the answer that the sponsor wants and then brag about it--a common poltical candidate trick or 2) be sure that the poll does not lead to an answer and to that by asking the same question several ways and building in questions that balance the response. From the article and the data presented Mn post did a great service by following Rule 2. I am not at all surprised at the response and breakdown. That is the Mn tradition of seeking responsible good government --thru valued compromise. The polls expressed by both sides previously were clearly Rule 1 which is not a repsonsible approach or it does not have integrty with the public. Thanks to Mn Post the state of Mn was teh winner.
Dave Broden

//This poll has a 4.8 percent margin of error. I love the second decimal place, by the way, which gives us the illusion of precision.

This is actually funny, all this snarkiness about this poll. Anyways, that 4.8 is just the result of the formula that produces the Standard Deviation. it's not unusual to get decimal points, especially when you have multiple variables it's just math. The only reason you don't see it more frequently is many pollsters just round the number or give you a range. I doubt it's there to create any illusions, it's just the number they got when they ran the R.

"This is actually funny, all this snarkiness about this poll. Anyways, that 4.8 is just the result of the formula that produces the Standard Deviation. it's not unusual to get decimal points, especially when you have multiple variables it's just math."

What's funny about the decimal point is the notion that such a wide margin of error could be determined so precisely. The 8 isn't a significant digit. That's why pollsters usually round the number.

"For whatever reason, during the 21 years Daves ran the MNPoll, his election-eve polls on Gubernatorial, Senate and Presidential races *always* showed the GOP doing worse - usually much worse - than it ended up doing:"

That may or may not be the case. For myself, I am never interested in the absolute results of polls. Out of context, I simply don't know what the numbers mean. Is 60 a lot or a little? I don't have information that answers that question. What can give polls some value is what they tell us about trends. How do people respond to the same question over time?

I have to confess, I am never happy with poll results. If they are in my favor, I worry they make my candidates over confident. If they are against me, I worry that they hurt morale. And all for something that even the people who are responsible for it concede "one can be 95 percent confident that results in the poll will vary by no more than 4.8 percentage points," whatever that means. Consulting the entrails of a goat would yield information just about as valuable, and more likely to be right.

Rounding the 4.8 may or may not be a good idea. It can backfire, if the actual numbers are to be made available and someone notices that actual number was 4.8 instead of 5, you're gonna get grief for that. I prefer just providing the number the calculation yields, readers can round for themselves.

One thing I'll have to say for Hiram's position, the Republicans have been running against the poll numbers for decades and winning. Or rather the Democrats have failed to exploit the poll numbers and keep losing. Even when the Democrats win they lose. For instance the majority of Americans now support the idea of some kind of government run health program, yet it was the only plan the Democrats steadfastly refused to consider. Meanwhile the Republican's have been making the false "silent" majority argument for decades and winning elections.

I think subsequent polls will more or less confirm Minnposts, with the exception of USA that KSTP keep hiring. They have completely lost all credibility however, their last set of questions was egregiously bias.

The statistical dialogue is good but those of you hung up on that are missing the message that the people of Mn have expressed. As I travel around the state I hear what the poll says--results desired, Mn traditional values, not more complexity to government etc. We will pay more if the value exists. I would expect that there would be some comments about the poll message and impact and less about the issue of statistics--by the way I do a lot about statistic and this is not a bad representation. This poll should show that the Mn tradition of progessive yet fiscally responsible government is the desire of all Mn and both parties and particularly the pundits of both side need to gt that message or we will continue with the deadlock and pushing solutions down stream.
Dave Broden

Question for Erik Black and Doug Grow:

Rob Daves and the Minnesota Poll have a documentably awful track record. As I showed in the extensive series of posts linked above, Daves' polling consistently shorted the GOP; the randomness of errors in Strib polls before Daves vanished when he took over.

And yet year in, year out, you (plural) present Daves' polling as "news". Even though it's quite clearly deeply flawed.

I'll anticipate your response: "Polls are snapshots in time". Yep, but the snapshots I compared were the Strib's traditional election-eve polls; the "snapshot" was as close to the "snapshot" captured on Election Night as the Strib could make it. And yet the MN poll's "Snapshot" is not only widely divergent from the one the electorate and the Secretary of State provide on election day, but the divergence is almost always pro-DFL and anti-GOP; the closer and more contentious the race, the more the polls short the GOP. (And your pal Larry Jacobs' polls are even worse).

So my question for you gents is this: if you had a source on, say, police corruption, and that source burned you with wrong, or grossly inaccurate information, *every single time you used him* for 24 years, would you keep on citing him as an authoritative source on the subject?

Why?

"Rounding the 4.8 may or may not be a good idea. It can backfire, if the actual numbers are to be made available and someone notices that actual number was 4.8 instead of 5, you're gonna get grief for that."

The problem is that 8 is not a significant digit. Probably the 4 isn't a significant digit either. Really, we have no idea what margin of error is because we have no correct answer to compare this poll against. We will never know what "actual" numbers are.

"the Republicans have been running against the poll numbers for decades and winning. Or rather the Democrats have failed to exploit the poll numbers and keep losing."

Democrats think numbers like these are real. Republicans simply see them as something to campaign against. Obviously, the Republican approach has been far more successful, more grounded in reality. I have literally seen DFL candidates refer to poll results as if they were something that exists in the real world, not just in the imagination of pollsters, and their enablers in the media. Needless to say, I am scathing when I hear that happen.

Why?

The reason newspapers do what they do is to sell newspapers. It's like the old Buffett story about fishing lures. Old timer shows friend a fishing lure he just made. The friend says "It's beautiful, how good is it in catching fish?" The old timer replies, "Well, I am not selling it to the fish." If reading polls please people as they do, that's good enough. It simply doesn't matter if they are accurate. And the fact is, the margin of error is so wide, you very rarely find a poll that can be classified as inaccurate.

//The problem is that 8 is not a significant digit. Probably the 4 isn't a significant digit either. Really, we have no idea what margin of error is because we have no correct answer to compare this poll against. We will never know what "actual" numbers are.

Hiram, I'm beginning to think you don't trust surveys. Listen, the .8 is simply a product of the equation, that in and of itself doesn't render it insignificant. Now it's true that if you put garbage in you will get garbage out, but you seem to saying that it's garbage simply because it's a survey. I guess you don't believe it but there's a lot of evidence that surveys can yield reliable data. I'm looking at this survey and I nothing's jumping out at me as bizarre or methodologically unsound, it's not just that I agree with it. Yes, we all have bias, but we are capable of compensating for it. The only methodological issue that's cropped up in the last few years has been cell vs. land line sample acquisition, but that's still not understood yet, so it's hard to discount something simply because it's a survey.

Hiram is raising an interesting issue but I'm not sure agree with his conclusion. We have to separate the issue of a surveys reliability from it's utility, or influence. Surveys can be reliable in the sense that they tell you what a population thinks, but that's no guarantee that politicians will act accordingly. Sure, people use polls to their own advantage, and they shop for results they like, but that's not the problem.

The problem is that politicians don't have to represent the majority of their constituents. Sports subsidies are a perfect example of this, despite large public opposition they keep passing the huge stadium deals. That doesn't prove the surveys are wrong, it simply illustrates the fact that politics aren't based on surveys. So on one hand, yeah, survey numbers are irrelevant because politicians use or ignore them at will. Decisions are not made according to popularity. But on the other hand, a reliable survey could be used to form good policy, frequently policies that run contrary to popular support are bad policy after all. I think these number are reliable, will be repeated in subsequent surveys, and reveal a strength Democrats could exploit, but I would make my decisions based on a single survey or surveys alone.

"the .8 is simply a product of the equation, that in and of itself doesn't render it insignificant."

The fact that is is the product of or related to an equation doesn't make it significant either. We don't poll equations, we poll people.

"I'm looking at this survey and I nothing's jumping out at me as bizarre or methodologically unsound, it's not just that I agree with it."

It tells me exactly what I want to hear. But then I know the KSTP survey will tell me exactly what I don't want to hear. Is the problem with my hearing?

sorry, meant say "wouldn't" make my decision... in #41.

"Surveys can be reliable in the sense that they tell you what a population thinks,"

What surveys tell you is how an arbitrarily selected group of people answer questions. Anything beyond that is extrapolation and guesswork. They most assuredly don't tell you what people are thinking.

In penance for what were undoubtedly grievous sins committed in prior lifetimes, I have actually done my share of canvassing, although I will say, I have never canvassed an equation, a margin of error, or indeed a number of any kind. Let me tell you, it's a hit or miss business. Most people won't answer your questions. Others will answer in a way that's difficult to understand. Still others are deliberately deceptive. Polling results depend on weather or whether the Vikings are on. For best results, I recommend canvassing with a pretty girl in a tank top on a warm summer day.

If MinnPost would release the numbers for the respondents that voted democratic in the last election--as was asked in the survey I bet it would show 80% voted for the dems.

"If MinnPost would release the numbers for the respondents that voted democratic in the last election--as was asked in the survey I bet it would show 80% voted for the dems."

Research suggests people lie about who they vote for to pollsters.

In Israel, I have heard it said, people tell the truth to pollsters but lie when they vote.

.
Oh, sweet mother of biased poll results. Daves and Associates? Who are the Associates? The same Associates who got decades of polls wrong? And Larry Jacobs to boot. How quaint. All the usual suspects.

People like Hiram believe polls and consider them to be accurate predictors of the future when the polls support their beliefs. But, when the polls don't, they dismiss them as biased and as having no predictive value.

Selective acceptance is a common form of self deception on both ends of the political spectrum. The fact is that the people of Minnesota largely blame the GOP for the shutdown. It's the DFL's job to make sure that they continue to feel that way and the GOP's job to get them to forget.

But, right now, the GOP has lost whatever mandate they had.

Interesting that the Republicans cannot find anyone whose taxes can be raised. When it comes to a sports facility, that the majority of us do not and never will use nor benefit from since once built it has been proven that they do not generate that much local revenue, they have no problem going along with additional taxes. Those taxes provide financing for these same billionaire's and millionaire's they are unwilling to tax at an equal percentage to maintain everyday, used by everybody, government services like education, fire, police, roads and transportation systems. Hypocritical? Oh, yeah!

A lot of space in these comments has focused on the poll cited - polls are a look at a moment in time which at best can only provide information about a "trend" and worst reflect just that moment and nothing else. The additional information in this article, quotes from the leaders, is more relevant to what the future holds for Minnesota and the budget, IMHO. The intransigence of the GOP leadership is very scary.

//It tells me exactly what I want to hear. But then I know the KSTP survey will tell me exactly what I don't want to hear. Is the problem with my hearing?

Therefore if a survey tells you what you don't want to hear it must be true? It's not about what you want to hear, it's about the answers to the questions. The people being surveyed have no idea who you are, or what you want to hear.

"Therefore if a survey tells you what you don't want to hear it must be true?"

I can tell you it's the poll I push. Favorable polls breed complacency. But of course, neither a KSTP poll or the polls here have any objective reality. They exist because someone paid for their creation. They are instances of news organizations manufacturing news-like information, rather than actually covering the news.

"The people being surveyed have no idea who you are, or what you want to hear."

Not me specifically, but lots of people give answers to pollster as a way of affecting the news and the perception of the news. I would never dream of answering a pollster by telling him what I actually think. Instead, I give the pollster the answer I think would best contribute to the political effect I want to achieve. I think most people answer polls that way when you get down to it.

Speaking as one who has done a lot of political canvassing a lot of different forms, I can tell you from experience that political posturing is the rule, not the exception. People give you the answers they feel it's in their interest to give to a much greater extent then they give you answers reflecting what they really think. I wonder how the math account for that when calculating sampling error. Is there a statistical equation for disingenousness?

Hiram, you're just getting too convoluted. Most people don't put that much thought into answering a survey question, most people CAN'T put that much thought into answering survey questions. Most Americans don't put that much thought into politics, they couldn't answer surveys the way you suggest even if they wanted to.

To answer your question regarding calculating disingenouity, you can build that into a survey by asking the same question in different ways, but most pollsters don't take the time to it.

I think it's interesting that Hiram keeps coming back to this notion that surveys are damaging because they breed complacency when they concur with pre-conceived beliefs. I don't know anything about the internal working of the Democratic party, but I don't think surveys are their problem. At any rate, this would betray a serious misunderstanding of a surveys function.

"you're just getting too convoluted. Most people don't put that much thought into answering a survey question, most people CAN'T put that much thought into answering survey questions."

It's more of an instinctive thing. But in canvassing, you see it all the time.

"To answer your question regarding calculating disingenouity, you can build that into a survey by asking the same question in different ways, but most pollsters don't take the time to it."

Answering questions in different ways, always get different results, and then you have apples and orange problems. There are solutions to some of the problems I raise and pollsters do employ them. Those just aren't the kind of pollsters hired by newspapers who have sort of an old fashioned mentality about these kinds of things.

"I don't know anything about the internal working of the Democratic party, but I don't think surveys are their problem."

I do know something about the internal workings of the Democratic Party, and surveys are more a minor irritant than they a problem, in terms of complacency at least. Since I don't think much of polling generally, they are not of much use or interest to me. But for Republicans they are more useful. The very poorly constructed KSTP poll on the shutdown was a major political asset for them during that whole mess. And in general, polling is a target of opportunity in their long term campaign to discredit any media which is not under their control.

Something else to keep in mind. Folks like me are often on the side of issues that don't poll very well. Voter ID is one such issue that comes to mind. Those kinds of polls are used as weapons. But what I know, is that the answers to those polls are dependent on the question is asked, and that if the question is asked differently, you can often get dramatically different results. How do things like that factor into "margin of error", or "sampling error", the pseudo scientific term of choice for pollsters intent on concealing the messiness and very unscientific nature of the activity they engage in?

//Answering questions in different ways, always get different results, and then you have apples and orange problems.

Just to clarify, it's not about answering questions in different ways, it's about answering a question about the same thing. This is a standard reliability metric that can be applied, and is applied in sophisticated questionnaires (something like the Minnesota Multiphasic Personality Inventory for instance). You're not comparing apples and oranges you're looking at consistency. You basically give the respondent a few opportunities to contradict themselves, the more they contradict themselves the more less reliable their responses are. I've been out of the field for a while now so I can't remember exactly what this is called, but there's a term for it. I'm not saying pollster's use it, but can be used.

"This is a standard reliability metric that can be applied, and is applied in sophisticated questionnaires (something like the Minnesota Multiphasic Personality Inventory for instance)."

When done with polling, what it tells you that polls aren't reliable, and they aren't consistent. It tells you the answer you get is very much influenced by the way the question is asked.

People are more complicated than polls.