Green Room

Salt Lake Tribune editors requested reweighted poll based on partisan split?

posted at 1:02 pm on November 6, 2012 by

John Podhoretz is on top of an interesting story involving the Salt Lake Tribune and Mason-Dixon.  It seems that the Tribune has admitted requesting that the pollster reweight the results of a poll when the outcome displeased the newspaper, asking to increase the weighting for Democrats, which they felt were undersampled.  And this is interesting, because most pollsters dismiss the complaints about partisan splits in their samples from poll critics:

Tribune editor Nancy Conway acknowledged the problem.

“We are as concerned about this as anyone,” she said Monday. “As soon as we understood there was a problem we worked to correct it.

“We had no reason to doubt the poll until we saw others conducted over the same period and could see differences in the numbers. That raised questions,” Conway said. “We contacted our pollster who did additional research on Salt Lake County demographics and found there was indeed a flaw.

“We knew right then that we needed to correct our mistake and that’s what we are doing,” Conway said.

John offers up a stinging indictment:

To recap: A newspaper pays for a poll. It doesn’t like the look of the results. So it asks the pollster to reexamine them and alter them by changing his “weights.” He does so; he may agree with the call (as the Mason Dixon pollster says he does in the story) or he may be simply serving the interests of his paying client.

And it will do so based on the partisan split—the very controversy that is dismissed so cavalierly by media types.

We only know about this one because of the highly unusual circumstances of its revision. The question you have to ask yourself now is: How many times does this happen before a poll is published?

Good question.  We may or may not find out today, eh?

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

Yet another data point in support of those who are paying attention and have called out the polls for their bias.

Skwor on November 6, 2012 at 1:10 PM

What are the checks and balances on the media..?

d1carter on November 6, 2012 at 1:11 PM

Interesting. This may explain the thickheaded insistence pollsters seem to have this cycle to mimic 2008 turnout. Any person being even only slightly honest knows there is no way it will be a D+7 election. Not after 2010. Not after the clear indicators that Republicans are nowhere near as unenthusiastic as they were during 2008. An important point because Democratic turnout was inflated in 2008 by historically low Republican turnout.

NotCoach on November 6, 2012 at 1:14 PM

Nate Silver hardest hit.

WordsMatter on November 6, 2012 at 1:17 PM

“We had no reason to doubt the poll until we saw others conducted over the same period and could see differences in the numbers. That raised questions,” Conway said. “We contacted our pollster who did additional research on Salt Lake County demographics and found there was indeed a flaw.

Supposedly they saw a flaw in the demographics, yet they don’t see the flaw in their logic. Of course the weighting is tied to demographics, but demographics is the base weight and is adjusted based on historical turnout and election season enthusiasm, mainly because it’s likely voters, not registered voters or voting age population.

If the want to be honest let them publish the original poll submission and the published poll with the revision. Until then it’s in the 100% bias category.

Dusty on November 6, 2012 at 1:23 PM

IN case people don’t know. The tribune is liberal. Not to mention anti-Mormon.

petunia on November 6, 2012 at 1:23 PM

“We are as concerned about this as anyone,” she said Monday. “As soon as we understood there was a problem we worked to correct it.

IOW, the Tribune got a call from the Democratic candidate complaining about it and they obliged him with the correction.

Which brings up a question: How stupid are the people at the paper that they don’t know the demographics at the start of the election season, or at least by May or so, such that they only caught the ‘flaw’ five days before the election? Talk about having a pulse on the community they “serve”.

Dusty on November 6, 2012 at 1:30 PM

Talk about having a pulse

I suppose that should be, “Talk about having a finger on the pulse”.

Dusty on November 6, 2012 at 1:31 PM

IN case people don’t know. The tribune is liberal. Not to mention anti-Mormon.

petunia on November 6, 2012 at 1:23 PM

Exactly right–you would think that in a relatively Mormon and somewhat right-of-center area like Salt Lake City, the Salt Lake Tribune would be more conservative and sympathetic to the Mormon faith. It is anything but.

I am so hoping that Mia Love will win her race for Congress from Utah’s 4th District!

Sir Rants-A-Lot on November 6, 2012 at 2:11 PM

We know the data are wrong, because the results were not what we expected. Therefore, keep collecting data until you get the right result. So, the reason we do polling is analytically verify what we already know to be true. Got it.

STL_Vet on November 6, 2012 at 2:17 PM

Noteworthy that this is happening in Utah – one of the states I feel more confident about protecting our Constitution.

KMC1 on November 6, 2012 at 2:38 PM

The Salt Lake Tribune has a long history of deception. A couple of weeks ago when it endorsed Obama, people noted that the only way Romney could have received the Tribune’s endorsement was if the Mormon Church officially endorsed Obama. Then, the Tribune would select Romney. It’s always amazed me that a paper that regularly denigrates the majority of the local population remains in business.

Alma on November 6, 2012 at 2:40 PM

Wow, manipulated statistics. Who would have guessed? I mean, this never happens right?

These are the same people who claim voter fraud is a non issue.

dogsoldier on November 6, 2012 at 3:01 PM

I remember this during the “Great Gun Control Debate”.

They would ask 7 or eight loaded questions, like “Do you favor Seizure of all Firearms” and “Do you favor laws that prevent babies from being shot in the head?”.

Then the news cast would sound something like this, “In our CBS Nation Wide Poll, we found that almost no one uspports the idea of Americans losing the right to own a gun, but almost all Americans are in favor of Sensible guns laws, similar to the (Insert Gun Registration Bill Here) Law.

We are being Setup everyday.

Bulletchaser on November 6, 2012 at 3:02 PM

We know the data are wrong, because the results were not what we expected. Therefore, keep collecting data until you get the right result. So, the reason we do polling is analytically verify what we already know to be true. Got it.

Kind of like global warming. No issue, move on.

Strike Twice on November 6, 2012 at 4:26 PM

I’m convinced this is all part of the MSM and DNC narrative for their after-election challenges – “Look, everyone loves us! The polls say so! So a loss on election day must mean something was changed.”

EdmundBurke247 on November 6, 2012 at 4:54 PM