I've said before in other posts that I'd vote for Obama if the polls indicated a close race; otherwise I intend to vote for Cynthia McKinney.
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
Full Version: Do you trust polls?
Lol you should have made this thread into a poll.
No I don't trust polls...not when they tell me to distrust Wayne Madsen...who has previously reported stories about election fraud.
Polls are subject to manipulation for the purpose of buttressing the credibility of outfits that employ plagiarists, liars and idiots.
Polls are subject to manipulation for the purpose of buttressing the credibility of outfits that employ plagiarists, liars and idiots.
QUOTE (Tagline @ Aug 23 2008, 11:52 PM) 
Polls are subject to manipulation for the purpose of buttressing the credibility of outfits that employ plagiarists, liars and idiots.
Really? Do you have polling data to back that claim up?
QUOTE (Tagline @ Aug 24 2008, 02:52 AM)
No I don't trust polls...not when they tell me to distrust Wayne Madsen...who has previously reported stories about election fraud.
Polls are subject to manipulation for the purpose of buttressing the credibility of outfits that employ plagiarists, liars and idiots.
Polls are subject to manipulation for the purpose of buttressing the credibility of outfits that employ plagiarists, liars and idiots.
Even a broken clock is right twice a day.
Polls are easily manipulated. If you ask the right question in the right framing of words to the right group of people, you can get the answer you want. It really is pretty easy to do.
So, if polling and stealing elections (especially where black box voting machines are concerned) go hand in glove, then haven't we already lost our democracy? In that case, it won't matter how overwhelming your support for Obama is. Since your vote won't matter, why bother?
At the risk of sounding all conspiratorial and all, really, what options do you have?
At the risk of sounding all conspiratorial and all, really, what options do you have?
I only trust polls that use proper random sampling, a large sample size, a proper methodology of choosing people in the sample so as to ensure they are dis-similar from each other (that goes with being properly random), and carefully designed questions that are not leading questions or push polling.
For this reason, yes, most Internet polls, which have many flaws, I consider meaningless, such as, for example, the ones saying Ron Paul was going to win the Republican primary.
Yes, because the kind of people inclined to vote Ron Paul are already reading the site and taking the poll, and it's about the furthest thing in the universe from a true random sample.
For this reason, yes, most Internet polls, which have many flaws, I consider meaningless, such as, for example, the ones saying Ron Paul was going to win the Republican primary.
Yes, because the kind of people inclined to vote Ron Paul are already reading the site and taking the poll, and it's about the furthest thing in the universe from a true random sample.
No... 2000 & 2004
www.rovecybergate.com
http://www.uncountedthemovie.com/
The Best Democracy Money Can Buy
by Greg Palast
www.rovecybergate.com
QUOTE
Tell Congress to investigate Karl Rove’s cyber strategy to illegally manipulate elections. Click here to send an email to your Congress Members demanding immediate public hearings on whistleblower allegations that Rove architected and directed illegal attacks on Democrat candidates through the improper use of corporate funds channeled through fake Web-based front organizations, the improper political use of the Justice Department to prosecute opposition candidates, and the use of Internet based IT networks to alter election results.
http://www.uncountedthemovie.com/
QUOTE
UNCOUNTED is an explosive new documentary that shows how the election fraud that changed the outcome of the 2004 election led to even greater fraud in 2006 - and now looms as an unbridled threat to the outcome of the 2008 election.
The Best Democracy Money Can Buy
by Greg Palast
QUOTE
This collection of reports touches on a number of familiar topics, including Enron, the presidential election of 2000 and the Bush family’s purported connection to Saudi Arabia.
QUOTE (Balor @ Aug 24 2008, 08:57 AM)
So, if polling and stealing elections (especially where black box voting machines are concerned) go hand in glove, then haven't we already lost our democracy? In that case, it won't matter how overwhelming your support for Obama is. Since your vote won't matter, why bother?
At the risk of sounding all conspiratorial and all, really, what options do you have?
At the risk of sounding all conspiratorial and all, really, what options do you have?
Well you can run a poll that has integrity by following the proper methodology. The problem we face today is methodology isn't important, getting the right results is more important to most of those who take polls than doing the job correctly.
We also have that nagging problem of young people not owning land line telephones and those of us (myself being included in here) who have caller ID and refuse to answer a call from an unknown or unpublished phone number. So polls don't really work like they used to.
There was a time when the two were highly related, but a lot has changed in the past 10-20 years.
polls are skewed... elections manipulated (recently at least)
The "polls" can ask anyone they want anything they want.
The have no effect on turnout at the only "polls" that count.
The have no effect on turnout at the only "polls" that count.
I only "trust" one poll... the only one that counts .. the one that is scheduled for November 4th. (Let's watch out for the "cheaters".)
QUOTE (rottmom @ Aug 24 2008, 12:22 PM)
We also have that nagging problem of young people not owning land line telephones and those of us (myself being included in here) who have caller ID and refuse to answer a call from an unknown or unpublished phone number. So polls don't really work like they used to.
There was a time when the two were highly related, but a lot has changed in the past 10-20 years.
There was a time when the two were highly related, but a lot has changed in the past 10-20 years.
One of the reporters on CNN was asked about this issue and they said that the pollsters were aware of the problem and that they take this into account when determining the results.
Unless they actually poll them, I don't see how they can accurately do this ... of course, who ever said they were trying for accuracy. My guess is that they simply assume that students and people with cell phones don't bother to vote, and that's what they take into consideration.
Another issue is that students are not in school yet, so they might be getting their parents opinions, but are totally missing the student vote.
I do trust polls, lot of money and human resources are put in to get them completed. They happen to be accurate most of the time, except for the NH Dem primary 2008 - BIG surprise - , and maybe 1 or 2 more times in my life time.
Some of you need to wake up and smell the roses. This election won't be a walk in the park. America seems to get more and more divided every day.
Some of you need to wake up and smell the roses. This election won't be a walk in the park. America seems to get more and more divided every day.
Hummm ... Maybe I'm just being paranoid, or spending too much time watching the convention, but up until the Biden announcement we had at least one new poll a day being released ... since then I recall maybe one poll, and that was taken before the announcement.
There was a new poll (National Gallop tracking poll showing McCain with a slight lead) this morning, I didn't catch the actual numbers because I got distracted when they showed it, but one of the commentators said this was the first poll since the Biden announcement ... almost 4 days since the announcement and this is the first poll? I don't get it.
There was a new poll (National Gallop tracking poll showing McCain with a slight lead) this morning, I didn't catch the actual numbers because I got distracted when they showed it, but one of the commentators said this was the first poll since the Biden announcement ... almost 4 days since the announcement and this is the first poll? I don't get it.
If the election is close, then the election will be stolen. Count on it. How will we know it is close? The polls will tell us. Garbage in, garbage out. Garbage we have to live with.
QUOTE (rottmom @ Aug 24 2008, 06:57 AM)
Polls are easily manipulated. If you ask the right question in the right framing of words to the right group of people, you can get the answer you want. It really is pretty easy to do.
Of course you are right on this. That is why one can trust the outcome of the poll, only if you understand what was asked. There are proper polling techniques, and there are polls designed to generate particular responses.
One might be better off asking whether you trust the reporting of polls.
I think it also makes sense to look at a group of polls, over time and over a range of techniques and pollsters.
There is a reason why I don't trust polls. They're not really accurate. They say they are, but they're not. The typical poll will ask approximately 1000 people their opinions. That's 1000 people out of 300 million, in 50 states. 1000/50 is 20 people per state. I challenge you to ask 20 random people in any state a question and have the answers accurately reflect the opinion of the electorate there.
Then there's the bias. If a pollster wants to skew results, they will take some of those 20 people/state, and give them to other states. Sometimes this is done "because x state is larger than y." Sometimes they want to reflect a certain result so they'll say ask more people in the bible belt than say the northeastern states. Then there's the way the ask the questions. Asking questions a certain way frames the debate and guides the answers. The old British sitcom Yes Minister demonstrates how this is done perfectly:
Opinion Polls: Getting the results you want
Then there's the bias. If a pollster wants to skew results, they will take some of those 20 people/state, and give them to other states. Sometimes this is done "because x state is larger than y." Sometimes they want to reflect a certain result so they'll say ask more people in the bible belt than say the northeastern states. Then there's the way the ask the questions. Asking questions a certain way frames the debate and guides the answers. The old British sitcom Yes Minister demonstrates how this is done perfectly:
Opinion Polls: Getting the results you want
Anyone who has ever walked past, hung up on or ignored a pollster should know why polls are generally useless.
There are only two occasions when polls will tell you anything meaningful:
1. Showing consistent trends over a large period of time
2. Showing overwhelming support or opposition to something
Otherwise, like my stats professor said, "statistics is useful, except for polls, which are shite."
There are only two occasions when polls will tell you anything meaningful:
1. Showing consistent trends over a large period of time
2. Showing overwhelming support or opposition to something
Otherwise, like my stats professor said, "statistics is useful, except for polls, which are shite."
QUOTE (jkun17 @ Aug 27 2008, 07:52 AM)
Anyone who has ever walked past, hung up on or ignored a pollster should know why polls are generally useless.
There are only two occasions when polls will tell you anything meaningful:
1. Showing consistent trends over a large period of time
2. Showing overwhelming support or opposition to something
Otherwise, like my stats professor said, "statistics is useful, except for polls, which are shite."
There are only two occasions when polls will tell you anything meaningful:
1. Showing consistent trends over a large period of time
2. Showing overwhelming support or opposition to something
Otherwise, like my stats professor said, "statistics is useful, except for polls, which are shite."
Not only sh*t, but dangerous sh*t because too many people are misled by them and too many politicians formulate policy based on them.
Disagreeing with what appears to be the consensus, I believe the overwhelming majority of polls from established reputable firms - those conducted typically for journalism outlets are accurate.
There is a science to sampling, and each firm develops their own formulas for interviewing X number of people in Y number of categories to come up with reasonably representative measure of certain opinions among the entire population. It's not as though the firms decide to interview 850, or 1,200 people, choosing those numbers arbitrarily.
I believe that over and over polls have been proven to be accurate. In addition, I confess noting the irony of some (both here on RRMB, and in he outside world) who will loudly and derisively ridicule the reliability of polling, and an hour later without a moment's hesitation while addressing a completely different subject, those same skeptics will passionately note that 70% of the country opposes the war in Iraq, or note that Congress has a 19% approval rating.
All that said, I also believe the vast majority of the polling done is simultaneously useless, and grossly misused - especially by the ignorant.
A fine example would be the polling done before the New Hampshire primary, an example often cited today by those who mock polling - even by RR - and generally regarded as recent "proof" that polls are "often wrong."
I don't claim to recall the specific figures, nor the names they were attached to, and I'm too damned laxy to go look it all up. But I will perhaps be forgiven for simply inventing figures for the sake of example, because I think you'll see the exact numbers and names don't really matter for purposes of the example.
In short, polling just before primary election indicated that Obama had a considerable lead over Clinton. Let's say it was something in the neighborhood of 30% for Obama and 20% for Clinton. Much was made of this difference. But in the end, Clinton won by a single percentage point. And so folks conclude, "The polls got it wrong!"
For those who looked at ONLY those two numbers, sure, it appeared the polls were wrong. Those 30% vs. 20% numbers were all folks talked about before the election, and so the results were clearly a big surprise.
Trouble is most folks, from pundits paid to know better, to the guy at the end of the bar, completely ignored that also in the polling data was a figure for those who would be voting, but who had not yet decided who they would support. It was a very high figure, and relative to the numbers in the scenario above, it would be fair to portray that "undecided" figure in the neighborhood of 20%- to 30%.
That percentage ultimately did vote, and they m,ade a choice, and their picks turned up in the results.
It's not the fault of the pollsters that the media and others portrayed the polling results inaccurately.
Beyond that clear record of the numbers being bungled, most polling is, I believe, pretty damned accurate, but utterly meaningless. last year's poll on how Clinton would do up against Romney doesn't mean a whole hell of a lot today, does it? That was money down the toilet!
Measuring Obama and McCain on a daily, weekly or even monthly basis is ludicrous. Again, I suspect most of the polls are accurate to within something like their stated margins for error, but they contribute absolutely nothing to knowing which of the candidates would make a better president, which is a liar, or which has the best plan for the future.
But the endless flow of numbers gives the networks something relatively inexpensive, something supposedly (but not really) easy to understand, and something that people flock to like flies to shit.
I'm always amazed when an anchor or pundit will announce with great gravity word that candidate X is behind candidate Y 41% to 44%, an important development because last week their positions were reversed, with Y at 40% and X at 43%.
Well both damned sets of numbers are within the poll's margin for error - a statistic that's ALWAYS noted, but only as a seemingly meaningless afterthought or technicality. In fact, the results of BOTH polls show the candidates are effectively tied, with no appreciable change during the week. But THAT wouldn't be so damned interesting to report, and hash over for the next 30 minutes, would it? It would put the kibosh on speculating whether candidate Y having slapped his wife at that fundraiser might have been responsible for the "precipitous drop." If they're still tied, no time goes to wondering if candidate X buying his kid a puppy might have helped bring about the "big comeback."
Finally, what happens every time there's a poll here at RRMB? A half dozen members observe that there wasn't an answer they were really comfortable with. Well, professional poster invest hours in crafting useful, meaningful and telling poll questions - but even they can't nail every category of answer. So they try to nudge survey subjects into picking the "closest" answer. Completely understandable, but how many important nuances are lost in the process that would make a difference to us when we're weighing the polling results?
I think campaigns themselves NEED polls, and Lord knows they think the ones they pay for are, for the most part, useful and reliable. But their interests and needs are much different than ours.
Polling is, I believe, mostly accurate. (I'm not getting into the bogus polls commissioned by interest groups, or other such games.) But most folks don't understand what the results are showing, and without exception the results are utterly useless in the process of choosing a president. if the nets spent half the time comparing and debating tax plans that they do poll results, we'd be much better off. But who'd watch that (other than some of us)? And who'd call and e-mail their friends to say, "Didja hear what Brokaw said about McCain's tax structure?"
Are You Kidding Me, MSNBC?
Talk about stacking the deck. This is one of MSNBC's most recent "text question" poll:
What do you feel McCain's speech accomplished the most?
A. Inspired the nation
B. Explained what motivated his presidential run
C. Outlined his policy priorities if elected
D. Positioned himself as the tough maverick for change
Notice anything missing from those options?
Talk about stacking the deck. This is one of MSNBC's most recent "text question" poll:
What do you feel McCain's speech accomplished the most?
A. Inspired the nation
B. Explained what motivated his presidential run
C. Outlined his policy priorities if elected
D. Positioned himself as the tough maverick for change
Notice anything missing from those options?
QUOTE (plodder @ Sep 6 2008, 12:51 AM)
Are You Kidding Me, MSNBC?
Talk about stacking the deck. This is one of MSNBC's most recent "text question" poll:
What do you feel McCain's speech accomplished the most?
A. Inspired the nation
B. Explained what motivated his presidential run
C. Outlined his policy priorities if elected
D. Positioned himself as the tough maverick for change
Notice anything missing from those options?
Talk about stacking the deck. This is one of MSNBC's most recent "text question" poll:
What do you feel McCain's speech accomplished the most?
A. Inspired the nation
B. Explained what motivated his presidential run
C. Outlined his policy priorities if elected
D. Positioned himself as the tough maverick for change
Notice anything missing from those options?
E. Caused you to develop a screaming headache and a return of your last meal to the top of your throat.
QUOTE (plodder @ Sep 6 2008, 12:51 AM)
Are You Kidding Me, MSNBC?
Talk about stacking the deck. This is one of MSNBC's most recent "text question" poll:
What do you feel McCain's speech accomplished the most?
A. Inspired the nation
B. Explained what motivated his presidential run
C. Outlined his policy priorities if elected
D. Positioned himself as the tough maverick for change
Notice anything missing from those options?
Talk about stacking the deck. This is one of MSNBC's most recent "text question" poll:
What do you feel McCain's speech accomplished the most?
A. Inspired the nation
B. Explained what motivated his presidential run
C. Outlined his policy priorities if elected
D. Positioned himself as the tough maverick for change
Notice anything missing from those options?
It doesn't even have the decency of including a "none of the above." lolA. John is a good little boy
B. John is a nice little boy.
C. Ad Nauseum
QUOTE (Fellixe @ Sep 6 2008, 12:56 AM)
E. Caused you to develop a screaming headache and a return of your last meal to the top of your throat.
F. Made one wonder all the more why someone who is in labor and dripping all over God's Precious Creation would take an 8+ hour flight from Texas to Alaska, because one can't have a fish-puller being born in TX, now can we.
G. Made me wonder how a Methuselah who speaks in a monotone whisper while reading a list of cliches and nasty innuendos is even on the podium speaking.
QUOTE (rottmom @ Aug 24 2008, 07:57 AM)
Even a broken clock is right twice a day.
Polls are easily manipulated. If you ask the right question in the right framing of words to the right group of people, you can get the answer you want. It really is pretty easy to do.
Polls are easily manipulated. If you ask the right question in the right framing of words to the right group of people, you can get the answer you want. It really is pretty easy to do.
I was polled the other day about my congressman, Mahoney. It was a "push" poll.
Q: Do you have a favorable, very favorable, unfavorable or very unfavorable opinion of Congressman Mahoney?
A Favorable
Q. Would you still be favorable of him if you knew he supported abortion on demand and partial birth/live birth abortions?
A. Yes. I love abortion. I think everyone should have at least 5 of them. In fact, if I'm not busy on Saturday, I think I will go have one.
(uncomfortable silence on the other end of the phone...)
Q. Would you still have a favorable opinion of Congressman Mahoney if you knew he was giving illegal aliens an easier path to be come citizens?
A. I love the illegals! I have 5 in my back yard right now mowing the yard. In fact I think we need MORE of them! They work wonders with landscaping!
Q. Ma'am would it be possible to get a serious answer from you tonight?
A. Look dumb ass, you called me. You know I am a super-voter Democrat. What the hell did you expect?
*CLICK*
Polls are bullshit.
Which polling organizations can you trust? Zogby? I am not really interested in how they ask their questions; their methodology might be very correct. I just want to know if they've been "bought" and produce the "results" desired by their purchasers.
It gets down to a matter of integrity I guess. If I feel polls are worthless, then I will most likely vote for Cynthia McKinney and the consequences be damned if McCain gets elected. On the other hand I hope the GOP is so badly trounced that it marks the end for them as a major party.
It gets down to a matter of integrity I guess. If I feel polls are worthless, then I will most likely vote for Cynthia McKinney and the consequences be damned if McCain gets elected. On the other hand I hope the GOP is so badly trounced that it marks the end for them as a major party.
QUOTE (Balor @ Sep 6 2008, 12:06 PM)
Which polling organizations can you trust? Zogby? I am not really interested in how they ask their questions; their methodology might be very correct. I just want to know if they've been "bought" and produce the "results" desired by their purchasers.
It gets down to a matter of integrity I guess. If I feel polls are worthless, then I will most likely vote for Cynthia McKinney and the consequences be damned if McCain gets elected. On the other hand I hope the GOP is so badly trounced that it marks the end for them as a major party.
It gets down to a matter of integrity I guess. If I feel polls are worthless, then I will most likely vote for Cynthia McKinney and the consequences be damned if McCain gets elected. On the other hand I hope the GOP is so badly trounced that it marks the end for them as a major party.
Balor the consequences you are referring to is the supreme court. I will say it once, I will say it another zillion times before election day. The next president will appoint at least one, probably two or possibly three Supremes. Obama could be gone in four years. If McCain-Palin wins, we will be stuck with the right-wing zealots they appoint to the bench for the next 30-40 years.
Do you really want that to happen?
QUOTE (Balor @ Aug 24 2008, 01:40 AM)
I've said before in other posts that I'd vote for Obama if the polls indicated a close race; otherwise I intend to vote for Cynthia McKinney.
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
But that's the point, the race will be close until the winner is declared because of McKinney, Barr and Nader. And the stupid medium keep talking about why hasn't Obama been able to run away with votes. This just shows how much Main Stream Media disrespects Independent Candidates. Have they not figured out yet that, 1)these independents candidates are getting votes, 2) no matter how you slice it, this election is still about race and you can't CURE racism.
So there is not much Obama can do but run as good as a he can. I believe the people that are objective and will not fooled by the "Palin" insurgent by the Rovian Repubs.
As for Palin, it doesn't matter whether she is a woman or not, her very arrogant, adamant beliefs are dangerous men as well as women. With the way women are acting over her and going crazy over her and has only known her for a week is scary. They have not even bothered to look into her background or beliefs. After looking at her speech at her church on YOUTUBE, it is frightening. Palin said she would band any type of aborton if she got in office. I am 58 years, female and the ERA I grew up in, women would get illegal abortion thru "backdoor operations and even so-called midwives were doing them and yes, SEVERAL women died back then. Palin also said the Iraq is a war from GOD.
It reminds of the smooth way that JIM JONES started out in the seventies. He was a so-called minister. He was influential, that when the Federal Government start getting reports of the corruption and abuse he exerted over his congregation, he convinced the entire church to move to Guyanna. When US were still getting these reports, a politician Ryan and some of aids went there. Some people got word to him and before he could leave on the plane they shot him and others dead. Then Jim Jones and his main aids, made a bunch of poison laced Kool-aid and convinced the people to drink it and 900 of his congregation committed suicide and ones that didn't was shot in the head. But there were 30 members that managed to escaple.
I hople these OVERWHELMED women take a second look at Palin before going completely, gah, gah...
QUOTE (Balor @ Aug 24 2008, 01:40 AM)
I've said before in other posts that I'd vote for Obama if the polls indicated a close race; otherwise I intend to vote for Cynthia McKinney.
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
Yes to your poll question.
In terms of which ones, any of the polls done by the major independant pollsters.
Election fraud had no bearing on the 2000 or 2004 elections.
Given what's at stake, I would recommend you don't waste your vote on a 3rd party candidate.
QUOTE (bushwa @ Aug 27 2008, 10:51 AM)
Disagreeing with what appears to be the consensus, I believe the overwhelming majority of polls from established reputable firms - those conducted typically for journalism outlets are accurate.
There is a science to sampling, and each firm develops their own formulas for interviewing X number of people in Y number of categories to come up with reasonably representative measure of certain opinions among the entire population. It's not as though the firms decide to interview 850, or 1,200 people, choosing those numbers arbitrarily.
I believe that over and over polls have been proven to be accurate. In addition, I confess noting the irony of some (both here on RRMB, and in he outside world) who will loudly and derisively ridicule the reliability of polling, and an hour later without a moment's hesitation while addressing a completely different subject, those same skeptics will passionately note that 70% of the country opposes the war in Iraq, or note that Congress has a 19% approval rating.
All that said, I also believe the vast majority of the polling done is simultaneously useless, and grossly misused - especially by the ignorant.
A fine example would be the polling done before the New Hampshire primary, an example often cited today by those who mock polling - even by RR - and generally regarded as recent "proof" that polls are "often wrong."
I don't claim to recall the specific figures, nor the names they were attached to, and I'm too damned laxy to go look it all up. But I will perhaps be forgiven for simply inventing figures for the sake of example, because I think you'll see the exact numbers and names don't really matter for purposes of the example.
In short, polling just before primary election indicated that Obama had a considerable lead over Clinton. Let's say it was something in the neighborhood of 30% for Obama and 20% for Clinton. Much was made of this difference. But in the end, Clinton won by a single percentage point. And so folks conclude, "The polls got it wrong!"
For those who looked at ONLY those two numbers, sure, it appeared the polls were wrong. Those 30% vs. 20% numbers were all folks talked about before the election, and so the results were clearly a big surprise.
Trouble is most folks, from pundits paid to know better, to the guy at the end of the bar, completely ignored that also in the polling data was a figure for those who would be voting, but who had not yet decided who they would support. It was a very high figure, and relative to the numbers in the scenario above, it would be fair to portray that "undecided" figure in the neighborhood of 20%- to 30%.
That percentage ultimately did vote, and they m,ade a choice, and their picks turned up in the results.
It's not the fault of the pollsters that the media and others portrayed the polling results inaccurately.
Beyond that clear record of the numbers being bungled, most polling is, I believe, pretty damned accurate, but utterly meaningless. last year's poll on how Clinton would do up against Romney doesn't mean a whole hell of a lot today, does it? That was money down the toilet!
Measuring Obama and McCain on a daily, weekly or even monthly basis is ludicrous. Again, I suspect most of the polls are accurate to within something like their stated margins for error, but they contribute absolutely nothing to knowing which of the candidates would make a better president, which is a liar, or which has the best plan for the future.
But the endless flow of numbers gives the networks something relatively inexpensive, something supposedly (but not really) easy to understand, and something that people flock to like flies to shit.
I'm always amazed when an anchor or pundit will announce with great gravity word that candidate X is behind candidate Y 41% to 44%, an important development because last week their positions were reversed, with Y at 40% and X at 43%.
Well both damned sets of numbers are within the poll's margin for error - a statistic that's ALWAYS noted, but only as a seemingly meaningless afterthought or technicality. In fact, the results of BOTH polls show the candidates are effectively tied, with no appreciable change during the week. But THAT wouldn't be so damned interesting to report, and hash over for the next 30 minutes, would it? It would put the kibosh on speculating whether candidate Y having slapped his wife at that fundraiser might have been responsible for the "precipitous drop." If they're still tied, no time goes to wondering if candidate X buying his kid a puppy might have helped bring about the "big comeback."
Finally, what happens every time there's a poll here at RRMB? A half dozen members observe that there wasn't an answer they were really comfortable with. Well, professional poster invest hours in crafting useful, meaningful and telling poll questions - but even they can't nail every category of answer. So they try to nudge survey subjects into picking the "closest" answer. Completely understandable, but how many important nuances are lost in the process that would make a difference to us when we're weighing the polling results?
I think campaigns themselves NEED polls, and Lord knows they think the ones they pay for are, for the most part, useful and reliable. But their interests and needs are much different than ours.
Polling is, I believe, mostly accurate. (I'm not getting into the bogus polls commissioned by interest groups, or other such games.) But most folks don't understand what the results are showing, and without exception the results are utterly useless in the process of choosing a president. if the nets spent half the time comparing and debating tax plans that they do poll results, we'd be much better off. But who'd watch that (other than some of us)? And who'd call and e-mail their friends to say, "Didja hear what Brokaw said about McCain's tax structure?"
There is a science to sampling, and each firm develops their own formulas for interviewing X number of people in Y number of categories to come up with reasonably representative measure of certain opinions among the entire population. It's not as though the firms decide to interview 850, or 1,200 people, choosing those numbers arbitrarily.
I believe that over and over polls have been proven to be accurate. In addition, I confess noting the irony of some (both here on RRMB, and in he outside world) who will loudly and derisively ridicule the reliability of polling, and an hour later without a moment's hesitation while addressing a completely different subject, those same skeptics will passionately note that 70% of the country opposes the war in Iraq, or note that Congress has a 19% approval rating.
All that said, I also believe the vast majority of the polling done is simultaneously useless, and grossly misused - especially by the ignorant.
A fine example would be the polling done before the New Hampshire primary, an example often cited today by those who mock polling - even by RR - and generally regarded as recent "proof" that polls are "often wrong."
I don't claim to recall the specific figures, nor the names they were attached to, and I'm too damned laxy to go look it all up. But I will perhaps be forgiven for simply inventing figures for the sake of example, because I think you'll see the exact numbers and names don't really matter for purposes of the example.
In short, polling just before primary election indicated that Obama had a considerable lead over Clinton. Let's say it was something in the neighborhood of 30% for Obama and 20% for Clinton. Much was made of this difference. But in the end, Clinton won by a single percentage point. And so folks conclude, "The polls got it wrong!"
For those who looked at ONLY those two numbers, sure, it appeared the polls were wrong. Those 30% vs. 20% numbers were all folks talked about before the election, and so the results were clearly a big surprise.
Trouble is most folks, from pundits paid to know better, to the guy at the end of the bar, completely ignored that also in the polling data was a figure for those who would be voting, but who had not yet decided who they would support. It was a very high figure, and relative to the numbers in the scenario above, it would be fair to portray that "undecided" figure in the neighborhood of 20%- to 30%.
That percentage ultimately did vote, and they m,ade a choice, and their picks turned up in the results.
It's not the fault of the pollsters that the media and others portrayed the polling results inaccurately.
Beyond that clear record of the numbers being bungled, most polling is, I believe, pretty damned accurate, but utterly meaningless. last year's poll on how Clinton would do up against Romney doesn't mean a whole hell of a lot today, does it? That was money down the toilet!
Measuring Obama and McCain on a daily, weekly or even monthly basis is ludicrous. Again, I suspect most of the polls are accurate to within something like their stated margins for error, but they contribute absolutely nothing to knowing which of the candidates would make a better president, which is a liar, or which has the best plan for the future.
But the endless flow of numbers gives the networks something relatively inexpensive, something supposedly (but not really) easy to understand, and something that people flock to like flies to shit.
I'm always amazed when an anchor or pundit will announce with great gravity word that candidate X is behind candidate Y 41% to 44%, an important development because last week their positions were reversed, with Y at 40% and X at 43%.
Well both damned sets of numbers are within the poll's margin for error - a statistic that's ALWAYS noted, but only as a seemingly meaningless afterthought or technicality. In fact, the results of BOTH polls show the candidates are effectively tied, with no appreciable change during the week. But THAT wouldn't be so damned interesting to report, and hash over for the next 30 minutes, would it? It would put the kibosh on speculating whether candidate Y having slapped his wife at that fundraiser might have been responsible for the "precipitous drop." If they're still tied, no time goes to wondering if candidate X buying his kid a puppy might have helped bring about the "big comeback."
Finally, what happens every time there's a poll here at RRMB? A half dozen members observe that there wasn't an answer they were really comfortable with. Well, professional poster invest hours in crafting useful, meaningful and telling poll questions - but even they can't nail every category of answer. So they try to nudge survey subjects into picking the "closest" answer. Completely understandable, but how many important nuances are lost in the process that would make a difference to us when we're weighing the polling results?
I think campaigns themselves NEED polls, and Lord knows they think the ones they pay for are, for the most part, useful and reliable. But their interests and needs are much different than ours.
Polling is, I believe, mostly accurate. (I'm not getting into the bogus polls commissioned by interest groups, or other such games.) But most folks don't understand what the results are showing, and without exception the results are utterly useless in the process of choosing a president. if the nets spent half the time comparing and debating tax plans that they do poll results, we'd be much better off. But who'd watch that (other than some of us)? And who'd call and e-mail their friends to say, "Didja hear what Brokaw said about McCain's tax structure?"
Perhaps, but I am still very skeptical of the manipulation of polls. Friday the polls showed McCain leading but today, Obama is leading again, so yes I am very skeptical of polls.
QUOTE (bushwa @ Aug 27 2008, 10:51 AM)
Disagreeing with what appears to be the consensus, I believe the overwhelming majority of polls from established reputable firms - those conducted typically for journalism outlets are accurate.
There is a science to sampling, and each firm develops their own formulas for interviewing X number of people in Y number of categories to come up with reasonably representative measure of certain opinions among the entire population. It's not as though the firms decide to interview 850, or 1,200 people, choosing those numbers arbitrarily.
I believe that over and over polls have been proven to be accurate. In addition, I confess noting the irony of some (both here on RRMB, and in he outside world) who will loudly and derisively ridicule the reliability of polling, and an hour later without a moment's hesitation while addressing a completely different subject, those same skeptics will passionately note that 70% of the country opposes the war in Iraq, or note that Congress has a 19% approval rating.
All that said, I also believe the vast majority of the polling done is simultaneously useless, and grossly misused - especially by the ignorant.
A fine example would be the polling done before the New Hampshire primary, an example often cited today by those who mock polling - even by RR - and generally regarded as recent "proof" that polls are "often wrong."
I don't claim to recall the specific figures, nor the names they were attached to, and I'm too damned laxy to go look it all up. But I will perhaps be forgiven for simply inventing figures for the sake of example, because I think you'll see the exact numbers and names don't really matter for purposes of the example.
In short, polling just before primary election indicated that Obama had a considerable lead over Clinton. Let's say it was something in the neighborhood of 30% for Obama and 20% for Clinton. Much was made of this difference. But in the end, Clinton won by a single percentage point. And so folks conclude, "The polls got it wrong!"
For those who looked at ONLY those two numbers, sure, it appeared the polls were wrong. Those 30% vs. 20% numbers were all folks talked about before the election, and so the results were clearly a big surprise.
Trouble is most folks, from pundits paid to know better, to the guy at the end of the bar, completely ignored that also in the polling data was a figure for those who would be voting, but who had not yet decided who they would support. It was a very high figure, and relative to the numbers in the scenario above, it would be fair to portray that "undecided" figure in the neighborhood of 20%- to 30%.
That percentage ultimately did vote, and they m,ade a choice, and their picks turned up in the results.
It's not the fault of the pollsters that the media and others portrayed the polling results inaccurately.
Beyond that clear record of the numbers being bungled, most polling is, I believe, pretty damned accurate, but utterly meaningless. last year's poll on how Clinton would do up against Romney doesn't mean a whole hell of a lot today, does it? That was money down the toilet!
Measuring Obama and McCain on a daily, weekly or even monthly basis is ludicrous. Again, I suspect most of the polls are accurate to within something like their stated margins for error, but they contribute absolutely nothing to knowing which of the candidates would make a better president, which is a liar, or which has the best plan for the future.
But the endless flow of numbers gives the networks something relatively inexpensive, something supposedly (but not really) easy to understand, and something that people flock to like flies to shit.
I'm always amazed when an anchor or pundit will announce with great gravity word that candidate X is behind candidate Y 41% to 44%, an important development because last week their positions were reversed, with Y at 40% and X at 43%.
Well both damned sets of numbers are within the poll's margin for error - a statistic that's ALWAYS noted, but only as a seemingly meaningless afterthought or technicality. In fact, the results of BOTH polls show the candidates are effectively tied, with no appreciable change during the week. But THAT wouldn't be so damned interesting to report, and hash over for the next 30 minutes, would it? It would put the kibosh on speculating whether candidate Y having slapped his wife at that fundraiser might have been responsible for the "precipitous drop." If they're still tied, no time goes to wondering if candidate X buying his kid a puppy might have helped bring about the "big comeback."
Finally, what happens every time there's a poll here at RRMB? A half dozen members observe that there wasn't an answer they were really comfortable with. Well, professional poster invest hours in crafting useful, meaningful and telling poll questions - but even they can't nail every category of answer. So they try to nudge survey subjects into picking the "closest" answer. Completely understandable, but how many important nuances are lost in the process that would make a difference to us when we're weighing the polling results?
I think campaigns themselves NEED polls, and Lord knows they think the ones they pay for are, for the most part, useful and reliable. But their interests and needs are much different than ours.
Polling is, I believe, mostly accurate. (I'm not getting into the bogus polls commissioned by interest groups, or other such games.) But most folks don't understand what the results are showing, and without exception the results are utterly useless in the process of choosing a president. if the nets spent half the time comparing and debating tax plans that they do poll results, we'd be much better off. But who'd watch that (other than some of us)? And who'd call and e-mail their friends to say, "Didja hear what Brokaw said about McCain's tax structure?"
There is a science to sampling, and each firm develops their own formulas for interviewing X number of people in Y number of categories to come up with reasonably representative measure of certain opinions among the entire population. It's not as though the firms decide to interview 850, or 1,200 people, choosing those numbers arbitrarily.
I believe that over and over polls have been proven to be accurate. In addition, I confess noting the irony of some (both here on RRMB, and in he outside world) who will loudly and derisively ridicule the reliability of polling, and an hour later without a moment's hesitation while addressing a completely different subject, those same skeptics will passionately note that 70% of the country opposes the war in Iraq, or note that Congress has a 19% approval rating.
All that said, I also believe the vast majority of the polling done is simultaneously useless, and grossly misused - especially by the ignorant.
A fine example would be the polling done before the New Hampshire primary, an example often cited today by those who mock polling - even by RR - and generally regarded as recent "proof" that polls are "often wrong."
I don't claim to recall the specific figures, nor the names they were attached to, and I'm too damned laxy to go look it all up. But I will perhaps be forgiven for simply inventing figures for the sake of example, because I think you'll see the exact numbers and names don't really matter for purposes of the example.
In short, polling just before primary election indicated that Obama had a considerable lead over Clinton. Let's say it was something in the neighborhood of 30% for Obama and 20% for Clinton. Much was made of this difference. But in the end, Clinton won by a single percentage point. And so folks conclude, "The polls got it wrong!"
For those who looked at ONLY those two numbers, sure, it appeared the polls were wrong. Those 30% vs. 20% numbers were all folks talked about before the election, and so the results were clearly a big surprise.
Trouble is most folks, from pundits paid to know better, to the guy at the end of the bar, completely ignored that also in the polling data was a figure for those who would be voting, but who had not yet decided who they would support. It was a very high figure, and relative to the numbers in the scenario above, it would be fair to portray that "undecided" figure in the neighborhood of 20%- to 30%.
That percentage ultimately did vote, and they m,ade a choice, and their picks turned up in the results.
It's not the fault of the pollsters that the media and others portrayed the polling results inaccurately.
Beyond that clear record of the numbers being bungled, most polling is, I believe, pretty damned accurate, but utterly meaningless. last year's poll on how Clinton would do up against Romney doesn't mean a whole hell of a lot today, does it? That was money down the toilet!
Measuring Obama and McCain on a daily, weekly or even monthly basis is ludicrous. Again, I suspect most of the polls are accurate to within something like their stated margins for error, but they contribute absolutely nothing to knowing which of the candidates would make a better president, which is a liar, or which has the best plan for the future.
But the endless flow of numbers gives the networks something relatively inexpensive, something supposedly (but not really) easy to understand, and something that people flock to like flies to shit.
I'm always amazed when an anchor or pundit will announce with great gravity word that candidate X is behind candidate Y 41% to 44%, an important development because last week their positions were reversed, with Y at 40% and X at 43%.
Well both damned sets of numbers are within the poll's margin for error - a statistic that's ALWAYS noted, but only as a seemingly meaningless afterthought or technicality. In fact, the results of BOTH polls show the candidates are effectively tied, with no appreciable change during the week. But THAT wouldn't be so damned interesting to report, and hash over for the next 30 minutes, would it? It would put the kibosh on speculating whether candidate Y having slapped his wife at that fundraiser might have been responsible for the "precipitous drop." If they're still tied, no time goes to wondering if candidate X buying his kid a puppy might have helped bring about the "big comeback."
Finally, what happens every time there's a poll here at RRMB? A half dozen members observe that there wasn't an answer they were really comfortable with. Well, professional poster invest hours in crafting useful, meaningful and telling poll questions - but even they can't nail every category of answer. So they try to nudge survey subjects into picking the "closest" answer. Completely understandable, but how many important nuances are lost in the process that would make a difference to us when we're weighing the polling results?
I think campaigns themselves NEED polls, and Lord knows they think the ones they pay for are, for the most part, useful and reliable. But their interests and needs are much different than ours.
Polling is, I believe, mostly accurate. (I'm not getting into the bogus polls commissioned by interest groups, or other such games.) But most folks don't understand what the results are showing, and without exception the results are utterly useless in the process of choosing a president. if the nets spent half the time comparing and debating tax plans that they do poll results, we'd be much better off. But who'd watch that (other than some of us)? And who'd call and e-mail their friends to say, "Didja hear what Brokaw said about McCain's tax structure?"
Just like the viewer ratings of speeches, first they said Palin almost had as many viewers as Obama, she had 37.2 and Obama close to 40, now last now Palin number changed and now she had more viewers that Obama.
QUOTE (QBC @ Sep 6 2008, 12:26 PM)
Election fraud had no bearing on the 2000 or 2004 elections.
Oh really? That was the point of the exercise. It was wrought with fraud.
The pole serves as a natural beauty, a perfect balancing center at the forefront on the stage for every stripper to earn a dollar and show why she is deserving of $20 a dance; so yeah, I support 'em.
QUOTE (BluesBrian @ Aug 25 2008, 09:24 AM)
I only "trust" one poll... the only one that counts .. the one that is scheduled for November 4th. (Let's watch out for the "cheaters".)
I can't go that far. Due to Diebold I'm not sure our votes are counted any longer. Two of the three major machines used across the country share the same address in the Caymans; not the same suite as I recall but the same building.
QUOTE (Balor @ Aug 24 2008, 01:40 AM)
I've said before in other posts that I'd vote for Obama if the polls indicated a close race; otherwise I intend to vote for Cynthia McKinney.
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
Lately, though, I've wondered, "Why should I trust the polls?"
Given the last two elections were "close," and ostensibly close enough to hide election fraud, I then had to ask, "were they really that close, or did the polls just say they were?"
I keep hearing it said we have to turn out for Obama in such huge numbers that there can be no doubt as to the outcome. Trouble is, how do you go about proving it after the election is stolen? If it is stolen, what are you going to do about it?
Diane Rehm did a show on polls last week. i guess she and her staff are worried as well. i learned a few interesting things from this show. first, some news organization polls can be less reliable, the times poll was mentioned by name, b/c they ask the key questions, such as who are you voting for, in the middle instead of up front, which can skew the results. some of the more traditional, independent polls, such as zogby and gallop, were alleged to be more reliable. second, pollsters and news organizations that interpret the polls sometimes ignore the strength or weakness of a polled persons' answers. finally and perhaps most importantly here, as randi as stated mobile phone users are not polled, only land line users. pollsters do not adjust polls to compensate for this difference. apparently research has shown that land line users and mobile phone users do not respond to polls differently, so failing to poll mobile phone users does not skew the results. maybe progressives are fooling ourselves, but it is impossible to imagine that cell phone users, who tend to have access to a lot more information, will not be voting differently from land line users in this election.
as for my opinion on what to do here, i plan to just vote for obama even though i'm not in a battleground state. this is election is just too important to risk it on a protest vote. i love cynthia mckinney and still have a great deal of respect for nader. but, the republicans had better be overwhelmed on nov. 4th or we will be looking at another remarkably close election in which the republicans win by one or two points.
QUOTE (RoyPDX @ Sep 6 2008, 01:58 AM)
G. Made me wonder how a Methuselah who speaks in a monotone whisper while reading a list of cliches and nasty innuendos is even on the podium speaking.
H. Left me wondering if this might be the day he kicks the bucket.
QUOTE (QBC @ Sep 6 2008, 09:26 AM)
Yes to your poll question.
In terms of which ones, any of the polls done by the major independant pollsters.
Election fraud had no bearing on the 2000 or 2004 elections.
Given what's at stake, I would recommend you don't waste your vote on a 3rd party candidate.
In terms of which ones, any of the polls done by the major independant pollsters.
Election fraud had no bearing on the 2000 or 2004 elections.
Given what's at stake, I would recommend you don't waste your vote on a 3rd party candidate.
Speaking of election fraud... How did you like Uncounted the documentary?
http://www.commondreams.org
QUOTE (JRunRun @ Sep 7 2008, 02:32 AM)
[size=1][/size]
Speaking of election fraud... How did you like Uncounted the documentary?
http://www.commondreams.org
Speaking of election fraud... How did you like Uncounted the documentary?
http://www.commondreams.org
I haven't been to my blockbuster in awile. I won't make a special trip for the video, but will pick it up on my next trip - assuming they have a copy.
QUOTE (QBC @ Sep 7 2008, 04:54 AM)
I haven't been to my blockbuster in awile. I won't make a special trip for the video, but will pick it up on my next trip - assuming they have a copy.
Do you have cable or satellite? It was showing on Starz quite recently.
I look for trends in polls. If 10 different polls show McCain gaining on Obama, I trust that to be accurate.
I would like to believe they're all wrong. But if Obama was leading by 30 points, I wonder how many people here would be questioning that.
I would like to believe they're all wrong. But if Obama was leading by 30 points, I wonder how many people here would be questioning that.
i was polled yesterday by moveon i think, was unusual and cant rememer all the questions but it was long
QUOTE (rrrocksmahsocks @ Sep 7 2008, 11:27 AM)
I look for trends in polls. If 10 different polls show McCain gaining on Obama, I trust that to be accurate.
I would like to believe they're all wrong. But if Obama was leading by 30 points, I wonder how many people here would be questioning that.
I would like to believe they're all wrong. But if Obama was leading by 30 points, I wonder how many people here would be questioning that.
Part of the problem with polls in this election will be the youth vote. When these pollsters poll, they ask "likely" voters. Most pollsters do not include the 18-24 age group in their "likely" voter model because the thought is that this group doesn't vote.
The other issue is pollsters who do attempt to poll them, most in that age group don't have a land line, they have cell phones. If they live on a college campus and they listed their parent's phone number for their registration, they won't be reached there either.
If the primaries are any indication, the 18-24 year olds will be out in big numbers in November. This is going to throw traditional polling methodologies off.
Do you trust polls?
No.
No.
QUOTE (brotherdavid @ Sep 7 2008, 12:08 PM)
Do you trust polls?
No.
No.
I
you Brother David. 
QUOTE (Balor @ Aug 23 2008, 11:40 PM)
"Why should I trust the polls?"
I would ask, "What kind of polls?
Opinion polls? No.
Exit polls? Yes.
QUOTE
I would ask, "What kind of polls?
Opinion polls? No.
Exit polls? Yes.
Opinion polls? No.
Exit polls? Yes.
And so there's no way to predict who will win before the election? Just ask people after they voted?
Now the polls I'm seeing on the TV news show that its a dead heat between Obama and McCain.
Aren't we just witnessing the set up for another election that will be stolen?
Where is the overwhelming support for Obama that will override election fraud? You can't tell it exists by the polls now can you?
Jeez, you may as well draft petitions calling for your states and regions to separate from the Union.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.