logo 
spacer
  

Welcome Guest ( Log In | Register )

If you have an opinion, you should share it! Register Now!

America's Debate hosts the best in news, government, and political debate. Register now to take part in the most civil and constructive debate on the Internet. Join the community, and get ready to be challenged!

Click here to start

> Sponsored Links

Register to remove these ads!
> "This Election Is Already Over - Obama Has Won"
akaCG
post Oct 2 2012, 01:40 AM
Post #1


*********
Advanced Senior Contributor

Sponsor
August 2012

Group: Sponsors
Posts: 4,846
Member No.: 10,787
Joined: November-25-09

Gender: Male
Politics: Conservative
Party affiliation: Independent



So declares Mr. Cenk Uygur (whose former MSNBC time slot is currently filled by Al Sharpton; who's currently hosting a show at Al Gore's CurrentTV) in a HuffPo piece today:
QUOTE
...
There's another poll out today showing President Obama with a nine point lead in Ohio. That's the fifth poll in a row showing him with a larger than a five point lead. The Quinnipiac University/CBS News/New York Times poll that came out last week had him with a ten point lead.

No Republican has ever won the presidency without winning Ohio. Plus, whoever has won Ohio has won the last 11 presidential races. Ten point leads aren't small, they're gigantic.
...
It's over. President Obama will get re-elected.
...

Link: http://www.huffingtonpost.com/cenk-uygur/o..._b_1927955.html

The first Ohio poll Mr. Uygur refers to is the one issued by The Columbus Dispatch, which shows President Obama up by a whopping 9%. Wow! That's just about twice the margin by which Obama beat McCain in 2008 in Ohio (4.6%)! Almost double wow!

Except that, when you take a look at said poll's internals, you find:

"Party ID" sampling: Dem + 8.8% (which is 5.2% higher than Gallup's party ID breakdown for the state: Dem + 3.6%).
"Voted for Obama in 2008" sampling: 52.6% (which is 1.1% higher than his Ohio result in 2008: 51.5%)
"Voted for McCain in 2008" sampling: 40.9% (which 6.0% lower than his Ohio result in 2008: 46.9%)

The "Is there a union member in your household" sampling seems Obama-friendly as well. I can't determine whether it's by just some or by A LOT, I'm afraid. But, given that union membership among the state's wage and salary workers was (according to BLS stats) 13.4% in 2011, ...

The second Ohio poll Mr. Uygur refers to is the one issued by Quinnipiac University/CBS News/New York Times last week, which showed President Obama up by an even more whopping 10%! Again, wow! That's more than twice the margin by which Obama beat McCain in 2008 in Ohio (4.6%)! Doubleplus wow!

Except that, again, when you take a look at said poll's internals, you find, for instance:

"Party ID" sampling: Dem + 9% (which, again, is 5.4% higher than the aforementioned Gallup's party ID breakdown for the state: Dem + 3.6%; and, of course, again, more than double Obama's margin in 2008.).

So, ...

Debate question:

Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?

Go to the top of the page
 
+Quote Post
 
Start new topic
Replies (1 - 17)
lo rez
post Oct 2 2012, 02:31 AM
Post #2


******
Senior Contributor

Group: Members
Posts: 388
Member No.: 10,616
Joined: August-8-09

From: Minneapolis, MN
Gender: Male
Politics: Liberal
Party affiliation: Independent



QUOTE
Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?


heavens, I would never trust someone so closely tied to Al Sharpton and Al Gore
Go to the top of the page
 
+Quote Post
akaCG
post Oct 2 2012, 03:30 AM
Post #3


*********
Advanced Senior Contributor

Sponsor
August 2012

Group: Sponsors
Posts: 4,846
Member No.: 10,787
Joined: November-25-09

Gender: Male
Politics: Conservative
Party affiliation: Independent



QUOTE(lo rez @ Oct 1 2012, 10:31 PM) *
QUOTE
Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?

heavens, I would never trust someone so closely tied to Al Sharpton and Al Gore

Why, for "heavens" sake, would you not trust someone like that?

Or, if one were to accurately assess your response as a sarcastic one, ...

Why, for "heavens" sake, would someone trust someone like that?

ps:
A response on your part that contains more words than the modern world's version of ancient Spartan "laconicism" (Twitter's 140 characters rule) allows for would be quite the refreshing change. Then again, I must say, a response on your part that contains exactly 141 characters (including spaces, of course) would be, like, ... awesome! In "points for style" terms, at least. biggrin.gif

ps2:
And, while you're at it, you could actually, like, actually answer the debate question and stuff. That would be, like, doubly awesome (not just style, but substance points as well)! biggrin.gif



This post has been edited by akaCG: Oct 2 2012, 03:50 AM
Go to the top of the page
 
+Quote Post
Paladin Elspeth
post Oct 2 2012, 04:02 AM
Post #4


*********
I want the 10th Doctor for President!

Sponsor
August 1, 2003

Group: Sponsors
Posts: 4,987
Member No.: 721
Joined: May-10-03

From: Between 2 Great Lakes
Gender: Female
Politics: Liberal
Party affiliation: Democrat



Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?

I don't buy it. Cenk Uygur has been wrong before. He can say what he likes, obviously, but the outcome of the election is still up to the voters (or the Electoral College). The last thing we need is voter complacency since one pundit has declared it a done deal. dry.gif

As I've said before (and the analogy is far from original), a poll is like a snapshot. The people in it move and go on with their lives. A blonde may dye her hair red. Children grow up. Change happens. I don't think it looks good for Romney, but things always change.

This post has been edited by Paladin Elspeth: Oct 2 2012, 04:20 AM
Go to the top of the page
 
+Quote Post
Curmudgeon
post Oct 2 2012, 04:22 AM
Post #5


********
I am an unpaid protester!

Sponsor
August 1, 2003

Group: Sponsors
Posts: 1,189
Member No.: 729
Joined: May-14-03

From: Michigan
Gender: Male
Politics: Liberal
Party affiliation: Democrat



Over the years, I would say a majority of the people who I have heard expound on their view of politics agree on one thing: "Politicians lie."

If the voters don't trust the politicians, why should the politicians trust the voters to "Tell the truth, tell the whole truth, and nothing but the truth."

If you believe that Fox News, Chris Christie, a foolish old man arguing with an empty chair, Paul Ryan and his running mate are all telling you the whole truth when they say that they have a plan to take your money and buy this election; please donate to their campaign, watch their ads and tell yourself that you invested well while you sip a cold drink and watch the news on Election Day. (After all, if you believe that your money can purchase an election, why would you bother to actually vote?)

In the meantime, I would encourage anyone who feels that the Ryan/Romney slate is out of touch with the generic population to cast their vote for Barack Obama. He at least, is a known factor.

The election is not over until the polls close on Election Day, and while the actual results may take time to count after the polls close... I am reasonably certain that once again, on Election Day, exit pollsters will tell us who won hours before official results are certified. After all, if an exit pollster tells us that: "I called the Election once Ann Romney told me that she had voted straight Democratic and filed for divorce." Who would you believe: the pollster or "Ann Romney?"

The main reason that we really trust the Polling companies is that they somehow manage to produce a working model and a close prediction. Pollsters that fail to do that really will have no customers except candidates who want to raise campaign funds, and then who could trust them?

Just as an afterthought akaCG, since you're so good at analyzing the polls, deciding who can be trusted, etc... Why don't you assemble a model to start predicting the outcome of the 2016 Election and start selling your prediction now to the prospective backers of that candidate. Their just might be a few dollars to be made there...

This post has been edited by Curmudgeon: Oct 2 2012, 04:24 AM
Go to the top of the page
 
+Quote Post
lo rez
post Oct 2 2012, 01:15 PM
Post #6


******
Senior Contributor

Group: Members
Posts: 388
Member No.: 10,616
Joined: August-8-09

From: Minneapolis, MN
Gender: Male
Politics: Liberal
Party affiliation: Independent



QUOTE(akaCG @ Oct 1 2012, 10:30 PM) *
QUOTE(lo rez @ Oct 1 2012, 10:31 PM) *
QUOTE
Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?

heavens, I would never trust someone so closely tied to Al Sharpton and Al Gore

Why, for "heavens" sake, would you not trust someone like that?

Or, if one were to accurately assess your response as a sarcastic one, ...

Why, for "heavens" sake, would someone trust someone like that?

ps:
A response on your part that contains more words than the modern world's version of ancient Spartan "laconicism" (Twitter's 140 characters rule) allows for would be quite the refreshing change. Then again, I must say, a response on your part that contains exactly 141 characters (including spaces, of course) would be, like, ... awesome! In "points for style" terms, at least. biggrin.gif

ps2:
And, while you're at it, you could actually, like, actually answer the debate question and stuff. That would be, like, doubly awesome (not just style, but substance points as well)! biggrin.gif


I don't know nearly enough about statistics to comment but what I do know is that if I was making an argument that this is faulty poll weighting, I probably would want to stick with that rather than slapping a huge red herring into the first sentence of my argument.
Go to the top of the page
 
+Quote Post
akaCG
post Oct 2 2012, 04:41 PM
Post #7


*********
Advanced Senior Contributor

Sponsor
August 2012

Group: Sponsors
Posts: 4,846
Member No.: 10,787
Joined: November-25-09

Gender: Male
Politics: Conservative
Party affiliation: Independent



QUOTE(lo rez @ Oct 2 2012, 09:15 AM) *
...
I don't know nearly enough about statistics to comment but what I do know is that if I was making an argument that this is faulty poll weighting, I probably would want to stick with that rather than slapping a huge red herring into the first sentence of my argument.

My mention of Cenk Uygur's professional credentials/associations is not a "red herring" distraction. It is integral to enabling those here who may not know who Mr. Uygur is to understand the ideological/partisan motivations that drove him to triumphantly/categorically declare the race over on the basis of deeply flawed polls.

And one need not have a sophisticated understanding of statistics in order to determine that said polls are indeed deeply flawed. All one needs is plain common sense, which says that if you end up with a sample that has 1% more Obama voters than he got in '08 and 6% fewer McCain voters than there were in '08, you're gonna end up with a result that has no relationship to reality.

Garbage in, garbage out.

Another prime example of which is the THE WMUR GRANITE STATE POLL that just came out, which has Obama with a mind-boggling 15 point lead:

"Party ID" sampling: Dem +9 (which is 10.5% higher than Gallup's party ID breakdown for the state: Rep + 1.5%)

"Voted for Obama in '08" sampling: 60% (which is 5.9% higher than his New Hampshire result in '08: 54.1%)
"Voted for McCain in '08" sampling: 40% (which is 4.5% lower than his New Hampshire result in '08: 44.5%)

Links:
http://www.wmur.com/blob/view/-/16809026/d...ential-poll.pdf
http://www.gallup.com/poll/156437/Heavily-...ted-East.aspx#2
http://en.wikipedia.org/wiki/United_States...on,_2008#Result

Go to the top of the page
 
+Quote Post
nighttimer
post Oct 2 2012, 05:15 PM
Post #8


*********
Advanced Senior Contributor

Sponsor
February 2007

Group: Sponsors
Posts: 4,660
Member No.: 504
Joined: February-16-03

Gender: Undisclosed
Politics: Undisclosed
Party affiliation: Undisclosed



QUOTE(akaCG @ Oct 1 2012, 09:40 PM) *
Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?


Cenk Uygur is a fool That needs to be stated first and foremost.

Secondly, the election is NOT over. Obama has a slight edge in most of the key battleground states, but nationally this race is almost a dead heat.

The momentum is on Obama's side. Most of the polls are on Obama's side. Most of the trends are on Obama's side.

What isn't on Obama's side is the economy. If it were up on its feet and running well, this wouldn't even be a competition. Obama would be crushing Romney by double-digits.

But the economy isn't up and running well. It's staggering, stumbling and unsteady and that is still the biggest club Romney has to hammer Obama on. At the first debate Romney has to use that club if he wants to reset the race and change the trajectory of the polls.

To beat an incumbent president you have to be incredibly popular (Reagan) and possess far more charisma and charm (Clinton) than the incumbent. That, or there has to be a signature issue where the incumbent has failed. There are reasons for Republicans to be encouraged. The economy is Romney's best chance to draw a clear difference between himself and Obama.

The other factor that could shift the election is how dramatically will the new Voter I.D./suppression laws effect the election. My guess is probably more than Obama hopes it does but not as much as Romney hoes it does.

It's not over. Only a fool would say that it is. Or Cenk Uygur. But I repeat myself.... rolleyes.gif
Go to the top of the page
 
+Quote Post
Raptavio
post Oct 2 2012, 06:06 PM
Post #9


*********
Advanced Senior Contributor

Group: Members
Posts: 3,515
Member No.: 10,458
Joined: April-27-09

From: Rosemount, MN
Gender: Male
Politics: Very Liberal
Party affiliation: Democrat



Okay. I admit it, I just can't do it.

I was going to remain silent for the next month, and watch patiently as dear akaCG, bless his heart, goes out farther and farther on the "Teh polls are biased!!!2!!!" limb, and then as it snapped out from under him on Nov. 6, revel in the schadenfreude and point and laugh while he tumbles to the cold, hard ground of political reality.

But I just can't do it. Try as I might, I just can't be that mean. Better to give his fanciful delusions a mercy killing now. akaCG is resting his hopes on the belief that the polls grossly oversample Democrats, and this belief is based on the howlers circulating the right-wing side of the political blogosphere and punditocracy, reaching their zenith of laughworthiness at http://unskewedpolls.com/ .

As I've mentioned before, I'm a software developer by trade. I've also mentioned that I have a professional and academic background in statistics -- I've been a student teacher of statistics processing at the collegiate and graduate levels and my current job is to a great degree about statistics. Partisan though I am, it's very easy to avoid putting a partisan slant on how one views statistics when one understands the underlying mathematical principles.

It pained me in 2004 to see my brother Democrats cling to the false hope that the polls showing GW Bush was headed toward electoral victory were oversampling conservatives. Pollsters, by and large (more on this later), do not weight their polling based on their measures of party ID, so consistent oversampling poll after poll could only be considered plausible due to deliberate malfeasance of the polling firms. As polling firms' entire success relies upon their ability to demonstrate accuracy in relation to actual electoral outcomes, that theory doesn't pass the laugh test. And of course the polls (with a few notable exceptions) reflected the actual outcomes. And Democrats who didn't understand how polling works were bitterly disappointed.

And Republicans now are falling into the same trap, with the same laughworthy implausibility, and making people like akaCG their patsies in promoting this meme. I admit it's amusing to watch these people who don't understand statistics worth a crap twisting themselves into knots to try to make this theory seem valid. But it just isn't so.

(In fairness: In 2010 I got caught in a similar trap, noting that demographics were changing in an interesting way and that many people no longer had land lines, while pollsters by and large did not survey cell phone users. Looking at the relative demographics of cell-phone-only people led me to conclude that Ds would be undersampled. I wound up missing one Senate race (Kentucky's) by a fair margin and the control of the House by a wide margin based on that assumption. But that theory didn't require an assumption of deliberate skewing on the part of pollsters, at least.)

I'll pause here to include Nate Silver's remarks on this right-wing meme:

http://fivethirtyeight.blogs.nytimes.com/2...-partisan-bias/

I encourage reading the entire article as it's very informative, but here's a money quote:

QUOTE
Pollsters will re-weight their numbers if the demographics of their sample diverge from Census Bureau data. For instance, it is typically more challenging to get younger voters on the phone, so most pollsters weight their samples by age to remedy this problem.

But party identification is not a hard-and-fast demographic characteristic like race, age or gender. Instead, it can change in reaction to news and political events from the party conventions to the Sept. 11 attacks. Since changes in public opinion are precisely what polls are trying to measure, it would defeat the purpose of conducting a survey if pollsters insisted that they knew what it was ahead of time.


To bring this to real-world examples, as I said above, most pollsters do NOT weight their polling based on party ID, for reasons Nate Silver summarized above. But there are exceptions. The most notable, of course, is Rasmussen. And since the 2008 elections, as Rasmussen's weighting got less and less accurate, they became the most inaccurate major pollster in the 2010 midterms.

But let's bring this down to brass tacks, okay? Here are the facts:

With all but two of the major pollsters (Rasmussen and Voter/Consumer Research) doing their polling without any party ID weighting, this means the party ID resulting from their polling is, essentially, a survey of party ID of the moment. To say "But the actual party ID according to THIS survey taken last month is different, so you oversampled Dems/Reps" is the logical equivalent to saying "But THIS survey shows a different percentage of people voting for Obama vs. Romney, so you oversampled Obama/Romney voters". In other words, it makes no logical sense.

When the pollsters are consistently discovering in their polls a D+5 to D+10 break, and you see that a single Gallup or Rasmussen poll shows a D+1 break, if you leap to the conclusion that the pollsters reporting a D+5 to D+10 break are oversampling Democrats, without considering that the D+1 break may be either the outlier or outdated, you're violating the most fundamental principle behind Occam's Razor and have fallen prey to the same confirmation bias that leads one to hilariously claim that a crowd was chanting "Romney! Romney!" when any darn fool with ears can hear the truth.

And the dilemma of deciding which of competing polls is accurate aside, the fact that these polls with which right-wingers take issue are being taken randomly and do not weight their survey responses based on any assumptions of party ID. Meaning the party ID of the respondents will, 19 times out of 20, be within the poll's margin of error (which, for those who don't know, is the margin by which the results of same poll taken by random sample 100 times would 95 of those times be within the results if the entire electorate were canvassed). Which means to believe the riotous claims of those claiming "unskewed" polls reflect far more Republicans in the sample must assert one of three things:
1) That other polls are consistently, by random sample, making outlier polls (those which statistically fall into the 5% that wind up outside the margin of error), which becomes more unlikely with every poll (two polls to fall outside the margin of error back-to-back by random chance would happen 1/4 of 1% of the time, three polls, 1.25 out of 10,000 times, etc.)
2) That there is a deliberate effort by multiple pollsters to falsify or "skew" their results (see above as to why this claim doesn't pass the laugh test)
3) That there is a fundamental flaw in the way every pollster that isn't Gallup or Rasmussen gathers data that creates oversampling of Democrats or undersampling of Republicans.

This last one is the most interesting but would require evidence of some fundamental change in either how these pollsters are gathering their data (all of them), or a fundamental change in the populace that causes this skew to show up in all the pollsters except Gallup or Rasmussen. No evidence, of course, has been presented to that effect, which leads us to again look at who was the worst pollster of the 2010 cycle -- Rasmussen, who weights their polls according to their assumptions of voter ID -- and the fact that Rasmussen's methodology has been shown in the most recent campaign cycle to be the least accurate of them all (including missing the Hawaii Senate race in 2010 by 40 points, the worst performance by a pollster in FiveThirtyEight's entire database) -- then one can't give the 'fundamental flaw' theory credence.

In conclusion: akaCG and every GOP partisan playing the "The polls are skewed!" game are grasping at thin, brittle straws that reflect either a fundamental misunderstanding of how political polling works, or an uproariously funny case of desperation and denial pretending to be reasoned statistical analysis.


---------------------------

All that said, it's never over until the ballots are cast. The race is definitely Obama's to lose at this point, but a major gaffe by the President or his campaign, a major victory by Governor Romney in the debates, or some unforeseen event (another stock market crash, another 9/11 type attack, etc.) could cause things to shift radically come election day. But barring those events, President Obama is on a glide path to an easy reelection, the Democrats are well on their way to holding the Senate, and while it looks like the GOP will continue to hold the House, it's likely to be by a much smaller (perhaps even razor-thin) margin come January.

Mr. Uygur's declaration is, therefore, premature to say the least; but he's not wrong in the underlying point that the current map shows Romney has little hope of finding a path to electoral victory.
Go to the top of the page
 
+Quote Post
akaCG
post Oct 3 2012, 01:52 AM
Post #10


*********
Advanced Senior Contributor

Sponsor
August 2012

Group: Sponsors
Posts: 4,846
Member No.: 10,787
Joined: November-25-09

Gender: Male
Politics: Conservative
Party affiliation: Independent



"Raptavio":

1.
I very much look forward to the battery of late October (the time when pollsters, as well as their clients, need to worry much more about their accuracy rankings than anything else; after all, pollsters' accuracy rankings are based on how close their last-before-the-election poll was, not on how close their poll(s) taken 45-30 days before the election was/were) polls that ... somehow/mysteriously ... will have gone from regularly featuring D +8, +9, +10, + 11, even +14 samplings to regularly featuring D +5, +4, +3, +2 and such samplings.

2.
I also very much look forward to the battery of late October (... see above ...) polls that ... somehow/mysteriously ... will have gone from sporadically featuring 54% women versus 46% men, 16/17% more 18-49 year olds, etc. than there were as of the 2010 Census (either national or state-level) to regularly featuring samplings thereof that actually have some resemblance to said 2010 Census stats.

3.
I also very much appreciate your ever-so-merciful decision to override your initial leanings toward "dancing on my grave" on November 7 and instead opting for "killing me softly" right now./

ps:
QUOTE
...
HH: And so if, in fact, you had gotten a hundred Democrats out of a hundred respondents that answered, would you think that poll was reliable?

PB: Probably not at 100 out of 100.

HH: Okay, so if it was 75 out of 100Ö

PB: Well, I mean ...
...
...
...
HH: Thatís not responsive. Itís just a question. Do you personally, Peter, believe that Democrats enjoy a nine point turnout advantage right now?

PB: What I believe is what we found.

HH: Geez, I just, and an eight point in Ohio? Iím from Ohio. Democrats havenít had an eight point advantage in Ohio since before the Civil War. I mean, that just never happens, but Peter, I appreciate your coming on. Iím not persuaded that Quinnipiac Polls havenít hurt themselves today, but I appreciate your willingness to come on and talk about it.
...

Link: http://www.hughhewitt.com/transcripts.aspx...74-4ce24952aceb

Heh.

This post has been edited by akaCG: Oct 3 2012, 01:58 AM
Go to the top of the page
 
+Quote Post
AuthorMusician
post Oct 3 2012, 02:18 AM
Post #11


**********
Glasses and journalism work for me.

Sponsor
November 2003

Group: Sponsors
Posts: 6,342
Member No.: 297
Joined: December-1-02

From: Blueberry Hill
Gender: Male
Politics: Liberal
Party affiliation: Democrat



Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?
I've always figured that the GOP has a rat turd's chance in Martha Stewart's kitchen of nabbing the POTUS slot. That leaves keeping the House, which is where the real money is going, or gaining the Senate, and by gosh, that may go super-majority for the Demos. That's due to redistricting meaning nothing in Senate races.

I believe the GOP has run out of dirty tricks. It is payback time, so eat it up with ranch dressing. Lots of it. Might want to throw some sour grapes sauce on that.
Go to the top of the page
 
+Quote Post
Raptavio
post Oct 3 2012, 03:20 AM
Post #12


*********
Advanced Senior Contributor

Group: Members
Posts: 3,515
Member No.: 10,458
Joined: April-27-09

From: Rosemount, MN
Gender: Male
Politics: Very Liberal
Party affiliation: Democrat



QUOTE(akaCG @ Oct 2 2012, 08:52 PM) *
"Raptavio":

1.
I very much look forward to the battery of late October (the time when pollsters, as well as their clients, need to worry much more about their accuracy rankings than anything else; after all, pollsters' accuracy rankings are based on how close their last-before-the-election poll was, not on how close their poll(s) taken 45-30 days before the election was/were) polls that ... somehow/mysteriously ... will have gone from regularly featuring D +8, +9, +10, + 11, even +14 samplings to regularly featuring D +5, +4, +3, +2 and such samplings.

2.
I also very much look forward to the battery of late October (... see above ...) polls that ... somehow/mysteriously ... will have gone from sporadically featuring 54% women versus 46% men, 16/17% more 18-49 year olds, etc. than there were as of the 2010 Census (either national or state-level) to regularly featuring samplings thereof that actually have some resemblance to said 2010 Census stats.


Well, this is typical akaCG maneuvering. Imply the most hilarious of the options -- that all pollsters not from the House of Ras or Gallup are conspiring, on their clients' behalf (notably, the mainstream media), to skew the polls in the favor of Democrats (ignoring the fact that said House of Ras was at the bottom of the heap last cycle in terms of accuracy) -- but lack the cojones to come out and say so openly.

Typical. Amusing, but typical.

In the meantime, I see your nonsense from Hugh Hewitt and raise you the Editor-in-Chief of your beloved Gallup:

http://pollingmatters.gallup.com/2012/09/t...g-focus-on.html

QUOTE
The discussion of the party identification composition of poll samples comes up in every presidential election with which I've been involved. Interested observers often opine that when a given poll shows that Candidate X is ahead, it cannot be correct because there is a higher percentage of voters who identify with Candidate Xís party in the sample than there should be, based on comparison to some previous standard.

There are several reasons why this is a faulty approach to evaluating a poll's results.

Party identification is basically an attitudinal variable, not a stable population parameter.

...

Here is how Gallup asks party identification: ďIn politics, as of today, do you consider yourself a Republican, a Democrat, or an independent?Ē

Note that this question does not ask, ďWhat was your party identification in November 2008?Ē Nor does it ask, ďAre you registered with one party or the other in your state?Ē Our question uses the words "as of today" and "consider." It is designed to measure fluidity in political self-identification.

We know that party identification moves over time -- sometimes in very short periods of time, just like other political variables. Generally, if there is a political tide toward either of the two major parties, all questions we ask that are of a political nature will move in that direction. This includes the ballot, job approval, party identification, among others.

So, it would not be surprising to find that if Barack Obama is enjoying a surge in popularity in any given state, that surge will show up on the ballot question, on his job approval measure, and on the measure of party identification. So, data showing that Obama is ahead on the ballot in a specific state poll and that Democrats have a higher-than-expected representation on the party identification question, are basically just reflecting two measures of the same underlying phenomenon.


But, since you're going to go down the tinfoil-hatted road of conspiracy theory, perhaps you'll just add Gallup to the list of pollsters in on the conspiracy.

And, since you're crimping your Reynolds Wrap beanie as we speak, I conclude you have not only failed to muster the courage of your convictions, but you have inured yourself to both reason and logic. So I tip my fur felt fedora to you, and bid you adieu. Enjoy the next five weeks, and bless your heart. We'll talk again Nov. 7.
Go to the top of the page
 
+Quote Post
akaCG
post Oct 3 2012, 03:51 AM
Post #13


*********
Advanced Senior Contributor

Sponsor
August 2012

Group: Sponsors
Posts: 4,846
Member No.: 10,787
Joined: November-25-09

Gender: Male
Politics: Conservative
Party affiliation: Independent



QUOTE(Raptavio @ Oct 2 2012, 11:20 PM) *
...
... We'll talk again Nov. 7.

Perfectly cool with me.

But, no matter how said talk turns out, the Wounded Warriors Project will be at least $270 better off on November 7, 2012 than they would have been otherwise.

And that's a good thing.

This post has been edited by akaCG: Oct 3 2012, 03:58 AM
Go to the top of the page
 
+Quote Post
Raptavio
post Oct 3 2012, 04:07 AM
Post #14


*********
Advanced Senior Contributor

Group: Members
Posts: 3,515
Member No.: 10,458
Joined: April-27-09

From: Rosemount, MN
Gender: Male
Politics: Very Liberal
Party affiliation: Democrat



QUOTE(akaCG @ Oct 2 2012, 10:51 PM) *
QUOTE(Raptavio @ Oct 2 2012, 11:20 PM) *
...
... We'll talk again Nov. 7.

Perfectly cool with me.

But, no matter how said talk turns out, the Wounded Warriors Project will be at least $270 better off on November 7, 2012 than they would have been otherwise.

And that's a good thing.

And there, at least, we are in full agreement.
Go to the top of the page
 
+Quote Post
Sleeper
post Oct 3 2012, 03:16 PM
Post #15


********
Who dat!

Sponsor
February 2004

Group: Sponsors
Posts: 1,692
Member No.: 386
Joined: January-17-03

From: Florida
Gender: Male
Politics: Independent
Party affiliation: Independent



Hey Rap, At least akaCG isn't putting his stock in Research 2000 like some people did in the past? rolleyes.gif

And if I remember correctly, he came to your defense when they were outed for the fraud that they were, because you had so steadfastly stood behind their numbers when Daily KOS used them.

I think we can see who the better man is here...
Go to the top of the page
 
+Quote Post
Paladin Elspeth
post Oct 3 2012, 04:43 PM
Post #16


*********
I want the 10th Doctor for President!

Sponsor
August 1, 2003

Group: Sponsors
Posts: 4,987
Member No.: 721
Joined: May-10-03

From: Between 2 Great Lakes
Gender: Female
Politics: Liberal
Party affiliation: Democrat



QUOTE(Sleeper @ Oct 3 2012, 11:16 AM) *
Hey Rap, At least akaCG isn't putting his stock in Research 2000 like some people did in the past? rolleyes.gif

And if I remember correctly, he came to your defense when they were outed for the fraud that they were, because you had so steadfastly stood behind their numbers when Daily KOS used them.

I think we can see who the better man is here...

Can we now? Can we make comments without making comparisons to character, whatever the "better man" is?

These two have been at it for years. I say we just wait 'til the smoke clears and let the facts do the 'splainin'. The question has been asked and answered in this thread. You can now see what remains here.

It's one man's opinion, and regardless of our answers, I'm pretty sure Cenk Uygur would be flattered that we actually devoted an entire thread to something he said. rolleyes.gif


This post has been edited by Paladin Elspeth: Oct 3 2012, 04:58 PM
Go to the top of the page
 
+Quote Post
Jaime
post Oct 3 2012, 05:30 PM
Post #17


Group Icon

**********
Elite Senior Contributor

Group: Admin
Posts: 5,941
Member No.: 4
Joined: July-25-02

From: Down where the River meets the Sea
Gender: Female
Politics: Independent
Party affiliation: None



How about we not make this personal? Keep to the topic, please.

Is the "It's over. President Obama will get re-elected" interpretation of said polls one that YOU buy into? Why or why not?
Go to the top of the page
 
+Quote Post
Hobbes
post Oct 4 2012, 03:37 AM
Post #18


Group Icon

**********
No More Mr. Nice Guy!

Group: Committee Members
Posts: 5,311
Member No.: 1,155
Joined: September-8-03

From: Dallas, TX
Gender: Male
Politics: Conservative
Party affiliation: Independent



My two cents on polling. First, it is very important to see who is being polled. It is quite common to see polls shift from registered voters to likely voters as the election gets closer. Early polls aren't done to see who is going to win the election, really, but to see how the candidates are perceived on various issues. Whether the respondent is likely to vote doesn't really impact that aspect that much. Otherwise, all these polls would need is one question: Who are you going to vote for.

But more important than who is asked is WHAT is asked. There are all kinds of ways to phrase a question, and the way a question is phrased can skew the answer condierably. The questions aren't usually shown, even, just the results, which themselves can be skewed depeniding on which results one chooses to show or how one chooses to interpret them.

Very few polls go about their business in what would be considered a scientifically sound method. Not usually any repetition of questions to see if the answers stay consistent, or reverse sampling to see if their answer also changes etc. They overcome this through sheer numbers of people polled. That is why looking at the makeup of those numbers can be so important---if the responders aren't fully correlated with who will actually vote, then the aggregate of their responses will be similarly biased.

Which doesn't mean I buy into all the polls being off due to that. Rather, I believe that the polls are off simply because polls are usually off prior to the election and that the results get closer to 'reality' as the election itself gets closer. Many respondents haven't made up their minds in early polls, and as they do, their answers change. All it takes is a few people out of a hundred to do this to completely change the results (3 people switching would be a 6 point change in the results). Hence, early polls tend to be much less accurate than later polls. So, as the election nears....we'll get a better idea of how the election is really leaning. I don't doubt that Obama is leading in many polls---he should be. But by how much exactly? I don't think a definitive answer is known yet. Which means those couple people out of a hundred are very very important (especially if they happen to live in Ohio or Florida).

This post has been edited by Hobbes: Oct 4 2012, 03:44 AM
Go to the top of the page
 
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 

  
Go to the top of the page - Simple Version Time is now: May 24th, 2018 - 03:43 AM
©2002-2010 America's Debate, Inc.  All rights reserved.