Meta-Blog

SEARCH QandO

Email:
Jon Henke
Bruce "McQ" McQuain
Dale Franks
Bryan Pick
Billy Hollis
Lance Paddock
MichaelW

BLOGROLL QandO

 
 
Recent Posts
The Ayers Resurrection Tour
Special Friends Get Special Breaks
One Hour
The Hope and Change Express - stalled in the slow lane
Michael Steele New RNC Chairman
Things that make you go "hmmmm"...
Oh yeah, that "rule of law" thing ...
Putting Dollar Signs in Front Of The AGW Hoax
Moving toward a 60 vote majority?
Do As I Say ....
 
 
QandO Newsroom

Newsroom Home Page

US News

US National News
Politics
Business
Science
Technology
Health
Entertainment
Sports
Opinion/Editorial

International News

Top World New
Iraq News
Mideast Conflict

Blogging

Blogpulse Daily Highlights
Daypop Top 40 Links

Regional

Regional News

Publications

News Publications

 
Polls all over the place
Posted by: McQ on Monday, November 06, 2006

Inveterate poll watcher Dick Morris says "bloodbath":
John Zogby's polling is tracking 15 swing House districts, and he finds Democratic leads in 13. Since Dems need only 15 to take control - and will doubtless pick up several not on Zogby's list - it seems we're in for several years of Speaker Nancy Pelosi.

In the Senate, only Tennessee seems to be holding for the Republicans. (There's no justice: Rep. Harold Ford Jr., the Democrat now losing to Republican Bob Corker, is the best of the crop of Democratic challengers).

Other Senate races? Ohio, Pennsylvania and Rhode Island are gone for the GOP. Republican fantasies of a rebound in Montana are falling short. It's down to Missouri and Virginia: Democrats need to win both to take control. In these states, pollster Scott Rasmussen has GOP incumbents Jim Talent and George Allen below the 50 percent mark - usually a sign of doom. Rasmussen has Allen tied with Democrat Jim Webb at 49 percent and Democrat Claire McCaskill ahead of Talent 49-48.

Even with those nail-biters too close to call, 2006 will go down in history as one of the worst years for the Republicans.
Quin Hillyer is of the opinion that Republicans will hold on:
Let's examine the Senate first, because it is easier to understand. It boils down to this: Democrats are trying to seize seats from states that traditionally vote Republican. Even in a bad year for Republicans, it is difficult for Democrats to overcome the triple advantages of incumbency, superior fundraising, and a population that usually leans rightward. Not only that, but the national economy — which, when it is strong, usually boosts incumbents tremendously — is arguably the strongest in the entire history of the world.

On the presidential level, Virginia and Montana almost always vote Republican, Missouri and Ohio usually do, and Tennessee does so more often than not. Pennsylvania, a state that tends to lean only a little leftward, has not elected a Democrat in a regularly scheduled Senate election since the 1970s. And in Rhode Island, a very liberal state, Democrats must overcome a habit of supporting three different generations of the (liberal Republican) Chafee family for statewide office.

Meanwhile, Democrats this year are defending several seats of their own that, except for the national anti-Republican trend, feature unique circumstances that should make them nervous.

Maryland has been trending a bit more conservative anyway, and this is the second straight election cycle where Maryland black leaders are expressing a serious dissatisfaction with the Democratic Party. With Republican nominee Michael Steele being a highly charismatic black man, Democrats have reason to fear his inroads into a normally Democratic constituency.

Washington state and Michigan, meanwhile, feature two of the least accomplished, least powerful members of the Senate — respectively, Maria Cantwell and Debbie Stabenow. Many Washington state voters are still sore that the Democrats snatched the governorship two years ago in a still-disputed election, and many Michiganders are unhappy with what is arguably the worst state economy in the whole nation, which has occurred under Democratic state elected leaders.
He discusses the House as well.

In both cases you can find polls which support the premise Morris and Hillyer each expound.

Why the disparity in these polls?

Billy Hollis has an interesting hypothesis:
I’ve noted that the polls in the last few elections have almost all tended to start out favoring Democrats, and then veer Republican. I think one potential explanation is the element of self-selection in today’s polls.

We all know Internet polls, in which the respondents are self-selected, don’t really tell us much about how the population at large feels. But it’s clear to me that even professional polls have some degree of self-selection by the respondents, because of the capability to opt out. People can avoid pollsters via caller ID, or refuse to take a poll, or whatever. Some people use mobile phones as their primary phone at least partially because unsolicited calls from telemarketers and pollsters are very rare on them.

Pollsters have noted the increasing turn-down rates they’ve been getting on calls for polls. So the element of self-selection has becoming more prevalent over in recent years.

Now here’s the hypothesis. Citizens who are dissatisfied with the those currently in power are, I think, more likely to respond to pollsters (not opt out) earlier in the election cycle. Thus they might have a tendency to be over-represented in early polls.

As the election nears, those who are more complacent about the party in power become more interested in the election and less likely to opt out, and to express a more definite opinion when they do take polls. It might be that the pollsters just can’t get these folks on the phone as much as they’d like early in the election season, but they can later, and that accounts for the drift in favor of Republicans.

Now I realize that the pollsters do the best they can to balance out their samples. But the more "self-selection" (because others opt out) is going on, the harder it is to create such samples. Maybe they just have not found a way in their methodologies to correct for such a effect that varies over time

Just an idea.
Thoughts?
 
TrackBacks
Return to Main Blog Page
 
 

Previous Comments to this Post 

Comments
I tend to agree with Billy’s hypothesis, as polls tend to fall into the same trap as Shere Hite did in the 1970’s. Her work is still considered textbook material for most statistics courses and is filed under the heading, "What Not To Do." I suspect pollsters could learn some valuable lessons from her badly flawed methodology.
 
Written By: The Poet Omar
URL: http://www.asecondhandconjecture.com
Two thoughts regarding polling...

1. How do they choose who they will call? I have been a registered voter in GA for 12 years and have only been polled twice. Once in 96 and once 3 weeks ago. None of my friends have ever been polled. It just seems strange.

2. When I was polled 3 weeks ago, the questions and the answers were pathetic. Odd questions with only A or B for available answers. My desired response never matched up with the options.

Anyways.... I know a bunch of people who’ve never been polled and they will be voting R or Libertarian. So the only poll that’s gonna figure that out is the final tally on Tuesday night.
 
Written By: meagain
URL: http://
Hillyer is engaging in pure wishful thinking. Santorum and Dewine are toast—not even breaking 40% in some polls (and Ohio is going to be a bloodbath for Republicans this year—check out the governor’s race) Burns and Chafee are well below 50%.

Allen has run a pitiful campaign (your colleague JH won’t be proud to have that one on his resume) and is also below 50%. Word is that he neglected his GOTV apparatus—which is a big problem considering that the Democrats have a very good ground game in Virginia.

It’s like he hasn’t even paid attention to the race for the past two years.

He’s essentially using ideology as a proxy for fact-oriented analysis.
 
Written By: Geek, Esq.
URL: http://
Dick Morris’s article has nothing to do with an honest assessment of the polling data. His sole purpose is to persuade the base to get out and vote GOP, or the terrorists win. He is a hack. His real point is in this passage:
What’s ahead for the next two years? Not new legislation so much as investigations, subpoenas, hearings etc. Washington will be as effectively paralyzed as it was during President Clinton’s impeachment trial. And, let us remember that it was in that incubator that Osama bin Laden was able to plan the 9/11 attacks.
 
Written By: Mona
URL: http://inactivist.org/
I’ve never been polled either.

Two reasons, AFAICT:

1) I live in Carrollton, Texas; a suburb of Dallas. Pretty conservative place; I don’t think our Congresscritter’s re-election is in much doubt. (I’m voting anyways.)

2) If I don’t recognize the number on the phone, it goes to the machine. Period.
 
Written By: Mark A. Flacy
URL: http://
I don’t see any assessments which I would call particularly convincing since you can indeed find a poll somewhere to support it (as well as polls which will dispute it). I’m inclined to give Billy’s hypothesis some credibility since I think it is getting harder and harder to poll. That’s why in many cases you’ll see X number of "adults" instead of "likely voters" or whatever as it is harder and harder to get in touch with likely voters.

I’m really looking forward to comparing poll results with actual results in this election.
 
Written By: McQ
URL: http://www.qando.net/blog
There’s actually empirical data indicating that Republicans are late deciders, hence the historical overperformance in the generic balloting. This probably accounts for some of what is discussed above.

The problems with polling, though, are real. Last cycle, much speculation was given to the effect of cellphones, as households increasingly abandoned landlines. My cell, for example, has a (202) area code; I’ll never be polled by random digit dialing for a Virginia race. Unlisted numbers, or alternative numbers like Vonage (where you don’t automatically go in the phone book) are also problematic — we have Vonage and aren’t listed; if our name shows up on the voter rolls the company won’t be able to find our phone number. This can have the effect of biasing the samples for polls, though it is unclear exactly how this biases the samples.

My sense is that we aren’t to the point where this is a massive problem yet, but we are getting there. As flawed as it is, Zogby’s internet research is an important experiment. If he can perfect it, there may be hope for polling yet.
 
Written By: Sean
URL: http://www.myelectionanalysis.com
Forgot:

Regardless, we heard a lot of these theories in ’04, and the final polling turned out to be more-or-less correct, especially on the race-by-race level.
 
Written By: Sean
URL: http://www.myelectionanalysis.com
Dick Morris’s article has nothing to do with an honest assessment of the polling data. His sole purpose is to persuade the base to get out and vote GOP, or the terrorists win. He is a hack. His real point is in this passage:
Yeah, it’s not like he ever worked for a Democrat, or anything like that.
 
Written By: Don
URL: http://
Sure, but all polls and surveys are self-selecting samples.

 
Written By: rightwingprof
URL: http://rightwingnation.com
If someone is watching Zogby....they may as well be watching a magic-eye picture to see what pops out
 
Written By: shark
URL: http://
What’s ahead for the next two years? Not new legislation so much as investigations, subpoenas, hearings etc. Washington will be as effectively paralyzed as it was during President Clinton’s impeachment trial. And, let us remember that it was in that incubator that Osama bin Laden was able to plan the 9/11 attacks
.

I agree with you Mona! Clinton willfully perjuring himself led to 9/11. I always knew he had blood on his hands and now you confirm.

Thanks!
 
Written By: shark
URL: http://
Inveterate poll watcher Dick Morris
Are you trying to say Dick Morris has no backbone?

[/undereducated and stuck in Irak]
 
Written By: JWG
URL: http://
I wonder if the reliability of telephone polling has changed since caller ID was introduced.
 
Written By: Dave Schuler
URL: http://www.theglitteringeye.com
In 2004, the theory was that all of us under-40’s who vote Democratic would be underrepresented because we only use cell phones.

I’ll believe in allegations that the polling is distorted towards one party or the other when they’re actually born out.
 
Written By: Geek, Esq.
URL: http://
Now I realize that the pollsters do the best they can to balance out their samples.
Really? How come they consistently oversample Democrats even though Republican registration has pulled even and in some cases ahead? Why do pollsters ask misleading questions?

I have no doubt that private polls are "doing their best" because they want to get an accuracte picture. But the MSM polls have no such interest. Their sole goal is to buck up Democratic attitudes.
 
Written By: SaveFarris
URL: http://
So Mona, you’re saying that the Democrats won’t advance the GWoT but merely focus on "payback?" Now isn’t this the Party you want McQ to join... "Vote for Us. We won’t DO anything, but we will destroy the war effort!" That’s a winning slogan....You’re Mona HAYDEN right, Tom’s sister?
 
Written By: Joe
URL: http://
Are you trying to say Dick Morris has no backbone?
Not that I know of ... inveterate.

(Yeah, I know you were joking JWG, but I thought I’d take the opportunity to highlight the word).
 
Written By: McQ
URL: http://www.qando.net/blog
Yeah, it was a good word.

I liked the example for the definition:

an inveterate liar
 
Written By: JWG
URL: http://
Mark F - you have to be kidding....Carrollton?
There’s an interesting statistical anomoly in here somewhere.
Between Rosemeade and Hebron, Old Denton and Josey.

But I have been polled, last election.
My feeling was the poll certainly slanted for the ’desired’ response.

 
Written By: looker
URL: http://
Dick Morris, like Joe Lieberman and many others such as Roger Simon, switched parties after 9/11. He was a hack before, he is a hack now. The NY Post doesn’t publish non-GOP hacks, except for fun.
 
Written By: Mona
URL: http://inactivist.org/
Dick Morris, like Joe Lieberman and many others such as Roger Simon, switched parties after 9/11. He was a hack before, he is a hack now.
Was Morris a Democrat in the ’90s? Prior to Clinton, he worked for Republicans, and he was a moderate voice in the Clinton White House, advocating signing popular Republican legislation.

The simple fact is that Morris is in the middle. He’s only a hack for himself. He’s no GOP hack, although he may currently lean GOP.

For a Libertarian, Mona sure has a lot of interest in shrinking the Democrat tent. In another place and time, she would have been working a guillotine, I suspect.

 
Written By: Don
URL: http://
Don, you must not understand the proper filter to interpret today’s political analysis. If someone ventures an opinion and it is neutral or favorable to Republicans, they’re obviously doing it because they’re paid to, or from other base motives. However, if they say something favorable about Democrats, they’re clearly doing so from a stance of complete objectivity. That’s just the way things work today, ya know?
 
Written By: Billy Hollis
URL: http://
He’s no GOP hack, although he may currently lean GOP.
No! Really? He (the hack) favors the GOP now— REVELATION.
 
Written By: Mona
URL: http://inactivist.org/
Is polling more accurate in nations like Australia that have compulsory voting?
 
Written By: Harun
URL: http://
Dick Morris, like Joe Lieberman and many others such as Roger Simon, switched parties after 9/11. He was a hack before, he is a hack now. The NY Post doesn’t publish non-GOP hacks, except for fun.
Great analysis, Mona. Thanks.
 
Written By: McQ
URL: http://www.qando.net/blog
For a Libertarian, Mona sure has a lot of interest in shrinking the Democrat tent. In another place and time, she would have been working a guillotine, I suspect.
Don - are you suggest Mona is a variation on Mel Brook’s "Madam DeFarge"?
 
Written By: looker
URL: http://

 
Add Your Comment
  NOTICE: While we don't wish to censor your thoughts, we do blacklist certain terms of profanity or obscenity. This is not to muzzle you, but to ensure that the blog remains work-safe for our readers. If you wish to use profanity, simply insert asterisks (*) where the vowels usually go. Your meaning will still be clear, but our readers will be able to view the blog without worrying that content monitoring will get them in trouble when reading it.
Comments for this entry are closed.
Name:
Email:
URL:
HTML Tools:
Bold Italic Blockquote Hyperlink
Comment:
   
 
Vicious Capitalism

Divider

Buy Dale's Book!
Slackernomics by Dale Franks

Divider

Divider