iPhone app iPad app Android phone app Android tablet app More

Featuring fresh takes and real-time analysis from HuffPost's signature lineup of contributors
Stephen Herrington

Stephen Herrington

Posted: October 12, 2010 10:41 AM

The "enthusiasm gap" is driving the midterm election narrative. You hear, everywhere from every MSM and polling source, that the generic ballot shows a 5-10 point lead for the GOP. That lead is constructed entirely on the "likely voter" model. Among likely voters, you hear, the GOP has a 5-10 point lead. Have you ever heard the results of polls of registered voters not qualified by the "likely" modifier?

In the latest CNN public opinion poll released 10/08, someone slipped up and mentioned that, among registered voters, the generic Democrats have a 6 point lead. A 6 point lead is about what it took to sweep Congress and the White House for the Democrats in 2006-08. So the sentiment of the country favors Democrats by 6 percent and the press reports only the pollster construct of right leaning likely voters as representing the sentiment of the nation.

What is a likely voter? Different polls have different methods and most are proprietary. Gallup has shared their battery of questions used to determine a likely voter. From Mystery Pollster :

1) How much have you thought about the upcoming elections for president, quite a lot or only a little? (Quite a lot = 1 point)


2) Do you happen to know where people who live in your neighborhood go to vote? (Yes = 1 point)

3) Have you ever voted in your precinct or election district? (Yes = 1 point)

4) How often would you say you vote, always, nearly always, part of the time or seldom (Always or nearly always = 1 point)

5) Do you plan to vote in the presidential election this November? (Yes = 1 point) [presumably updated for 2012]

6) In the last presidential election, did you vote for Al Gore or George Bush, or did things come up to keep you from voting?" (Voted = 1 point) [presumably updated for 2010-12]

7) If "1" represents someone who will definitely not vote and "10" represents someone who definitely will vote, where on this scale would you place yourself? (Currently 7-10 = 1, according to this "quiz" on USA Today)


It's easy to see that the questions used to determine a likely voter skew in favor of older people and those with long term political involvement. Note that only one question, number 7, directly asks if you plan to vote and voting history is the most weighted measure by far. What is apparent in the battery is that complacency is judged an immutable pattern of the U.S. voter, 2006-08 notwithstanding for some reason.


Fair reporting on the subject of this current election would be that the press has decided that the same American public that voted out the GOP in 2006-08 is intent on putting them back in office in 2010. Despite the tribulations of the first two years of Obama, Americans seem still in favor of Democratic rule, but their "enthusiasm" is measured by only two of the Gallup battery questions, numbers 1 and 7. Question 1 can be interpreted as having not had to think about the election too hard in order to make a choice. Question 7 could turn on the notion that if things appear to be in hand, then I won't bother. Rather gaping holes in both questions for the polls and press to rely entirely on them as a litmus of political fact. Rather thin criterion for the claim that America is a center right nation, these two questions, but that is what the MSM has made of it in practical terms.

News organizations are depriving the public of a vital piece of information that is significant. That piece of information is that the public actually supports the Democrats more than it does the Republicans, just maybe not enough to bother with voting. Only reporting the horse race aspect of the election, which party's voters are most likely to show up by historical statistical standards, serves to skew the perception that America is leaning right when it's not. Issue polls bear this out, America is the most left on issues that it has been since the sixties. What comparing the likely voter and registered voter numbers show is a GOP willing to use big money enabled anger to impose its will on a public that, in majority, doesn't want it to.

Because of this distortion in reporting poll outcomes, some on the left will stay home or vote a third party because they think the Republicans will win anyway. The tragedy of this is that the "unenthused" left majority will suffer the all too predictable pains of a GOP Congress legislating against their interests and wishes. The public wishes government to be left of center and not right, according to the polls.

Democrat are by avocation democratic, so if the majority is leaning politically right they are more likely to accept a right wing future than Republicans are to accept a politically left leaning future. The right is just always more enthused about forcing everyone to live according to their rules than is the left. But then, actual enthusiasm is not a central component of the likely voter calculation, history is.

Pollsters always push the "likely voter" qualification because it is intuitively the more exact predictor. It is an exact predictor until "enthusiasm" swings, right up to and including the day of the election. It is the old "if the vote were held today" codicil to predictions that allows them to predict in polling at all. But the pitfall of that is a misrepresentation of the political landscape that the press and polls undertake at the risk of being proved wrong, like the market crash of 2008 propelled Obama to a 6 point lead and win out of a virtual tie. Happily for the press and pollsters, the public has the long term memory of a mayfly (lifespan one day).

Conspiracy theorists might jump to the conclusion that the conservative owned press is willfully misrepresenting the public temperament. More likely is that the American press has just become too lazy and arrogant to care whether what they are fed by pollsters who should know what is happening in fact do not know what is happening, or even whether they are being fed propaganda crap by them. The drumbeat of the Democrats being beat like a drum is everywhere that Fox News can feed a tag line. The somnambulism of the left, having won something but not so much that it appears to have upset business as usual, is innervated further by dire predictions of losing even what has been accomplished. Time for the left to surrender, according to Fox.

Fox, CBS, ABC, NBC, CNN and Republicans, listen up and listen tight, surrender is not a word any American takes lightly.

 
The "enthusiasm gap" is driving the midterm election narrative. You hear, everywhere from every MSM and polling source, that the generic ballot shows a 5-10 point lead for the GOP. That lead is cons...
The "enthusiasm gap" is driving the midterm election narrative. You hear, everywhere from every MSM and polling source, that the generic ballot shows a 5-10 point lead for the GOP. That lead is cons...
 
 
  • Comments
  • 73
  • Pending Comments
  • 0
  • View FAQ
Comments are closed for this entry
View All
Favorites
Recency  | 
Popularity
Page: 1 2 3  Next ›  Last »  (3 total)
HUFFPOST SUPER USER
WilliamGossett
04:19 PM on 10/15/2010
I hope this article is right.

Have noticed a swing towards push-polls these days which are skewed to favor the GOP. It is an effort to scare away the Dem voters.

I think the FL low polls for Kendrick Meek are accurate. He was a poor choic e for the Dems.
12:50 PM on 10/15/2010
It used to be that when you took a statistics class in college, one of the first demonstrations of how statistics could be skewed was when the lecturer would statistically show that storks deliver babies.

As referenced in the article, polls are only as good as the questions asked. I have not seen very many good polls out there (that is as reflected by the questions asked) (and I have responded to many polls in the past). Too many questions are either leading or too general (leaving the results open to question).

Ignorance is expensive. It is so important to our country that our citizens be educated and informed and not afraid of thinking for themselves.
photo
HUFFPOST COMMUNITY MODERATOR
tacevad
American SS Card Carrying Socialist
09:14 AM on 10/15/2010
the only poll that really counts...the voting booth.
07:58 PM on 10/14/2010
It'd be amusing to see the Huffalot Post presume to differentiate itself from the mainstream media if there weren't so much at stake. Not that the Dems are any great shakes, but the [ahem] elephant in the room is the widespread election fraud that has become standard in our country. And THAT'S what's not being reported-- in this article, either, where it most definitely belongs. http://bradblog.com + http://blackboxvoting.org + http://electiondefensealliance.org/ + richard hayes phillips, bob fitrakis, clint curtis, death of Michael Connell, Stephen Spoonamore, Smartech: the evidence is there. Why aren't YOU (PuffHost) reporting it? Why do we only hear about it when it happens in Afghanistan, Mexico, Ukraine, Iran, etc etc? What's your excuse for ignoring the mountains of evidence presented by Marc Crispin Miller? The real scandal is that the 2004, 2006, and 2008 were LANDSLIDES for Democrats, but millions of votes were stolen. Those are the polls that matter, not the manipulative ones asked of a manipulated public and then manipulated to "conclude" whatever has already been decided it will say. here's a fresh, steaming hot pile of election fraud: http://electiondefensealliance.org/believe_it_or_not_MA_senate
12:16 AM on 10/26/2010
I remember going through election records in LA County and seeing evidence of more votes than voters.
I have seen the evidence of more votes than voters in Ohio in 2004.

So when the "polls" tell me that people are too dumb to remember who got us into this mess, and want to vote the surrogates of those rich $ucks back in, it DOES ring an alarm bells that they are arranging cover for another heist.
06:59 PM on 10/14/2010
The teabaggers have helped the democrats tremendously. They scare sane people away.
04:27 PM on 10/14/2010
Want another few percentage points back from poll scewing? Take into account that approximately 15% of households only use cell phones, while polls only call landlines. These households are typically minorities and younger voters, demographics who mostly vote for Democrats.
01:58 PM on 10/14/2010
Good article.

I would like to see someone compare November detailed election results patterns (vote shares broken out by ballot type - absentee, precinct,..) with the likely voter (LV) and registered voter (RV) polls available just prior to election day in close contests (say within 10% margin). Then compare these polls and election results patterns with the extent to which election outcomes are publicly verifiably accurate in each contest's election jurisdictions.

This would reveal several types of anomalous patterns produced by any fraudulent vote manipulation that could occur in the myriad of US jurisdictions not conducting sufficient post-election audits of all ballot types or using inauditable, unaccountable, hackable, unverifiable voting systems, if the fraudulent manipulation occurs in only some types of ballots. The pattern variations could be compared to contests that were not close (i.e. not a likely target for manipulatuin) to see if the same anomalies occurs there.

A short list of conditions required for producing publicly verifiably accurate election outcomes:

http://kathydopp.com/wordpress/?p=174

Unfortunately no US states meet all the necessary conditions, but some are a lot better than others.
12:48 PM on 10/14/2010
Invisible 6 point lead?

Wizard of OZ ? Alice in Wonderland ?

Ha Ha
photo
HUFFPOST SUPER USER
frank day
Obama cares about all of U.S.
11:59 AM on 10/14/2010
I look for a big turnout on Nov2nd.
That alone will derail the Republican effort.
Demographics have shifted, the pollsters aren't reflecting it.
Look for people to get more interested in the week running up to the election.
photo
dakotawoman
I dreamed I saw Joe Hill. . .old time Progressive
11:58 AM on 10/14/2010
Can't it work to Dem advantage?

If liberals believe that we are losing, might it not encourage us to try harder and get to the polls early?

And, conversely, might thinking that they are ahead make cons complacent themselves so that they stay away?
HUFFPOST PUNDIT
reveal5
12:15 PM on 10/14/2010
Good questions.
photo
HUFFPOST SUPER USER
FogBelter
Illegitimis non carborundum
02:32 AM on 10/14/2010
Polls are a lot of things but don't represent actual voting on Election Day, so at their best polls are a snapshot in time, at their worse they are a tool of social engineering.
photo
George Hanshaw
There are none so blind as those who will not see.
01:57 AM on 10/14/2010
Wish in one hand, crap in the other - then see which hand gets full the fastest......
photo
HUFFPOST SUPER USER
Doug Watt
Not ready for 2012
01:54 AM on 10/14/2010
Thanks for this excellent post, I'll spread it around.
photo
HUFFPOST COMMUNITY MODERATOR
tacevad
American SS Card Carrying Socialist
09:13 AM on 10/15/2010
the timing of your post right after George Hanshaw's "wish in one hand" one is just about priceless:)
photo
HUFFPOST SUPER USER
Doug Watt
Not ready for 2012
01:56 PM on 10/15/2010
HAH!!! I didn't see that, thanks!
photo
HUFFPOST COMMUNITY MODERATOR
Callyson
Trying to come up with a new creative microbio
01:04 AM on 10/14/2010
This is all the more reason not to give in to despair, which is just what the Reeps want us to do (and is what the trolls' comments are trying to make us do.)
Phonebank! Walk precincts! Talk to your friends and neighbors!
And--no surrender!
11:25 PM on 10/13/2010
http://richardcharnin.com/2010ElectionForecastModels.htm
T
The House and Senate forecast models provide a comprehensive analysis of Registered Voter (RV) and Likely Voter (LV) polls.

The Senate model employs simulation analysis of the latest RV and LV polls to forecast average GOP net gains, associated win probabilities and trends. The built-in sensitivity analysis displays the effects of various undecided voter allocation and vote-switching scenarios.

The House model provides a summary comparison of the latest RV and LV Generic polls, win probabilities and a moving average projection. As in the Senate model, the sensitivity analyses displays the effect of various undecided voter and vote-switching assumptions on forecast vote shares, House seats and win probabilities.

Democrats always do better in the full RV sample than in the LV sub-sample. LV polls exclude millions of registered voters who actually vote - and most of them are Democrats.

Polling websites generally display only Senate LV polls. CNN/Time has provided both RV and LV samples, but only the LVs are listed at realclearpolitics.com. The Senate RV model forecast model is therefore a mix of RV and LV polls. Without a corresponding RV poll for every LV sub-sample, a comparable analysis is difficult.

Unlike the Senate, House Generic polls have been primarily RV samples (except for Rasmussen which only provides LV sub-samples). But the ratio will shift to virtually all LVs as Election Day approaches.