Yay! It's really 43%-55% approve-disapprove. Stellar.

It also might be a good idea to review Rasmussen's "truly reliable" polls on the 2000 election You know, the Bush landslide.

Thanks for having absolutely nothing to add Pug...

I heard an announcement recently that something like 34% favor Hillary for President. There was a big whoop-t-do about it for about a day.

The message was clearly that , Hillary ALREADY has 34% of the people ready to vote for her in 2008!!! Get out the banners. This is great. She way ahead of the game.

Then I read the actually poll.

The poll had a couple of other names on it - and asked something like 'Who would you vote for' between the persons offered on the poll.

John McCain had 50 plus % and Hillary around 34%

The news left out the John McCain part.

LOL

I don't have time to look for that poll right now, but I'll see if I can find it in a while. I know someone will say it isn't so.

lol

You're absolutely right. Well, I can't say for certain that I saw that exact poll, but there was the poll about a month ago that said that already 51% of people said they would definitely not vote for Hillary, compared to only 16% who said they would. I don't know their sampling rate, but I think that 51% number is huge enough to make it clear she's got no chance of winning the general election.

http://www.rightwinged.com/2006/01/drudge_flash_only_16_would_vot.html

Calm down. Photos are always used in this way. Also, 34% is awful but 43% ain't exactly great - it is nearning the absolute bottom that can be reached short of Bush being impeached. People feel like the administration lied. It may be true, it may be false, but that's what politics is all about - perception, not reality. Karl knows that better than anyone! The Dems have sucked at getting their message out and have paid the price. It cuts both ways.

Also, 34% is awful but 43% ain't exactly great - it is nearning the absolute bottom that can be reached short of Bush being impeached.

(sarcarsm)Yeah because low poll number are an impeachable offense(/sarcasm)

Wow, that is one of the most idiotic comments I've received.

Now as for the photos, that wasn't really the main topic of the post, but I'll touch on it briefly. No they aren't "always" used like that, but I'll grant that they often are... but the press has been busted time and time again for running photos deliberately used to make the President look stupid. They've even photoshopped (the Condi devil eyes photo comes to mind) and been busted their too.

Now I don't have a huge problem with the photo Drudge ran, although it is still kind of dumb because that is the one thing that everyone can agree on about the President is that he doesn't care about polls at all. But come on, look at the photo CBS used! He looks like he's at a funeral, what a joke!

And no one said 43% is great, which is why I went in to detail, hoping to avoid idiotic and predictable comments like yours... The point is it is abundently clear that CBS wanted to make news rather than report it, so they skewed a poll in order to be able to report this new "all time low" of "34%"... well, they accomplished their task.. Drudge was right on it, and O'Reilly just has the TP memo and opening segment all about it... Luckily Tony Snow pointed out the 40% Dems polled to 27% Republicans, although I would have thought the Factor staff could have done that research themselves.

Thanks for playing Mark

Weighting corrects for the fact that the number of Democrats/Republicans/Independents polled was not exactly the same as the mix of Democrats/Republicans/Independents in the US population. In any poll (or in any statistical sample in any science), the sample measured won't be exactly the same as the total population for which you're attempting to generalize. If you understand how your sample differs from the entire populations, honesty dictates you apply a weighting factor. That's exactly what CBS did.

Your claim to have "busted" CBS is illegitimate. Your claim that Rasmussen Reports is more accurate is unsupported and highly suspect considering your previous inaccuracy.

I don't have any doubt that the majority of Americans feel this president is a complete failure.

sorry bud, but I already got the explanation from others... even if you take the "weighted" stuff, it's still disproportioned because in the last election it was 37% Republican and 37% Democrat... not to mention that this is a poll of "adults" rather than "likely voters" so it's really pointless anyway. It is quite clear what CBS tried to do here, and why their numbers are 5-10 points lower than most other polls.

We are giving CBS NEws too much credit. It is the worlds largest Liberal Blog.

That would put Daily Kos at #2.

Give it a couple of years for the last remaining CBS news viewers to die off and then Kos might be able to Merge with CBS News and provide it with new complient eyeballs.

You claim to have busted CBS - yet you've articulated no reason to doubt the CBS statement that 34% of Americans approve of this president.

You made three assertions.
1. You claimed the results are skewed because more Democrats were polled than were Republicans. You now seem to concede this point - this is what happens when a poll is performed randomly. CBS recognized this fact and applied a correction factor.

2. You claimed that the correction factor applied was incorrect "because in the last election it was 37% Republican and 37% Democrat." The CBS question is how many American approve of this president. CBS did not ask how many likely votes approve of this president. Therefore, using the Dem/Rep mix from the last election would be inappropriate.

3. Your last claim was that a poll of American adults doesn't matter - only a poll of likely voters matters. What does that have to do with the CBS statement that 34% of Americans approve of this president? It would be pointless to ask likely voters if they'd vote for GW - he's not running for election. The CBS poll is not attempting to determine how likely voters would vote. CBS made no attempt to extend these poll results to voting patterns.

I thought you conservatives prided yourselves on using reason rather than emotion. If so, use a logical argument to convince that CBS is wrong. Based on what I've heard from you so far, you have no reason to disagree with the statement that 34% of American approve of GW's performance.

What?! I don't get what you don't understand mjoeg... Follow the links to other blogs if you don't get it... but it's quite clear what CBS did here. They WAY oversampled Democrats, and that is just a fact. You are drinking some serious Kool Aid if you actually believe that was a random sampling. You want reason, here's the Rasmussen poll showing steady low to mid 40s numbers
http://www.rasmussenreports.com/Bush_Job_Approval.htm
how about all other polls that show the same thing
http://www.pollingreport.com/BushJob.htm

Look, no one is trying to brag about the President's numbers... that's not the point here. The point is this is just CBS being See BS again, and it's obvious to any honest person. Not sure why I'm defending the obvious, because if you don't see it it's because your'e choosing not to.

You don't get it? I'll review. You claim the CBS poll is inaccurate. You gave three criticisms of the statistical methology used by CBS. I showed the logical falacy of all three of your criticisms. You didn't respond to my criticsms. Instead

1. You bring up a fourth criticism of the statistical methology - you say the sample wasn't selected randomly. But you don't even try to provide evidence. The burden of proof is on you. You made the claim the the CBS poll was inaccurate.

2. You tell me to go to another website if I'm insterested in a rational agrument. Pathetic. I'm on your site because you claim that you broke this story. If you have valid criticism of the CBS poll, bring it. I doubt that you're intellectually capable of determining if the statistical methods used by CBS were valid or invalid. I can't fault you for being ignorant, but I do fault you for spouting off about something that's beyond your reach. That's my problem with you.

3. You ignore reason, instead you start talking about belief. I'm arguing with facts and logic. I believe in right and wrong. You're telling me you just don't believe. It's all relative to you. In western societies, we have a long and successful history of valuing rational and logical arguments over feelings. Do you come from an eastern mystical society where your feelings are considered your truth? How womanly, liberal and relativistic of you! You are unable to support your position but you say you know it because you feel it?!!? Did god whisper it in your ear?

If you care to argue about statical methods - that was the subject of your post - step up, my little friend. If not, admit you don't know what you're talking about.

I only directed you to another site because I am obviously not articulating to your liking.

See BS didn't weight their results fairly.. simple as that. They simply portray the country as having many more Democrats than Republicans to get their desired outcome. They may have added in "weighting", but it was still inaccurate... maybe not as inaccurate, but still inaccurate.

If you think CSB didn't weight the results fairly (let's use accurately instead of fairly - it's facts, not feelings that matter in statistics), find out what is the correct weighting factor. What is the actual percentage of Republicans/Democrats/Independents amoung American adults in February 2006?

You're getting closer to having a valid point. It is possible that CBS applied an inaccurate weighting factor and that it skewed the results. But it's just a possibility. You haven't presented any evidence. I don't know what the correct weighting factor is. You don't know what the correct weighting factor is either. I'm sure if you did some research, you could find out. You could even use the correct weighting factor to recalculate the CBS approval numbers. Do that and you may have some evidence to back up your claim that you've busted CBS. On the other hand, you may find out that CBS used an accurate weighting factor.

My only point is that claims are valid when backed up with facts; claims backed up by feelings are of no value. CBS released their statistical methods to support their claim that 34% of American adults approve of GW. If you want anyone to believe your claim that the CBS results are inaccurate, you need to be equally as rigorous in finding and presenting facts.

I would welcome a logically rigorous analysis of the CBS results. But no one cares about unsubstantiated opinion.

Latest approval polls:
Fox: 39
Gallup: 38
Q-poll: 36

Awww, is the rightwingers feeling sorry for themselves. They too afraid to admit they were wrong. And that they were ignorant enough to believe the BS coming out of the Whitehouse for the last 5 years. I guess republicans like to be lied to by our government.

Independent Rasmussen Poll

How Poll Questions Effect Answers

Who's Stronger on National Security?

Rasmussen Reports says 43 percent indicate they trust the Democrats more on national security today while 41 percent prefer Bush.

I will leave your babbling about poll numbers with words from a True Patriot. Think about it! Hitler was popular too.

"If Tyranny and Oppression come to this land, it will be in the guise of fighting a foreign enemy." ~James Madison

"It is a universal truth that the loss of liberty at home is to be charged to the provisions against danger, real or pretended, from abroad." ~James Madison

"No nation could preserve its freedom in the midst of continual warfare." ~James Madison

Rather than criticize their corrupt or incompetent politicians, Republicans criticize the media that exposes them.

lol, you guys are hilarious.

As always, rightwingers are off in fantasy land, ignoring the truth, and too proud to admit that they were ignorant enough to fall for the lies from the WhiteHouse. I bet if a poll on IQ was done for republicans, they would average around 54, slightly above the presidents 40.

And just like the current administration, this comment section will be censored and stripped of any negative feedback.

You wish so you could whine about something else... but I allow you people to make fools of yourselves and comment as you wish... I guess when you've got nothing else to say, complain about being censored when you haven't, right? Wow.

You poor folks are delusional and pathetic. You can spin like dreidels but it doesn't change the fact that America is finally waking up to the abysmal incompetence and frighteningly shocking arrogance that epitomizes the Bush administration. What exactly haven't they mucked up?

If you would remove the blinders you've worn in the name of partisanship politics, and shake off the neocon fairy dust, you might come to the same realization that the vast majority of Americans have - The only thing that drives this administration is power and the corporate money that feeds it.

Of course the fact that some folks continue to defend this administration despite its myriad of incompetences and law-breaking, only speaks to how well they have managed to brainwash with the use of fear and the aid of 24-hour news channels. That is perhaps, their only true legacy, and in 20 years, most decent, rational-minded folks will look back on Bush's tenure as American's darkest hour.

Randy,I think what Mark meant was that 34% is as low as polls can get without the president becoming impeached(as in his poll numbers would surely not go up during such an event)Thanks for playing Randy and your still a douche.

Ah.. the always civil left ladies and gentlemen...

Rob, I don't see how what you said is any different. Unless you're just not being clear, you comment seems to say the same thing, which is still ridiculous BS. How can the President be impeached simply based on low poll numbers? I see your "explanation" in parenthesis which would make sense, except that the previous sentence is still saying that the President could be impeached for low numbers, read your own sentence:

I think what Mark meant was that 34% is as low as polls can get without the president becoming impeached

Beyond that, your little explanation in parenthesis is actually quite ridiculous and pointless anyway... Of course the President's poll numbers wouldn't go up if he were impeached. These are job approval numbers, and if he wasn't in the job, there wouldn't be a poll!

No sir, thank you for playing!

This is a textbook case of being a nimrod.Your counter argument is (sarcarsm)Yeah because low poll number are an impeachable offense(/sarcasm)
NOBODY said that.The spirit of Marks post wasnt that he would be impeached FROM low poll numbers but that they couldnt get any lower UNLESS he was being impeached.I really hope you aren't the beacon of light for the rightwing movement cause were all screwed if this is your attempt at critical interpretation....and your still a douche

you know what? I actually just went back and read Mark's comment and realize that it could be interpretted either way:

Also, 34% is awful but 43% ain't exactly great - it is nearning the absolute bottom that can be reached short of Bush being impeached.

Maybe you're right about what he meant to say, but he never clarified... You're trying to say that he meant: "The only way Bush's poll numbers could go any lower is if he were impeached".... I can't blame him if that was the message he meant to convey, and in his own mind thought he did. Everyone's guilty of saying things that make sense in their own head, but aren't clear to other people in the way they say it. This is often the difference between good and bad teachers.

Anyway, then you said:

I think what Mark meant was that 34% is as low as polls can get without the president becoming impeached

That further goes towards my original interpretation of Mark's comment. Again, perhaps that's not what you meant to say and you could have been more clear...

But again, all of that is moot, because who takes job approval numbers of citizen George Bush? If he's not President there is no poll!!! Get it? It just doesn't make any sense which is why I would be forced to interpret the comment's of both Mark and you the way I did, because if the President were to be impeached there wouldn't be poll numbers! Are you getting this?

By the way, after the last F911 screening in your freshman polisci class, did they tell you that saying "you're a douche" is an argument winner? Bravo on that... You got me!

It wasn't inserted to win an argument.It was only an observation inserted after the fact.If it works for Ann Coulter it can work for me too.Every Republican I know loves 'ad hominem' attacks.Just something I love about being a conservative.

Perhaps a course in statistics would have helped you understand polls. But, instead, you took the 101 class in Intelligent Design. And probably failed.

yeah, cuz that proves that CBS used proper weighting in their poll Blastin... great argument as always. If you have proof that the country is 28% Republican to 37% Democrat (the weighted figures) then bring it forward, otherwise stop whining...

hold it, randy, you broke the story about CBS using flawed statistical methodology. you need to bring the evidence. we're still waiting. you must be preparing a bombshell - it's taking you several days.

just for kicks, Harris polls found that in 2005 the country was 30% Republican and 36% Democrat. With all GW's screwups in the last year, 28% Republican and 37% Democrat in 2006 seems reasonable.

http://www.harrisinteractive.com/harris_poll/index.asp?PID=630

well "seeming reasonable" and having information to support it are different things. also, what is this footnote about, that I just found at the bottom of the CBS poll:

This poll was conducted among a nationwide random sample of 1018 adults, interviewed by telephone February 22-26, 2006. The error due to sampling for results based on the entire sample could be plus or minus three percentage points. An oversample of African Americans was also conducted for this poll, for a total of 207 interviews among this group. The margin of error for African Americans is plus or minus seven points.

I don't know how that factors in exactly, but since it is known that blacks extremely disproportionately vote Democrat vs. Republican and that is an extremely disproportionate amount of blacks sampled, it's a bit odd.

Anyway, on to real stuff.... I'll answer your Harris poll with a NEWER Pew poll, done in February that says the breakdown is 30% Republican and 33% Democrat. (pg. 13)

http://people-press.org/reports/questionnaires/270.pdf

There ya have it. You can rely on Harris if you choose to, but Pew has numbers from just last month.

i agree that an uncorrected sample that oversampled african americans would not represent the real American population. i'm not sure that's what was done - there are several questions that compare african american sentiment to general amerian sentiment. CBS may have sampled african americans to be able to make those comparisions only. the published methods don't make it clear if the oversample of african americans was included in the results of all questions or not ("an oversampling of african americans was alsoconducted for this poll"). you can't make assumptions one way or the other, but in science you lean toward the conservative case (i.e., the results are invalid).

i recalculated the approval numbers using approval % for Repub, Dems and Independents from CBS and using 30% Repub, 33% Dem, 37% Independent from the Pew poll - ends up at a 36.3% approval rating.

furthermore, if black republicans and independents approve of GW less than other republicans and independents, result would be off by more. i have seen another poll that says 77% of republicans approve of GW - CBS says 72%. this would skew results by 1.5%.

Assuming most but not all blacks dispprove of GW, you might be able to bump it up to 38-39%.

you may have a valid criticism - i'm glad we using facts now. the CBS results could be off by 3-5% points. that is, if CBS used the oversample of african american for all results.

I'm not whining. I just think it's hilarious how far people will go to defend a failed presidency. Can you actually name one good thing the president has done for this country. Even his tax breaks are going to cause problems for many generations to come. I suppose you actually think he's trying to spread democracy with this war in Iraq. And now that our country is nearly broke, how is he going to fund a war with countries that are ACTUALLY A REAL THREAT to the U.S.

nice dodge now that I brought numbers to shut you up... classic liberal... but if you wanna do that, then I will wait for you to slam Clinton for making the same, and stronger arguments for military action in Iraq and specifically pointing out why he chose them over other WMD nations
http://www.rightwinged.com/2006/03/flashback_presidents_comments.html

Back to reality. Since 1998, much intelligence has been shown that Saddam did not have any WMDs. Bush ignored this intelligence, and decided to focus on claims HE KNEW were false. Bush was told repeatedly that Saddam had no WMDs right up to the start of the war. And I'm not a liberal. I'm actually quite conservative. And, If Clinton would have started this war by lying to the American people, I would slam him too. I believe a president should at least have enough integrity to not lie and talk down to his people. Bush lies consistently. That is a fact. This war has killed more Iragi civilians altogether than Saddam ever has, and Iraq is far worse off than it ever was. The president assured us this would be walk in the park, ignoring all expert claims to the contrary. And, by the way, there were no WMDs found, and Iraq had nothing to do with 9/11. The majority of the highjackers were from Saudi Arabia, and I don't see anyone giving them a hard time. And, I don't see anyone going after the real mastermind of 9/11.

I'm not a liberal

lol, that is the classic far lefty statement before commenting on all conservative blogs or radio shows... so pathetic and transparent. Besides, only a liberal could be idiotic enough to say some of the things you just said, like:

This war has killed more Iragi civilians altogether than Saddam ever has

Stupid as they are, most liberals wouldn't even make such an idiotic comment

And, I don't see anyone going after the real mastermind of 9/11.

I'm sure the troops along the Afghan/Pakistan border would be glad to hear that from you Blastin.

Iraq had nothing to do with 9/11.

I guess not, sense Clinton launched military strikes there long before 9/11 for the same arguments, and even answered the question about why Iraq and not other WMD countries, which you all pretend you want Bush to answer.

The rest of your "Bush lied" crap is just pathetic and ridiculous. Enjoy freshman polisci today... I heard your professor will be letting the class check out the special features on the F911 DVD! Wahoo!

You are so far out of touch with reality it's insane. Bush has been caught lying so many times. He assured us no one in his administration leaked the Plame name. That was a lie. If it wasn't a lie, then why didn't anyone in his administration tell him otherwise, after all it was his vice president that authorised the declassification of the information so Libby could leak it. And the leakers were not fired as he promised. Cheney consistently lied about Iraq's involvement with 9/11. They both lied about WMDs when they knew there wasn't any, that's a fact, there's no disputing it. He lied about the port deal, the 45 day delay is a scam, the deal already went through as far as he is concerned. He lied about Katrina and continued his vacation for three days after he know people were dying and the federal response wasn't working. I admit there's people looking for Osama, and I sincerely hope they catch him, but if the effort we put into Iraq went towards catching OBL, then I'm sure he'd be in custody right now.

And just what is your own personal reasons for wanting the Iraq war. Is it to spread democracy? For the oil? To keep America safe? Do you even know, or are you blindly following the Bush administrations guiding voice?

And I still would like to see someone come up with one good thing the current administration has done for this country. Basically my whole point is, it's no wonder his poll numbers are low, he doesn't do anything good at all. I'm surprised the numbers are as high as they are. If they did a poll at Haliburton, I'm sure it would 100% favorable. But, the whole world doesn't work there.

And to show I am a bit conservative, my biggest hope is that the current administration wouldn't put so much effort into covering their butts from all their mistakes and lies, and actually started trying to do some good for this country. I know we can't back out of Irag right now, but we sure could use a different form of diplomacy to help clean up this mess. And the far left is spending way to much time nitpicking about every little mistake. But, I can see why they are so angry, there's so many mistakes being made. At the same time it's hard to understand why the right stands by this administration so whole heartedly. Doesn't the right want something to come out of the 8 years this administration has coming to it? So far, I think a grade of 34% is about all the Bush administration deserves, and that's being liberal (hehe).

Wow bud, this one's for you:
http://www.rightwinged.com/images/photoshops/koolaid.jpg

Anyway, it would be a full time job to answer your lefty talking points... you know full well this is just what your type does. You go on to conservative blogs spew out all the normal "Bush lied" crap from a million different unsupported angles, and know that you could keep a bogus argument going with your ranting and raving for a year and I've been there before and it's like, why bother? Nothing will make you be honest, and you've chosen to believe what you do, so I'm not going to waste my time. Suppose I were to answer your question about what Bush has done good... since you've already decided that he hasn't done anything, you'll never let up and I'll be spending all my time entertaining you when there are much more important things going on. Why don't you spend time figuring out how to get your guys elected and instead of whining over the same things you've been whining about for years.

Well, you passed the test. I've been looking for conservative bloggers to help write articles on my new website: www.itsfizzy.com. And you passed with flying colors. You stood by your man faithfully no matter how bad things get. So if you have some spare time, please help me get www.itsfizzy.com running.

In all honesty I think you really lost that debate Randy.There were many segues you could've used but you chose the classic 'Deny em and slam em' technique.Kudos.

Also realize this is a VERY old conversation.

huh?

I'm not really sure what you're saying Rob... I provided a link that proves the sampling rate that CBS used was BS... then Blastin comes in with the normal insane talking points like:

They both lied about WMDs when they knew there wasn't any, that's a fact, there's no disputing it.

Anyone who says that loses any ounce of credibility (not that Blastin had any, anyway). But there was no other debate left. Even mjoeg, when presented with that evidence admitted that the CBS numbers were suspect (by the way, this was like 2 weeks ago anyway). Blastin' brought nothing but a bunch of idiotic whining liberal talking points that are nothing more than annoying if it weren't for the fact that the media tries to push them too. But it's the traditional method they use. They launch out at least a half dozen different talking points so that you'll be busy tearing down each one and they just throw out more and it just becomes a scavenger hunt for links to destroy the liberal "arguments"... It can go on for months if you want it to, so long as libs don't ever have to address the original point about the biased study. Notice that this is what the post was about, and Blastin is just complaining about defending the President. That's irrelevant to the post, but nothing out of the ordinary for a liberal to do. Diversion and talking points is the name of the game.

Randy,
You pointed out that the methods CBS reported weren't clear. The methods said CBS "also oversampled African Americans". Did they use the oversample for all results or only for a subset of the questions? Neither you nor I could definitively say.

I ran across some pertinent information a couple week back, but Ive been busy. There is a more recent CBS poll. The methods in that study state the CBS oversampled miltary families. And the results? 34% of the sample approve of GW's job performance.

So either
1) the oversample data is used (as it should be) only for those questions that begin "among African Americans" or "among military families", or
2) military families, more than the general population, disapprove of GW's job performance

I think you'd go with #1, as would I. We're left with no evidence that the CBS poll was biased.

And what about all the other polls that show similar results - approval ratings in the mid to high 30s? Do you think all these polls are biased? Even Fox News and the Wall Street Journal?

By the way, you referred to me in another post as a liberal. I've shown myself to be moderate and reasonable. What gives?

I don't recall, though I'm sure I did call you a liberal... Perhaps you have always come off as a moderate and your comments seemed like piling on with other libs... not sure...

As for the CBS poll, regardless of the oversampling (which it would be nice to have a link to back up what you've said) for blacks or military families, the fact is that they oversampled Democrats to Republicans, based on what we know about the make-up of the country, from the Pew poll I cited:

http://people-press.org/reports/questionnaires/270.pdf

No one is claiming that W's numbers are great, we all know they are in the toilet. But it makes headlines to have a "NEW LOW", so I contend that it fits in with CBS known bias that they sought to return a "new low" to create, rather than report a story. Pew is actually the only other polling company to return that low (actually 33%) of a number, though I don't have all the information to disect their methodology... Notice that all others have stayed 36%+, mostly 37-39.

Again, based on the political make-up of the country, CBS oversampled Democrats even after they "weighted" it, and I believe it was deliberate to get the lowest of the low numbers to create a story. They should explain why they polled so many more Dems than Republicans if they have a real explanation, which I doubt. Perhaps Rather's unimpeachable sources could fill us in.

by the way, for support of my claims that the media wants that "new low" story, the AP tried to do the same thing, only flat out lied about a "new low", when it was just a return to low:
http://www.rightwinged.com/2006/03/poll_wars_ap_misleads_with_pol.html

Randy
Your accusations are as follows:
1) CBS used incorrect statistical methods
2) CBS did so to purposely mislead
3) Therefore, CBS lies in order to manipulate public opinion.

In support of your first point, you've throw four criticisms of CBS's statistical methods at the wall. But none have stuck. I've shown you the logical fallicy of each of your criticism. Now you've moved on to a fifth criticism. I tire of this. Your new argument is weak. The fact that a Pew poll found a slightly different Republican/Democrat ratio than the CBS poll proves nothing. Ever poll has a margin of error. If you and I were to randomly call a thousand people (that's the size of most of these polls), we wouldn't find exactly the same number of Republicans and Democrats. Does that mean one of us is using inappropriate methods and intentionally attempting to mislead? Not at all. We have no reason to believe the Pew poll is more accurate than the CBS poll. And in fact, Pew found an approval rating lower than CBS. The Pew poll therefore supports the CBS conclusion that 34% of Americans approve of GW's job performance.

Your claim that CBS used incorrect statistical methods is unsupported by evidence. So your claim that CBS did so intentionally (which you haven't even tried to support with evidence) is irrelevant. Your conclusion that CBS lies to sway public opinion, therefore, is baseless.

Randy, it's disturbing that you don't seem to let the lack of evidence impact your claims. Are you finding facts and forming opinions or are you only looking for facts to support your existing opinions?

I hear a lot of righwingers (though I haven't heard it from you) say that Democrats just throw claims at the wall hoping something will stick or that Republicans will tire of defending themselves. I agree Democrats and Republicans use that tactic. Isn't it precisely what you're doing here? Honestly, you don't know much about statistics, but the CBS results just don't feel right to you, so you find something that make sense to you and claim you've proved CBS intentionally lies. When that doesn't pan out, you find something else. Repeat again and again. When there's no evidence left, you just stick with your belief as if your errored accusations are enough evidence.

I agree with you that every media outlet wants to report on sensational events. Are they trying to sway public opinion or are they trying to make a money? Years ago the media told us repeatedly that some pesticide used on apples was going to kill. Turns out they were wrong. But I haven't heard the media reporting apples are safe to eat lately. Why do they only report the bad news coming out of the apple orchards? Obviously, the media hate apples. And America. But they love oranges.

CBS has a proven track record (i.e. Rathergate, The Wall Within, etc.) ... take it or leave it.

I dont' see where you've proven anything here. The fact is, Pew asked a question as to political affiliation, where as CBS seems to have an assumption about the make-up of the country. They should explain why they used the rate they did. It wasn't like they oversampled Democrats just slightly.. it was by a rather large margin.

If you believe you'd debunked the theory here, you're welcome to it... But given CBS obvious and known bias, it seems quite clear to me that they set out to hit a new low and oversampled just enough to do that. Just as I showed you the AP was overexcited about a "new low" that they lied in their headlines, when it wasn't a "new" low at all. Other polls were all registering in the upper-mid 30s to low 40s and CBS knew that the bit "34" number would get the extensive coverage it did. Just like they knew the National Guard fake story would... but in that case they were just too excited about bogus documents to check them out... this time I believe they purposely sampled the way they did to get a lower number without it being so shockingly low that it would come under mainstream scrutiny.

Again, Pew found a LOWER approval rate than CBS. If you're trying to argue that CBS found an inaccurately low approval rate, the Pew polls hurt your case.

What's all this about Rathergate, the Wall Within, blah blah blah. This tactic reminds me of something you said just recently:

"They launch out at least a half dozen different talking points so that you'll be busy tearing down each one and they just throw out more and it just becomes a scavenger hunt for links to destroy [their] "arguments"... It can go on for months if you want it to"

"If you believe you'd debunked the theory here, you're welcome to it." Great I'll stick with the facts.

"But given CBS obvious and known bias, it seems quite clear...I believe they purposely sampled..." And equally great, you stay solely in the realm of opinions. Tin foil hats and koolaid for all conspiracy theorists. Particularly those who don't let evidence stand in the way of their beliefs.

First of all, I began replying once already but trying to pull up this CBS poll's pdf, it caused and error and made by browser shut down... Tried it again but it's not pulling up.

Anyway, for starters, you see it as Kool Aid, I see this as reaffirmation of what we already know about CBS. As for the Pew Poll, we're talking about two different polls. I don't know what methodology was used in the 33% poll.... But I know that the other poll asked a specific question about party affiliation, and it was the most recent I could find with that question, and it shows CBS used an unbalanced sampling rate. Simple as that. Now if you work for CBS or can get someone to explain why they did that, that's fine. But I think it would have been smart for them to include that in their poll, no?

As for the "more recent" poll, for starters I don't see a link to the full PDF poll with questions, etc. Just their summary article. But it points out something interesting you left out (for whatever reason)... They didn't oversample military families, they did them as a separate poll. And the CBS piece mentions nothing about Bush job approval among those families, it only mentions their feelings on whether Iraq is going well or not.

http://www.cbsnews.com/stories/2006/03/13/opinion/polls/main1396372.shtml

Fifty-eight percent of Americans in the military (or with family members who serve) and 52 percent of those who have served or have family members in Iraq think things there are going badly for the U.S., about the same percentage as other Americans. Just over four in 10 of each group thinks the war is going well.

So your point about that poll is totally irrelevant unless you have more information on it.... I suspect you don't or you would have already provided it.

Back to the oversampling of blacks from the first poll... I don't recall (and again the PDF isn't working right now) it saying (as you indicate might be an explanation) "among African Americans"... I could be wrong, but I don't remember seeing that.

Please stop wasting my time forcing me to dig out links to debunk your claims... If you've got some smoking gun that proves CBS's sampling rate correct, bring it. If you have something to prove your "military families" thing, bring it. Otherwise stop wasting my time on this old post.

Since you’re having trouble finding the links:

The latest CBS poll (http://people-press.org/reports/questionnaires/271.pdf) found that 34% of Americans approve of GW’s job performance. CBS oversampled military families and also oversampled Republicans. 358 Republicans were polled while only 366 Democrats were polled. That’s 32% Republican, 32% Democrat polled.

If I were like you, I’d claim that I had busted CBS – that they’re trying to artificially inflate GW’s approval numbers by oversampling Republicans. There is no way the country is evenly split between Republicans and Democrats. But I’m honest and I can see that CBS weighted the results to 34% Democrats, 28% Republicans.

If I were like you, I’d then say CBS intentionally applied an incorrect weighting – that they should have weighted to 36% Democrats, 25% Republicans because some poll somewhere found that 36% of American were Republicans. But I’m honest and I know polls typically have a +/- 3% margin of error. Results vary slightly. In fact, you seem to favor the Pew poll for some reason, so I’ll go with that. The latest Pew poll (http://people-press.org/reports/questionnaires/271.pdf) finds 34% Democrats and 28% Republicans. And these results are slightly different than Pew’s February numbers.

If I were you, I’d then complain CBS oversampled military families and that somehow confounded the results. But I’m honest and I see CBS applied a weighting for the oversampling of military families (in exactly the same way they applied a weighting factor for the oversampling of African Americans - look it up http://www.cbsnews.com/htdocs/pdf/poll_bush_022706.pdf).

If I were you, I’d say despite the lack of evidence, I believe/feel/conjecture/speculate that CBS is sneaky and hid they way the inflated GW’s approval numbers. But I’m not like you. If there is not a shred of evidence, I reject the conspiracy theories.

It's not a conspiracy theory bud, CBS has a record of liberal media bias.. now you're becoming quite a waste of time... but let's just establish a couple things here:

YOU have just continued to assert this oversampling of military, but you have offered NO EVIDENCE to back up that claim. Now you're claiming to link to things that you're not even linking to... read your first paragraph:

The latest CBS poll (http://people-press.org/reports/questionnaires/271.pdf) found that 34% of Americans approve of GW’s job performance. CBS oversampled military families and also oversampled Republicans. 358 Republicans were polled while only 366 Democrats were polled. That’s 32% Republican, 32% Democrat polled.

You're linking to Pew dude, not CBS... Nice try though.

As for the African American weighting on the actual CBS poll.. it should be noted that they've recently added this, and it wasn't in there before, and some other things look different now too, but I'm not sure... I am sure that they didn't have that weighted/unweighted African American section in there to begin with, and they should have, instead of casually mentioning an oversampling of African Americans in an earlier note, and that's it.

Now back to Dems to Republicans... CBS sampled 37% Dems to 28% Republicans, agreed? Where is that number from? Why do they not provide an explanation for using these percentages of party affiliation? As I said, the most recent poll that asks the question was the PEW poll (a different one than the approval poll you're citing) says the numbers are actually 33% to 30%, though I can't tell if this is registered voters, or adults (my guess is adults). CBS poll is random "adults". I guess that the pew affiliation quesiton is is adults, mainly because in 2004, registered voters were split 37/37, and Fox has been reporting a poll the past couple days that now tilts by one percentage point in favor of Democrats.... Any way you slice it
CBS OVERSAMPLED DEMOCRATS!!!

But back to your claims about oversampling military and military families in a more recent poll... Are you ever going to try to back that up? Are you aware that NOT ONE OF THOSE LINKS has anything to do with that? After you initially made the claim, I went and dug up the link to the story that obviously confused you (see previous comment) and noted that they didn't oversample military, they simply had a side poll for military, and all that CBS reported was their feelings on how the Iraq war was going, not on approval.

Now if you have another link that you'd like to provide to make an attempt to support your claim, go right ahead, but know that you have yet to link to anything about military families being polled or oversampled... You've linked to a different Pew poll (than the party affiliation one) that gives a 33% approval, but doesn't tell us whether it's adults or registered voters, or explain their sampling rate by party, and you've linked to the original CBS poll PDF, which again they've added the African American weighting information to. What don't you understand? You just made a complete waste of a post and now I've had to spell it all out for you.... please either dig up a link to support your claim, or just stop already

on other quick point.. that Pew poll with the 33% approval asked the party affiliation question (again, I don't know if these are adults or registered voters, but I guess it doesn't matter much)... it says 34% to 28% Dems/Repubs (which again, leads me to believe it was probably a poll of adults)... but it still doesn't match up with CBS's rate (37/28), which again, they provide no basis for. All CBS would have to do is add a note for why they sampled at that rate and there wouldn't be an issue here.. but they don't and that's the point.

I'm wasting your time, or you're frustrated because you're losing this argument?

1. "YOU have just continued to assert this oversampling of military, but you have offered NO EVIDENCE to back up that claim."

Absolutely. I'm being sarcastic. The oversampling of the miliary families did NOT invalidate the latest CBS poll results. Just as the oversampling of African Americans did NOT invalidate the previous CBS poll results, as you claimed it did. It's got to be one way or the other - either both polls were biased by the oversampling (in opposite directions) or neither poll was biased. Yes, oversampling = a side poll. In both polls.

2. If CBS added details to their first poll 's methodology PDF, isn't that a good thing? You've been complaining that CBS doesn't explain the methodology in enough detail. Then when they explain themselves, you complain that they recently added detail.

3. "As I said, the most recent poll that asks the question was the PEW poll (a different one than the approval poll you're citing) says the numbers are actually 33% to 30%"

WRONG. There's a more recent Pew poll. It's the one for which I provided the link. The most recent Pew results are 34%/28%.

4. So now all the evidence you're left with to prove that CBS is biased is that they used a 37%/28% weighting factor when other polls have found that a 34%/28% weighting factor to be appropriate. Assuming that's 34/28 is more accurate (for the sake of argument only), Bush would have a 34.5% approval rating instead of a 34% approval rating??? That's all you've got? CBS says the margin of error is +/-3% but you're claiming that because the CBS results could have possibly been off by as much as 0.5%, this proves a liberal media bias?

5. "All CBS would have to do is add a note for why they sampled at that rate and there wouldn't be an issue here.. but they don't and that's the point"

Wrong again. The point is that you claimed that you could demonstrate that CBS intentionally used inappropriate statistical methods, proving that CBS has a liberal bias. Now you're backing off your claims. Now your only claim is that CBS didn't provide enough information for you to determine if they used appropriate methods. That blows your conclusion - you have provided no evidence whatsoever here to demonstrate that CBS is biased.

The correct URL for the latest CBS poll is http://www.cbsnews.com/htdocs/pdf/031306_poll_iraq.pdf

For starters, next time provide the links you claim to be providing before posting... Now I've checked out your poll that seems to have oversampled military families, although as I said, pointing to the article that is tied in with this, CBS says it was a separate poll:
http://www.cbsnews.com/stories/2006/03/13/opinion/polls/main1396372.shtml

A separate poll finds the negative feelings about the war even extend to Americans with close military ties to Iraq (either serving there themselves or having a family member there).

So I'm not quite clear on that seeming contradiction... but assuming you were right, notice what CBS does in the part where they note the oversampling:

This poll was conducted among a nationwide random sample of 1136 adults, interviewed by telephone March 9-12, 2006. The error due to sampling for results based on the entire sample could be plus or minus three percentage points. The poll includes an oversample of people who have served in the military in Iraq or have an immediate family member who has served there. There were a total of 247 interviews with this group (see separate release). They were weighted to their proper proportion within the total sample.

Do you see any such clarification here (from the previous poll):

This poll was conducted among a nationwide random sample of 1018 adults, interviewed by telephone February 22-26, 2006. The error due to sampling for results based on the entire sample could be plus or minus three percentage points. An oversample of African Americans was also conducted for this poll, for a total of 207 interviews among this group. The margin of error for African Americans is plus or minus seven points.

Absolutley not! Why couldn't they simply say the same thing there? It was obviously journalistically necessary to make clear that the families were weighted fairly, why not the blacks and why not say the same about why the oversampled Democrats. Even if it were 34.5, it still doesn't explain their sample rate.

Anyway, you might have wanted to check out the newer poll your citing before posting it... It's reenforcing my point.. In the newer one they use a rate of 34% Dems to 38/39% Republicans... Again, the more accurate percentage... Though they have a huge amount of independents, but don't break down, as many polls do, the usual "leanings" of those independants.. But that's fine, I won't worry about that much because it's speculation... Let's just assume this newer one is on the up and up all the way (despite the fact that virtually every other poll still shows upper thirties to low 40s)...

But the fact is, YOUR NEWER CBS poll has vindicated MY POINT. Why wouldn't they use the same Repub/Democrat sampling rate for both polls? Or if there was a swing of more people becoming independants and leaving the Democratic party, wouldn't that be something for them to note? Simply, why are they using two rates, especially without explanation at all? THAT'S THE POINT!

The number isn't really all that important, because as I've said, no one is questioning that his numbers are in the toilet.

1. "Why wouldn't they use the same Repub/Democrat sampling rate for both polls?"

You're showing your total ignorance about how polls are conducted and statistics applied. Polls are not conducted at a predefined rate. Unless one polls randomly (that is, not knowing the demographics of those you're calling), one cannot apply statistics. This is stats 101. You should go to the library and find a book. If you have any evidence to show that the polls were not conducted randomly, bring it.

2. Why didn't the news article that accompanied the release of the first poll include the phrase "They were weighted to their proper proportion within the total sample"?

Because it's unnecessary. The methods provided by CBS not only show that the results were weighted in both polls, but they show the exact weighting factors applied. Anyone who can read can see that the results were weighted in both polls. It's right there in print.

3. The newer CBS polls vindicates your criticism of the earlier poll.

Wrong again. Quite the contrary. Both polls found a 34% approval rate. What this shows, Randy, is that any differences in the polls had no impact on the results - 34% of Americans approve of GW's job performance. Again statistics 101. Give it a try.

You have no evidence that even hints that the CBS poll is anything less than 100% accurate.

lol, no one is saying that the sample is random and thus different for each poll... but the weighting is to correct oversampling to show the results at the proper sampling rate... the weighting should match for both polls, it's as simple as that. Why would they correct/weight using 37% in February, and 34% in March? Again, if they make a point to explain that they weighted the military families to the proper proportion, why wouldn't they say the same about the blacks in the previous poll? You say it's not necessary, but don't explain why... Wouldn't it make sense that if they add the footnote in one poll, they'd do it in another?

As I said, the actual number isn't the issue.. the issue is CBS needs to explain why they decided that 37% of the population were Dems in February, but 34% (the more accurate, according to other polls) were in March.