Green Room

Re: Marist

posted at 6:30 pm on November 3, 2012 by

Duane, I have to say I’m surprised that pollsters like Marist have not adjusted their likely-voter screens as Election Day nears, and I’m not sure that polling criticism isn’t the reason. In elections past, sample data didn’t get as much scrutiny, perhaps because in 2008 the outcome wasn’t in that much doubt. I’m wondering if pollsters like Marist simply are doubling down on earlier bad calls, afraid to move because it would undermine the long-term trending in their poll series.

In cycles past, we would see media polls like the WaPo/ABC and CBS/NYT suddenly adjust their samples in the final week in order to have bragging rights after the election. Either these pollsters are convinced beyond any doubt that the turnout model is going to be similar to 2008, or they are more reluctant to shift samples at the end with all of the increased scrutiny. We’ll see how that works out on Tuesday, but at least we can be sure that the final assessments in 2012 will apply to the totality of their polling, and not just the final week.

Recently in the Green Room:

Blowback

Note from Hot Air management: This section is for comments from Hot Air's community of registered readers. Please don't assume that Hot Air management agrees with or otherwise endorses any particular comment just because we let it stand. A reminder: Anyone who fails to comply with our terms of use may lose their posting privilege.

Trackbacks/Pings

Trackback URL

Comments

Ed, I think you’re giving Marist way too much credit.

This is brazen push-polling, nothing more.

fiatboomer on November 3, 2012 at 6:35 PM

I think by this point the pollsters know they’ve committed themselves, and instead seek strength in numbers. Instead of just PPP being by themselves as the fools, they’ll have company in Marist, Q, SUSA, WAA, MLS where they can all just say what an odd anomaly this election was.

Everybody getting it wrong (with Ras and maybe Gallup, Susquehanna being able to claim victory as being the closest to correct in their own venues) means everybody can pretend it was an oddity with the election cycle. Maybe blame Sandy or early voting or something, instead of having to admit that they had no justification for using D+9 samples to begin with.

Gingotts on November 3, 2012 at 6:39 PM

For the first time in this entire election cycle, I saw that Democratic pollster Doug Shoen on FOX and he was not entirely confident as to how things would work out. All three of the pollsters on the panel said it would likely be determined by turn out, and they agreed that the conservatives were much more enthused about getting rid of Obama, than Obama voters were to keep Obama. To this I say, “uh, yeah”.

BettyRuth on November 3, 2012 at 6:48 PM

Jay Cost has been saying that the likely voter screens are bad because of lower polling budgets. The tighter the likely voter screen, the more expensive it is to conduct the poll. If you are eliminating more voters from your sample when you are narrowing it down to LVs than you have to make more calls in order to maintain a valid sample size. This costs more money.

Mark1971 on November 3, 2012 at 6:50 PM

Speaking of Dem push pollsters; PPP admitted earlier that Obama is down 10 points or more from 2008 in many of the blue states they have “polled”.

Yet somehow, Obama is not down at all in OH and leads nationally… Riiiiight

Norwegian on November 3, 2012 at 6:52 PM

Most every poll is skewing to Obama. It’s a blantant lie and they know it. Along with all the alphabet networks (except for Fox). Many of our most prestigious newspapers too, are skewing to the left, hoping against hope to effect the direction of this election.

The result is inevitable, Romney wins on Tuesday. It may not be a route but it will be decisive. If my likely scenario happens how do we ever give the media any respect. How can we ever believe them again?

Tonynoboloney on November 3, 2012 at 7:01 PM

A major polster’s primary financial concern is remaining in good standing with the Main Stream Media outlets that use (and therefore give credibility to) those polsters.

Any truly reliable polster predicting a solid Romney would quickly become anathema to the those who prop up their credibility would be brought into question, and eventually they would lose their ability to make a living at being polsters.

What is telling in this is several non-profit polling groups (some of which are university research groups) have from the start predicted a Romney win. Yet, hardly anyone talks about those polls, and those polls are rarely discussed in main-stream news reports.

Lawrence on November 3, 2012 at 7:12 PM

Why are polls skewed this late in the game? Simple to explain. Tuesday, they will be reports about turnout, the media will report that turnout is similar to 2008. It will be a lie folks! Purpose, voter disenfranchisement!

jjnco73 on November 3, 2012 at 7:16 PM

Why are polls skewed this late in the game? Simple to explain. Tuesday, they will be reports about turnout, the media will report that turnout is similar to 2008. It will be a lie folks! Purpose, voter disenfranchisement!

jjnco73 on November 3, 2012 at 7:16 PM

This is all they have folks!

jjnco73 on November 3, 2012 at 7:18 PM

I think by this point the pollsters know they’ve committed themselves, and instead seek strength in numbers.

Gingotts on November 3, 2012 at 6:39 PM

Crowd psychology. There is a degree of impunity that arises from being part of a crowd. People will do things that they would never do on their own. Think lynch mobs, riots, investment bubbles, etc. It’s a fascinating phenomenon.

RedWinged Blackbird on November 3, 2012 at 7:21 PM

The problem with this election is that if Obama wins, this behavior will be rewarded. The liberal pollsters will do this again in 2014 and 2016.

This, compounded by very few Republican pollsters left, will continue the trend of the politicization of polls. And the worst offenders are Marist and PPP, who release non-stop polls and skew all samples with the trash they release.

milcus on November 3, 2012 at 7:25 PM

The problem with this election is that if Obama wins, this behavior will be rewarded. The liberal pollsters will do this again in 2014 and 2016.

This, compounded by very few Republican pollsters left, will continue the trend of the politicization of polls. And the worst offenders are Marist and PPP, who release non-stop polls and skew all samples with the trash they release.

milcus on November 3, 2012 at 7:25 PM

And at that point we will have left the period in American history in which we had free elections and a fair vote count. Did anybody follow the election in Ukraine last week, and the odd discrepancy between the split, generally anti-Russia desires of the electorate and the clearly pro-government, pro-Russian outcome? This will be our future.

Gingotts on November 3, 2012 at 7:37 PM

i honestly wish i were as optimistic as, well, almost everyone from Rove to Barone to Kristol to Cost

but if mitt wins, there will be hell to pay….the left, in all its institutional might, will deny the results. Actually what the left has now is brute force of opinion management. leftists policies are stupid and hellish.

and that is why obama deals straight up contempt for the non-Left. Revenge for running against ME. So if Romney wins we will have condemnation of the political process. If obama wins, we will have condemnation for anyone who stands against him.

if mitt wins, the corruption of the polls, the tv, the op/eds will all be shown to be a lie…but if obama wins it will be a great Victory of the People

r keller on November 3, 2012 at 7:38 PM

milcus on November 3, 2012 at 7:25 PM

..on the other hand, if Obama loses, this will be the death knell for the MSM and pollsters — well, almost. Along with Obama’s regime of incompetents, they have so much tried to wrest the narrative away from the good, honorable people in main stream America to put it on a liberal socialist agenda and they know what the stakes are.

If When Romney wins, the economy will get back on track and jobs will increase — not only because of Romney’s business acumen and bipartisanship but also because of Obama’s incompetence and divisive mien — and 2016 will make people far more content with the incumbent.

It will be so-o-o-o-o over for them.

The War Planner on November 3, 2012 at 7:41 PM

Anyone relying on RCP poll averages is a fool. They have so many outliers in there that distort the average. This will be the last election that many reasonable people will look to RCP as the standard for poll watching. My prediction is that Gallup may be the only pollster to come close to Romney’s national victory margin, assuming their resumption of polling since the hurricane hasn’t affected their methodology. The rest of them will be so far off that I wouldn’t give them the time of day to explain why they missed it. I am including Rasmussen in that group. And this includes most of the individual state polling as well. No excuse by any of them will be acceptable.

The Good Doctor on November 3, 2012 at 8:07 PM

At this point, I think the polls are skewed to try to save the Democrat down ticket races.

Conservalicious on November 3, 2012 at 8:11 PM

I figure that about this time tomorrow night, Obama will be receiving from his pollster the same bad news that Pat Caddell gave Jimmy Carter on election eve in 1980- “it’s over”.

The Good Doctor on November 3, 2012 at 8:20 PM

The result is inevitable, Romney wins on Tuesday. It may not be a route but it will be decisive. If my likely scenario happens how do we ever give the media any respect. How can we ever believe them again?

Tonynoboloney on November 3, 2012 at 7:01 PM

We already don’t believe the media. What will happen is that we also won’t be able to believe any pollsters either. It’s possible we may end up seeing “alternative conservative pollsters” develop the same way we’ve had to develop talk radio and the blogs as alternatives to the MSM.

Of course, this assumes the pollsters are all in fact wrong, something I’m not entirely sold on yet.

Doomberg on November 3, 2012 at 8:43 PM

I’ll take door number 3 – Marist is taking its cue from the Obama Presstitute Organs and trying to create “truth”.

Steve Eggleston on November 3, 2012 at 9:18 PM

Marist is not quite a business, those guys exist off fees from sucker students who attend that third rate university. They may feel free to pursue their political agenda rather than a reputable polling business that is profitable.

bayview on November 3, 2012 at 9:34 PM

At this point, I think the polls are skewed to try to save the Democrat down ticket races.
Conservalicious on November 3, 2012 at 8:11 PM

I also wonder how much is adjusted for voter fraud.

conservative pilgrim on November 3, 2012 at 10:11 PM

In elections past, sample data didn’t get as much scrutiny, perhaps because in 2008 the outcome wasn’t in that much doubt.

Not to give you a hard time, Ed, but if I recall correctly, your final projection had McCain winning.

But yeah, Marist’s polls don’t pass the laugh test.

LukeinNE on November 3, 2012 at 10:17 PM

Hey guys–

There’s a reason why pollsters use previous polling data to base their samples for current polls– it’s because the future hasn’t happened yet.

I think we’re all aware there’s the 2010 midterm Tea Party wave that took over the House, but Midterms don’t compare well to Presidential elections because the turn-outs are generally quite different. Presidential elections have a different attraction to the population, so it draws more people.

When it comes to this election cycle, the pollsters could draw from 2008 or 2004 or 2000 for basis sets for the sampling of the electorate. The issue is we don’t know how to model the 2012 election because we don’t know what it will be until Tuesday.

And in 2016, it’s going to be a different again, and we’ll be complaining about sample populations.

Nethicus on November 3, 2012 at 10:46 PM

Nethicus on November 3, 2012 at 10:46 PM

That’s all well and good as a matter of science, but the purpose of polls is to predict what will happen. If they’re not predictive, they’re useless. This is a fact that all the pollsters have readily admitted when pressed in the last two months.

These pollsters get the vapors when Republicans suggest that they should adjust their samples to at least approach something that might bear a passing resemblance to reality – quelle horreur! That would be tampering with data! An affront to the science of polling! That’s all well and good, but these people are skewing for things other than party ID – race, gender, income, cellphones – that they can fix if they wanted to. They don’t.

HitNRun on November 3, 2012 at 10:57 PM

Many pollsters have stopped adjusting their D/R/I split in the final days for years now. It’s not new. It’s just that there are more new pollsters and they just don’t care. In 2010, they didn’t adjust in the final days. In the WI recall, they only adjusted slightly. In the 2008, they did the exact same thing against Hillary and she kept winning the primaries, sometimes setting the polls off by more than 30%. She was behind on the caucus’ so that didn’t help her. But the pollsters go against Dems in primaries too.

This isn’t new. It’s just more blatant.

MrX on November 3, 2012 at 11:46 PM

It’s very simple. If Obama goes down, the Liberal gatekeepers of our popular culture want Obama’s supporters to be outraged and to cry foul play.

After Watergate, every Republican scandal, real or faux, was not only compared to Watergate, but had the word -gate attached to it. Does anyone think that the fact that the Democrats were on the losing end of the Florida recall in 2000, and all the outrage that they felt over it, is not something that they’d love to replay every time a Democrat Presidential nominee loses? They claimed that the 2004 election was stolen as well. Add to that the fact that Obama is Black, and it’s the perfect storm for all sorts of bedlam.

If Obama goes down, they want there to be hell to pay. This is the main reason that the polls are skewed, more so than creating an illusion of inevitability for Obama’s re-election.

ardenenoch on November 3, 2012 at 11:59 PM

This is the presidency that all shame was thrown out the window. The MSM, the “fact-checkers”, the pollsters. None of them care anymore.

John the Libertarian on November 4, 2012 at 12:15 AM

These polling numbers are not trying to suppress turnout for Romney. Thy media is really trying to limit the damage in down-ticket races so the Dems don’t get totally killed in the house and senate races. Obama is going to lose big on Tuesday, the media and pollsters know it and are trying to get dem turnout up to save a few congressmen.

spec_ops_mateo on November 4, 2012 at 1:01 AM

I’m waiting to see what election day exit polls will be saying–that is, the ones they’re still doing. For how many states will we hear what we heard on WI recall day–the race is tight, neck’n’neck, and then we find out Walker wins bigger than the first time.

stukinIL4now on November 4, 2012 at 1:11 AM

Changing their likely voter screens now would make them worse, though. You shouldn’t change the methodology during the middle of the race, or at least not at a time when so many other changes are happening – doing so impedes your ability to measure actual changes in opinion. Better to have a bent yardstick that is consistently bent.

The point of polling isn’t simply to predict the winner, it’s to measure changes from one week to the next. A pollster who consistently uses the same flawed model week after week will still provide a better picture of what’s happening – this is why even the crazy D+9 polls show romney improving after the first debate, etc. Adjusting the model during the race, especially near the end when things are so fluid, defeats this purpose.

RINO in Name Only on November 4, 2012 at 5:46 AM

Just waitin for LibFree and Gumby to weigh in . . .

BigAlSouth on November 4, 2012 at 6:39 AM

Marist is taking its cue from the Obama Presstitute Organs and trying to create “truth”.

1. They bought the White House argument to totally ignore 2010 and look only at presidential elections
2. They bought the WH argument to consider 2008 D turnout not as a high point, but as part of an upward trend.
3. Relatedly, they project a greater population base that is more likely to vote D. There’s something to that, unfortunately. More poor people, whether from immigration or job loss, means more D voters in general.

The polls are definitely skewed Democratic, but I fear by not as much as conservative pundits assume.

G. Charles on November 4, 2012 at 7:00 AM

THE NEW YORK DAILY NEWS Endorsed Romney.

The new Trib poll in PA has Romney at 48%- Obama 48%.

And, did I mention that the freakin’ NY DAILY NEWS ENDORSED ROMNEY?

So, a D+9 sample in that Marist poll? It’s just. not. happening.

mountainaires on November 4, 2012 at 8:10 AM

And, don’t forget: Just one week ago…

Newly released figures from Gallup show that the demographics of the American electorate (age, race, sex, etc.) have changed very little since 2008 except in one way:

Party affiliation has swung dramatically toward the Republican party, and away from the Democratic party, during President Obama’s term.

Gallup’s tallies on the composition of the electorate show that, among likely voters, Democrats held a 10-point advantage over Republicans in party affiliation in 2008 — 39 to 29 percent.

Four years later, Gallup’s tallies show that, among likely voters, Republicans hold a 1-point advantage over Democrats in party affiliation — 36 to 35 percent. That’s an 11-point swing in just four years.

Gallup writes that “the largest changes in the composition of the electorate compared with the last presidential election concern the partisan affiliation of voters.” It adds, “In fact, the party composition of the electorate this year looks more similar to the electorate in 2004 than 2008.” In 2004, according to Gallup, Republicans enjoyed a 2-point advantage over Democrats in party affiliation — 39 to 37 percent.

When independents who lean toward a given party are included, the changes in party affiliation are even more striking. In 2004, Gallup showed that Republicans and Democrats were tied in party affiliation when leaners were included — at 48 percent apiece. In 2008, Democrats enjoyed a 12-point edge when leaners were included — 54 to 42 percent.

In 2012, Republicans enjoy a 3-point edge when leaners are included — 49 to 46 percent. That’s a 15-point swing since 2008.

To be sure — despite these swings — Gallup still shows Republicans with only a 1-point edge in party affiliation (without counting leaners). But given how well most recent polling has shown Mitt Romney doing among independent voters, the GOP nominee would presumably be quite happy to supplement his apparent advantage among independents with even a 1-point advantage in turnout.

In short, if Gallup is right about likely voters’ party affiliation — and especially if it’s right about the 11-point swing toward the GOP since 2008 — that’s very good news for Romney and very bad news for Obama.

Source: Weekly Standard

mountainaires on November 4, 2012 at 8:14 AM

CNN’s John King was stunned by Romney crowd in Ohio

In a report from Ohio on Friday, CNN’s John King noted that in comparison to 2008, the Romney campaign’s ground game and the verifiable enthusiasm King’s seeing, give Romney a real chance to take the state:

http://landing.newsinc.com/shared/video.html?freewheel=91074&sitesection=breitbartprivate&VID=23868461

mountainaires on November 4, 2012 at 8:17 AM

Oops, sorry, wrong link above. Wish we could edit.

Original story with video:

http://www.breitbart.com/Big-Government/2012/11/03/Obama-Cleveland-80k-compared-to-4k

mountainaires on November 4, 2012 at 8:20 AM

D+7, D+9, D+12

Lies, Damned Lies, and Bullsh!t.

Quote me.

BigAlSouth on November 4, 2012 at 9:01 AM

Civil war looms. It is just a matter of geography and degree.

Mason on November 4, 2012 at 10:42 AM