Monday, October 1, 2012

So Just How Deep DOES
That Rabbit Hole Go?
Take A Look At What
We're Dealing With:

Right after some polls showed over 70% of Democrats think the economy is doing just fantastic & the country is headed in the right direction, one of them popped into my Twitter TimeLine to argue about polls.  

Here's how it went: 

So this Liberal Democrat jumps into the conversation I'm having to claim he sees no evidence whatsoever that Hope N' Change fever has abated, or that Republicans will turn out in greater numbers in this election than turned out in 2008.  

OK.  Here we go.
Here's the Pew Link about only 9% of people contacted by pollsters agreeing to give a response to their questions: 

http://pjmedia.com/tatler/2012/09/30/we-are-the-91-only-9-of-americans-cooperate-with-pollsters/

Since Cody asked 'based on what measure' am I saying more Republicans will turn out for the 2012 Presidential election than voted in 2008, I supplied him with one: 

Here's the pic I gave him: 
Unless something happened since July that I missed, the Hopey Changey stuff is still pretty much dead.  

Cody then made the excellent point that a pollsters rep is made on being ACCURATE.  Why yes - in their very last poll before before the election is actually held.  Before that? Not so much.  As I now attempted to explain to him.

To demonstrate just absurdly some of the media's national polls had been skewing lately, I posted this chart: 


Cody didn't even blink on seeing MSM polls giving Obama a national lead of over 10 pts.  At this point I was pretty much sure I wasn't dealing with a rational person.  

Note the dodge Cody tries:  Where's your source for Hope N' Change fever ending, or that more Republicans will turn out this year than in 2008?  Oh Gallup? You do have a source?  Just one?  Meh. 

Yeah. He asked for evidence.  He got it.  He uses the "Oh is that all?" dodge.  

To counter what Gallup had to say about declining Democratic voter enthusiasm while Republican enthusiasm has risen to 2004 levels, Cody countered with this link: 

http://pollingmatters.gallup.com/2012/09/the-recurring-and-misleading-focus-on.html

That's a column written by Gallup's editor in chief Frank Newport on PARTY IDENTIFICATION.  That has nothing to do with VOTER ENTHUSIASM.  Dem's have ALWAYS held a voter ID edge over Republicans.  And they still manage to lose plenty of elections in the past couple of decades.  Maybe Cody hasn't gotten around to noticing this.

Not only that, even when Republicans trail in voter enthusiasm, such as they did in 2004 when Dem enthusiasm was a sky-high 68% according to some polls [including the Gallup one I used] and Republican enthusiasm was only 51%, Bush still beat Kerry handily.  

I pointed out to Cody the link he sent had to do with party ID, not voter enthusiasm, which is what the Hope N' Change BS was based on.  He totally ignored this.  

In response to his claim MSM pollsters have no reason to skew their polls, I asked him if he had noted all the MSM polls the last 2 months or so that predicted an electorate on Nov. 6 that would be +9 Democrat or greater.  He responded by referring me back to Frank Newport's column at Gallup. 

Let's remember: 2008 was WAVE ELECTION for Democrats.  That +7 adv. they had in the vote was pretty damn big.  You have to go back to one of Reagan's landslides to find a Presidential election where one party outvoted the other by such a big honking margin.  

Cody is claiming that on Nov. 6 Democrats are going to MATCH or even EXCEED the vote advantage they achieved in 2008.  This while every poll shows Democratic voter enthusiasm has tanked dramatically since 2008.  

Go read that Frank Newport column again that Cody linked to. Is there ANYTHING in it where Newport provides evidence that we are gonna see another huge wave election for Democrats in 2012?  All he discusses is why using PARTY ID alone to predict results is just stupid. 

Here's the only point in that column Newport discusses a 10 point lead by one candidate or party, which is probably what Cody latched onto in his Google search: 

Basically, if an observer is concerned about a poll’s results, that observer should skip over the party identification question and just look at the ballot directly. In other words, cut to the chase. Don’t bother with party identification sample numbers.  Look directly at the ballot. 
For example, we know that in Ohio:
  • Obama won by 5 points in 2008
  • Bush won by 2 points in 2004
  • Bush won by 3 points in 2000

Now if a given poll in Ohio in this election shows Obama with a 10-percentage-point lead, one should just ask, “How likely is it that Obama would be ahead by 10 points if he won by five points in 2008?” -- forgetting party identification, which we assume is going to be higher for the Democratic Party if Obama is ahead, anyway. The discussion of the ballot in the context of previous ballots is, in fact, a reasonable discussion. It may be unlikely that Obama will double his margin in 2012 from what occurred in Ohio in 2008. Or maybe not.  But the focus should be directly on the ballot, and discussions of reasons why it might be different than one expects should not involve an attempt to explain the results by focusing on changes in party identification  -- which is basically a tautological argument. 

What kind of answer does Newport expect when he asks the question, "How likely is it that Obama would be ahead by 10 points if he won by five points in 2008?"  With a President who's approval rating has been underwater for over a year, and even after what the electorate did in 2010, Cody reads that and thinks Newport expects the answer: "Pretty damn likely!"

This is seeing what you want to see instead of what the person is actually saying.  Newport is pointing out if Obama won Ohio by 5 points in 2008, and Dem enthusiasm has cratered since then while Republican enthusiasm is up, how LIKELY is it that Obama is leading there by DOUBLE what he won by 4 years ago?  Anybody who thinks the answer to that questions is "Pretty much a certainty!" can't read very well.  Or has a very significant case of solipsism.  

So the 'evidence' Cody is supplying to 'prove' that Democratic vote enthusiasm hasn't declined while Republican enthusiasm has risen, and that most of the Republicans that stayed home in 2008 will also stay home in 2012 doesn't even say what he claims it does.  


Cody wasn't done yet, though.  Again, I try to get him to answer: what does HE think of MSM polls that keep projecting a +10 Democratic electorate or greater?  Since he's using an aggregated number combining all the MSM poll results along with Gallup and Rasmussen.  

I don't want to hear what he THINKS Frank Newport has to say about it, because I had already read the article Newport wrote and knew it didn't even say what Cody was claiming it did.  

Instead he RT's an earlier tweet of mine, wanting to claim I provided no evidence that Hope and Change Fever has died out.  Er, weren't we just discussing the Gallup poll on Voter Enthusiasm?  I linked the Gallup chart again.  Does it or does it not show Dem enthusiasm dropping from 61% in 2008 to 39% this July?

When the person I'm talking to on Twitter

1. Pretends I didn't offer any evidence for something when I did and
2. Links me an article to 'prove' something the article doesn't say

I realize I'm pretty much wasting my time with the discussion.  So here came the first threat of a block.  

At this point, still refusing to say what HE thought of MSM polls skewing +10 or greater to Dems nationally, he changed the subject to what an AWESOME job many Democrats think Obama is doing.  

Yes folks. over 70% of Democrats polled DO believe Obama is doing a fantastic job: 

http://www.nationalreview.com/campaign-spot/328904/voters-keep-missing-conventional-wisdoms-memo
Among all likely voters, 56 percent say the country is on the wrong track. This number has fallen because 72 percent of Democrats and 73 percent of African-Americans now say the country is on the right track. Yet two in three independents still think the country’s on the wrong track.
So yeah, this is the kind of myopic solipsism we're having to deal with here when it comes to the Obots.  

Finally Cody decides to claim I must have not read Frank Newport's article.  That's how he'll get out of having to rationally discuss Media polls that claim a +9 or greater Democratic edge in the electorate this November.  Naked appeal to authority, now stop asking me, etc. etc. 

Only as I've already showed, Newport isn't saying what Cody thinks he's saying.  Newport certainly isn't arguing for +9 or greater advantage for Democrats nationally right now.  He's says NOTHING that could possibly be translated into a low Republican turnout matching 2008.

So I press him: I want to know what YOU think of polls that say Dems will top the +7 in 2008 with a +9 or greater.  

Finally he answers & claims pollsters are helpless and have no control over how many R's, D's and I's end up in their samples.  

This is just laughably absurd.  Imagine Pew saying "Oh hey, that +19 Democrat sample we used in early August, we just couldn't find enough R's and I's to answer the phone."  After the initial round of calls they ended with +19% more Dems in the sample than R's? No problem - do more calls until you get more Republican responses.  Or you could always, you know, WEIGHT the poll to realistic proportions by dropping a lot of the Democrats.  

It never occurs to Cody why Pew and MSM pollsters will put out a poll that was already heavily skewed to Dems, and then in the WEIGHTING they do of it, INCREASE the skew by adding more Dem's and dropping Republicans.  

So no, pollsters DO 'have control' over how many D's, R's and I's end up in their samples and they CERTAINLY  have control over how they choose to weight the sample.  So this is nothing more than a dodge.  This is the 3rd time he's dodged.  I let him know if he's going to keep doing this I'll just end this conversation and block him.

Once he says I must not have read Newport's article again, the one that's not saying what he thinks it does, it's time to end this.  But not before he gets in a parting shot by linking me to Nate Silver's FiveThirtyEight blog at the New York Times which attempts to argue that there's no CONSISTENT partisan bias in Media polling...in the last 21 days leading up to the election.  

http://fivethirtyeight.blogs.nytimes.com/2012/09/29/poll-averages-have-no-history-of-consistent-partisan-bias/

As I explained in an earlier post, these MSM's pollsters won't quit the skewing games and get serious about their sample numbers actually reflecting reality until just before the election takes place.  

http://drawandstrike.blogspot.com/2012/09/why-you-wont-see-accurate-polls-from.html

That LAST poll a pollster puts out is the only one anyone will remember or really discuss.  Thus it's the only one where the pollster's reputation is REALLY on the line.  But in August, September, early October?  Will anybody remember a few years from now how Pew & others claimed they were seeing a +10 pt or greater adv. for Dem's back in September?  Not really.  

If Cody thinks Democrats will make history again on Nov. 6 2012 and match or exceed that historic turnout advantage of 2008, all I can say is I hope somebody is there to hold his hand on Nov. 7 and tell him it'll be all right.  

4 comments:

  1. You came off much worse in that exchange - like you're demented or something. This is all really getting to you, isn't it?

    ReplyDelete
    Replies
    1. Time cures most illusions. Plenty of illusions are going to die on Nov. 6. People confidently pontificating to whoever will listen that The Hope and The Change is alive and well and that the 1,000,0000's of Republicans that stayed home in '08 are going to stay there again?

      I'll be here to hold their hands and tell them it'll be OK.

      Delete
  2. "...all I can say is I hope somebody is there to hold his hand on Nov. 7 and tell him it'll be all right."
    You're a better man than I, Brian. I'll be down on the sidewalk chanting "Jump! Jump! Jump!..."

    ReplyDelete
  3. Complete and utter fail.

    ReplyDelete