A tale of two polls

Most political observers know to stop reading when they see the phrase, “According to a poll from Zogby Interactive…” and yet, yesterday, I couldn’t believe the commotion caused by a poll that obviously didn’t make any sense.

About 24 hours ago, two polls came out -- a Zogby Interactive poll (with questionable methodology) showing Hillary Clinton struggling against the top GOP candidates, and a Gallup poll (with more reliable methodology) showing the opposite. Guess which one got too much attention?

While the Zogby poll was mentioned by multiple reporters and pundits, the only mentions the Gallup poll got on TV were from Hillary advisers who had to bring it up themselves on the air in order to inject it into the conversation.

Of course, every political reporter, editor, and producer in the country knew that Zogby Interactive results were unreliable, but they trumpeted the results anyway.

Wouldn’t responsible journalism require news outlets to a) note why professional pollsters discount Zogby Interactive data; and b) also highlight the Gallup numbers with equal enthusiasm?

Related Reactions

Advertise Here

Login or Register to post comments.

78 comments

"Wouldn’t responsible journalism require news outlets to a) note why professional pollsters discount Zogby Interactive data; and b) also highlight the Gallup numbers with equal enthusiasm?"

You'd think, wouldn't you?

Here's another tale of two polls, where the difference is more striking.

A new CNN poll of Florida gives a radically different result from another poll released earlier today, which showed Mike Huckabee surging into second place. The CNN numbers: Giuliani 38%, Romney 17%, McCain 11%, Thompson 11%, and Huckabee 9%. The margin of error is ±5.5%.

The other poll, from GOP firm Insider Advantage, had Rudy at 26% and Huckabee with 17%. It had a ±3.8% margin of error.

Obviously, at least one of these polls is wrong.
http://tpmelectioncentral.com/2007/11/another_poll_huckabee_not_second_i...

dont let this go away C and L. Post this again first thing in the morning.

Wouldn’t responsible journalism require news outlets to a) note why professional pollsters discount Zogby Interactive data; and b) also highlight the Gallup numbers with equal enthusiasm?

I'm afraid that for the most part "responsible journalism" is pretty much an oxymoron these days. At best, the hacks just want a man-bites-dog angle; at worst---well lets not go there right now. The ironic thing is that the only reason a poll (valid or not) showing HRC struggling against GOP candidates is man-bites-dog is that the OM (old-media) has spent so much effort to date to anoint her the "front-runner". Basically, the OM is getting bored with their own story line and are trying to drum up some interest: what's a horse-race if the leader is 10 lengths ahead? They could get off their fat duffs and delve into the candidate's view and history, but that would be too much like work.

"Wouldn’t responsible journalism require news outlets to a) note why professional pollsters discount Zogby Interactive data; and b) also highlight the Gallup numbers with equal enthusiasm?"

oh, you are a dreamer, aren't you?

Here’s another tale of two polls, where the difference is more striking.

A new CNN poll of Florida gives a radically different result from another poll released earlier today, which showed Mike Huckabee surging into second place. The CNN numbers: Giuliani 38%, Romney 17%, McCain 11%, Thompson 11%, and Huckabee 9%. The margin of error is ±5.5%.

The other poll, from GOP firm Insider Advantage, had Rudy at 26% and Huckabee with 17%. It had a ±3.8% margin of error.

Obviously, at least one of these polls is wrong.
http://tpmelectioncentral.com/.....lorida.php

I think the stated "margin of error" on most of these political polls is utter nonsense.

Responsible journalism ... do they teach that anymore ?

They also control the message. How about a real debate where real people ask real questions that have not been vetted by the network?

Oh the poor Hillary supporters had to bring it up themselves while on national television or syndicated radio shows...oh how ever will they get their message out about how wonderful life would be under a democratic war-monger.

Sorry, I couldn't resist. My candidate won't even get to bring up issues at an actual debate...forum...whatever.

Zogby is literally bought and paid for...I mean how else could they pay their staff?

Kucinich!!! Forever simone!

Anyone who believes polls aren't skewed to further a cause or manipulate the public is an idiot . This includes Gallup , Zogby or the Pepsi Challenge .

As absolutely worthless as all polls are, we are constantly bombarded by them in the MSM, Fux and the left. Somebody finds enough people to say they would vote repug rather than vote for Billary. Why does anyone find that surprising? I'm sure I can find enough people to tell me they think chimpy and darth are aliens to make it seem statistically significant. We see polls every day that 80% of Americans want the war to end, but hundreds of billions more will be spent and hundreds of thousands more lives will be lost. Those polls don't mean shit.

Just because a poll "shows" something doesn't make it a fact. And why should anyone be surprised that a controversial poll gets attention? Our country is full of complete and utter morons who don't comprehend enough statistics let alone reality to understand that a poll doesn't mean shit. The only polls that makes any difference are the elections we have every two years, and they seem to be rigged as well.

the gop can't win the 2008 elections without the support of pundits and liars to help them lie their way ALL THE WAY to election day. that is if bush allows us an election in 2008.

God knows I am not a Hillary fan, but I can say this in her defense as well as any Democrat running for the office of President of the United States.

There is not one republican, including this idiot Ron Paul, who just thinks he is fooling the hell out of American people, that has a snow ball chance in hell, to even come close to winning this election.

This media will stop at nothing to get some traction for these ass holes, whom have simply destroyed what many of us always felt was the greatest country in the world and if any of these flea bags think that they will come close enough to even smell the white house to finish the bull shit that Bush will leave behind, they are going to simply start a mutiny in this country. This is real

Even they could not steal enough votes to do that. Folks are angry as hell at all of these bastards, but there is complete revulsion for ALL the republicans.

Just looking at these bastards, you can feel the vile build up in your system and the urge to somehow violate what ever laws they left in this country.

No I am afraid that Hillary wins this one.

iraqconcilable @ 10:

Anyone who believes polls aren't skewed to further a cause or manipulate the public is an idiot . This includes Gallup , Zogby or the Pepsi Challenge .

everything in America is skewed, that is what free speech gets you.
a lot of deception and a bunch of fucking bastards in the whitehouse.......

Responsible Journalism... hmmm .... Don't think this is the week for that

We're living in very strange times. Most of the journalism hacks and lackeys remind me of John Hurt as Winston Smith in 1984. Sitting in his cubicle tweaking and editing incoming news stories to fit Big Brothers program before sending them back out for public consumption.

The media, and I am not defending it, latches on to the Zogby poll because it is more newsworthy than the Gallup. The News is no longer about what is actually News; it is purely for entertainment purposes. CNN, MSNBC, CBS, and Faux it us with a headlines like "Hillary Behind in Latest Polls," because it will render more viewers than "Hillary Still Ahead in Polls," it isn't new News.

I hate to blame the internet, but I am going to blame the internet for the media as we now know it. Anybody and everybody can be connected to events at all times now; many in real time. On top of that we can read blogs, such as this one, who feed our political leanings. I read Crooks and Liars because I am a die-hard liberal and they report the stories that I keep tabs on; it makes perfect sense but perspective has been lost.

No longer do we need to look at issues for ourselves; we can feed our own needs by reading online publications that feed our political needs. I won't stop at politics. Music, Art, Film and Sexual Desires can now be fed through specialized sites. What ever happened to discovery? The world is at our fingertips but many of us choose to put our blinders on. When did we lose our way?

L.A. Confidential @ 16:

We're living in very strange times. Most of the journalism hacks and lackeys remind me of John Hurt as Winston Smith in 1984. Sitting in his cubicle tweaking and editing incoming news stories to fit Big Brothers program before sending them back out for public consumption.

did you see or hear what turdblossum said today? this bastard is trying to re-write the
last six years and blame the war on congress.......it's a real shame that some of these bastards can't disappear in the night to bush's vacation home at gitmo.

Yes of COURSE real media outlets should report on real polls, not voodoo ones. However, since we don't actually GET real news in America anymore, it's no surprise that the major INFOTAINMENT outlets would dwell on a sensational and unexpected poll result.

However, DON'T KNOCK IT! For the last 15 years the Republicans have made a specialty out of creating their own 'faith-based' political world, and then selling it to the public with talking points, which proved to be easier than dealing with a 'reality-based' world. Like the price of WORLDCOM STOCKS, this only becomes a problem when you run head-first into a reality that can't be ignored. In the case of WORLDCOM, the $200 stock finally banged into its reality-based price of $0.13 when somebody actually looked in the books.

By the same token, the Republican rhetoric-based view of the world, rating Republican shares at $200 based on bumper-sticker slogans and facile 'family values' ran head-on into the reality of the Iraq war, and plummeted to its true value of $0.13, once their hypocracy and incompetence was exposed. However, if Republicans think it's STILL a good idea to live in a faith-based view of the world, and believe their own fake polls that say Hilary Clinton would lose to Mike Huckabee, and think that all they have to do is sell it to the public, and it doesn't actually MATTER if it's true or not...who are we to argue!?

dadams @ 18:

L.A. Confidential @ 16:

We're living in very strange times. Most of the journalism hacks and lackeys remind me of John Hurt as Winston Smith in 1984. Sitting in his cubicle tweaking and editing incoming news stories to fit Big Brothers program before sending them back out for public consumption.

did you see or hear what turdblossum said today? this bastard is trying to re-write the
last six years and blame the war on congress.......it's a real shame that some of these bastards can't disappear in the night to bush's vacation home at gitmo.

Yeah. That Congress pressured Bush into war. I almost choked when I saw that one.

Hillary up or down in the polls doesn't matter in a rigged system. The candidates that can make a real difference in this country, e.g., Ron Paul, Kucinich, Gravel, have been ignored or played down by the media and leaders of their own political parties. As a result, we pretty much have a system in which the "acceptable candidates" are forced upon the public because the public accepts the false premise through herd mentality that no one else can win. The Democratic and Republican leadership control which candidate wins by not giving other certain candidates the same public attention and forcing the primaries to be decided by a few small early states which allows them to control the process. In addition, those parties do all they can to keep Independents out of the political system. The polls are another tool to make the process seem democratic by reinforcing our thinking that there is really a free and fair electoral process.

Wouldn’t responsible journalism require news outlets to a) note why professional pollsters discount Zogby Interactive data; and b) also highlight the Gallup numbers with equal enthusiasm?

HAHAHAHAHA!
BWAHAHAHAHAHA!
LOLx1000!

Ahem...

Woo...!

Wait. This isn't funny. This is our country we are talking about. Sorry about that, Steve.

"Journalists". Pfffft. Show me an honest journalist and I'll most likely be able to point to the same person and show you an actor on break from his role as the very fictional "last honest reporter" in the made for tv movie of the week.

zogby polls are full of shit.
the questions are literally steered toward some predetermined conclusion.
why are the questions never revealed and have to be hunted down?
fuKKK zogby.
the last zogby poll I read about had americans in favor of attacking Iran.

no credibility whatsoever!

The same polling methodology that predicted President Alf Landon.

Here's how they think...hmmmmmm...which one will sell more tickets....this one will get more attention...so they go with that one,right,wrong,truth or morals do not apply. "It's all about the money", a very slimy ex friend of mine said.

It would be nice to have responsible journalism.

And while we're at it we should cure cancer and end world hunger, too. I'd also like a unicorn.

Bill @ 24:

The same polling methodology that predicted President Alf Landon.

I loved that show!

justabill @ 2:

Here's another tale of two polls, where the difference is more striking.

A new CNN poll of Florida gives a radically different result from another poll released earlier today, which showed Mike Huckabee surging into second place. The CNN numbers: Giuliani 38%, Romney 17%, McCain 11%, Thompson 11%, and Huckabee 9%. The margin of error is ±5.5%.

The other poll, from GOP firm Insider Advantage, had Rudy at 26% and Huckabee with 17%. It had a ±3.8% margin of error.

Obviously, at least one of these polls is wrong.
http://tpmelectioncentral.com/2007/11/another_poll_huckabee_not_second_i...

Obviously? No, not to a group of people who believed that massive tax cuts combined with massive increases in spending with money borrowed from the Chinese would lead to a more prosperous economy, and a LOWER national debt. I think people with minds like that would probably say 'if Guliani can have both 28% and 38% and still be in FIRST place, then there's no problem believing that Huckabee can have both 9% and 17% and be in SECOND place.' QED.

Chris Matthews made these questionable poll results his centerpiece Big Number of the Day. "FIVE GOP candidates are leading Hillary in a national mach-up... EVEN HUCKABY, by a small margin..." WHAT A BUNCH OF BUNK!

When I saw that this evening, I thought to myself... "That is f...ed up! Can't be right." Guess I was right and Tweety was wrong... AGAIN."

Chris Matthews shows his obvious Republican bias over and over again, like his question for Buchanon tonight: (to paraphrase:) "If you were on a mid-evil RACK, writhing in pain, with your arms being pulled off... being forced to vote for Obama or Hillary...(laugh, laugh, laugh, big joke...) which one would it be?" I was insulted. What a bullshit way to carry on a discussion about politicians and issues ON NATIONAL TV. Here's my response:

If someone held a gun to my head, telling me I had to vote for Guiliana or Romney... they were my only two choices and I have to pick one or the other... Which will it be?

I would say: "PULL THE TRIGGER."

Joshua James @ 17:
"newsworthy" as a term does not apply here. "sensational" is more appropriate the attack on hillary, while purporting her to be the frontrunner, is the real story. this is just more from the noise machine that masquerades s press.

murdock and the msm want hillary to be the frontrunner, or appear to be, so they can attack her and thereby the democrats in general. this whole schmear is to pretend that americans will choose a republican over a democrat.

all general polls for party vs. party have the dems coming out ahead, so there is a need to granularize it and try to predetermine the election even before the primaries have taken place. they are attempting to create the impression that americans would prefer one of the racist, fascist white men in repug garb, rather than a woman, a minority or some peacenik pansy pinko in the democratic party.

What's so wrong with the "methodology"?

Ok, they didn't call people on their land line phones or stop them in the mall. They e-mailed their respondents from a random list of demographics and had them take an online poll. (Uh, oh...Thems usin them new fangled compooters.) Oooo. "questionable" It wasn't self initiated...like the Ron Paul polls.

In the study you cite (Actually I had to go through three sites just to get to it...here...from those "responsible journalists" over at the WSJ, no less.) Ok...Zogby's Interactive missed by 8.3% while the other polling firms missed by 3.4, 4.8 and 4.4. in one gubernatorial race. They all picked the winner but missed the numbers.

Look, polls are polls...they are transient, changable and not scientific. Paleeze. If you're going to slam one...slam em all...not just the ones that piss you off.

The MSM is a disgrace.

Wouldn’t responsible journalism require news outlets to a) note why professional pollsters discount Zogby Interactive data; and b) also highlight the Gallup numbers with equal enthusiasm?

You want responsible journalists?

Yeah...

So do i...

it's not about news... and it hasn't been for a long time...

it's about drama...

i mean, they are competing against soap operas...

Let's not get too carried away with this. Hillary actually ties with Giuliani and McCain.

Clinton -- the dominant front-runner for the Democratic presidential nomination -- would appear to have at least a slight advantage over any Republican candidate among registered voters if the election were held today. She has a five-point edge over Giuliani (49% to 44%) and a six-point edge over McCain (50% to 44%), but neither lead is statistically significant. Clinton runs much more strongly against the lesser-known Thompson (53% to 40%) and Romney (54% to 38%).

Part of their statement by the way, is a load of horsehockey. If the difference is within the margin of error ("not statistically significant), it is not a difference. If they took the poll over agin, the results could be reversed. The entire poll is baloney since they are asking only 800 people to represent the entire USA. Even a candidate who actually had a statistically significant ead would only be able to claim that they could win the popular vote -- and we all know how much that means.

The whole poll thing is a charade. Hillary's hope was that strong poll numbers would "suck the air" out of the race for all of the other candidates, and thus the otherwise meaningless polls would become a self-fufilling prophecy. Obama's impressive fundraising killed that idea, so now she is kind of up a creek.

Really... since when is Zogby so disreputable? It's no winger poll outfit, that's for sure. So, is it just disreputable because it's harshing on Hillary's buzz?

A quick google search hasn't enlightened me to why you think Zogby is unreliable? Maybe you can offer some evidence...?

As Zogby noted in a press release, its online polls identified 18 of 19 Senate winners correctly. But its predictions missed by an average of 8.6 percentage points in those polls -- at least twice the average miss of four other polling operations I examined. Zogby predicted a nine-point win for Democrat Herb Kohl in Wisconsin; he won by 37 points.

http://www.pollster.com/blogs/numbers_guy_rating_the_pollste.php

I hardly find that comforting. So the odds are 94% that Giuliani beats Hillary but it is by 12 points instead of 4 points? That seems to be what they are implying about Zogby's record.

What if it is the Gallup methodology that is flawed because people won't admit to another person live on the phone that they won't vote for a chick?

rekroc @ 31:

The MSM is a disgrace.

you obviously are a fakenewscrap (to help you that's fnc) fan

"Wouldn’t responsible journalism require... "

this is American press were talking about ...

I'm sorry. I know it suits the narrative of many of the people here, but 90% of the polls have Hillary beating every Republican since the summer. Obama, too, to some extent, but Hillary across the board. And that's, you know, real polls. Has been, still is. Zogby "Interactive" is about as good as the mail-in to Literary Digest, a conservative magazine, that showed Alf Landon in a landslide.

I gotta quit deleting those e-mails when I get them from Zogby Interactive asking me to participate in their survey.

Look. Most political polling hardly rises to the level of pseudo-science. I have done some political calling, and over %80 of the people I called were not home. Even smaller numbers wanted to hear what I had to say. The number of people who have landlines, happen to be home, take calls from strangers, and are willing to waste 15 minutes talking to a pollster has to be miniscule. The people who do take such calls are unlikelyl to represent the general population. The reported margin of error is always a joke -- it is just a crude mathematical approximation that makes so many assumptions of perfect method that it has no relationship to the real world.

Whenever you see a political poll, the first thing that you should do is look for the sample size. (In the gallup case, it is 800). Then take any of the "points" and multiply that by the sample size (which is often exaggerated in ways beyond the scope of my discussion.) Now rewrite the poll in actual people units rather than percentages.

{Consider: it looks really impressive if I tell you that my recent polling shows Obama leading Hillary by 75 points. It is less impressive if I tell you that I talked to four people.}

So the Gallup poll was 897 registered voters.

Hillary "leads" Giulani 49 to 44. What that really means is that 439 people said "Clinton" and 394 people said Giuliani. Her "lead" is a WHOPPING 45 PEOPLE! Out of almost 900. Representing the entire country.

Does that make my point clear? It is the same point if my guy (Obama) is ahead. The polls are crap.

I've never heard anything bad about Zogby up until they reported this poll result. I wonder if it being challenged by those not happy with the results. I'd rather not consider the polls at all. let's look at more relevant matters such as fundraising $, number of individual donors, campaign event turn out and other more measurable things. If those things were the guage Obama would be winning hands down.

Some info from Zogby:

Why don't you ever call me or my friends? None of my friends, or those in my circle agree with the polls!
JZ: "Phone numbers are chosen purely by random, ensuring that every household in the US (or wherever we are polling) have as much chance of being selected as any other. With tens of millions of adults in the US, it is still rather easy to be missed - but stay in there and maybe stay home more often. You probably have more of a chance of being called than having a visit by Ed McMahon."

How can polls be so accurate when you only ask such a small number of people?

JZ: "It's pure probability and statistics. The same theory is involved as when you take a blood test and the clinician draws only a small sample rather than draining all the blood out of your body."

What is a "margin of error"?

JZ: "The margin of sampling error means that if we do the same poll 100 times, in 95 cases out of 100, we will get the same results plus or minus a certain percentage. Generally, a sample of 400 gives you an MOE of +/-5%, 600 gets you +/-4%, 1000 gets +/-3%. We need to be sure our sampling is random and scientific. It all relates to probability and statistics. If there are a million marbles in a jar, some black and some white - how best do I determine the number of each color, short of counting every marble? If I draw 1000 out at random, chances are I will get the same numbers of each 95 times out of 100 within a margin of error of +/-3%. However, there are other sources of survey error - e.g. how questions are phrased, etc."

I've heard so many conflicting stories - is Zogby International a Republican pollster, a Democratic pollster or an Independent pollster?

JZ: "We are independent and nonpartisan. I am personally a Democrat, but the firm does a lot of work for media (like Reuters America, New York Post, St. Louis Post Dispatch, etc.) and we work for both parties."

Where do you poll for Zogby America? What regions, area codes or time zones?

JZ: "If we are polling the US, we poll from a sample drawn from all households with telephones in 48 states. We, like others, do not poll Hawaii or Alaska because of time differences and because Republican Alaska cancels out Democratic Hawaii. As well, out of a sample of 1000 likely voters there would only be a total of 1 from both states (combined)."

Can I be put on a list so that you can call me for a future poll?

JZ: "I'm sorry but our polls must be random. As soon as we "place" people in our pool of respondents we run the risk of skewing results."

I read a lot of polls and yours is so different from the others - what makes your answers so different (and accurate)?

JZ: "We poll only likely voters who are different from just all adults. In addition, we poll all day long - 9am to 9pm local time (to the region we're calling). Finally, we apply weighting for party identification to ensure that there is no built-in Democratic bias in our sampling."

Zogby's pools are Pravdaist in nature

dadams @ 38:

rekroc @ 31:

The MSM is a disgrace.

you obviously are a fakenewscrap (to help you that's fnc) fan

Maybe you should refrain from jumping to conclusions, because it can often make you look stupid.

Phone numbers are chosen purely by random

We poll only likely voters who are different from just all adults.

BWAHHAHHAHHA! Crap, I tell you, pure crap. As soon as you start trying to figure out who is a likely voter, the jig is completely up. You may as well just make up the numbers. I also loved the little Hawaii/Alaska dance.

Jim H @ 40:

I'm sorry. I know it suits the narrative of many of the people here, but 90% of the polls have Hillary beating every Republican since the summer. Obama, too, to some extent, but Hillary across the board. And that's, you know, real polls. Has been, still is. Zogby "Interactive" is about as good as the mail-in to Literary Digest, a conservative magazine, that showed Alf Landon in a landslide.

This is a fine example right here. "%90 of the polls". You really looked at every crappy little poll and calculated that %90 of them show Hillary beating all of the Republicans? When even the Gallup poll that she trots out shows at best a tie? "Obama to some extent." Care to quantify that? "Hillary across the board" of %90 of the polls? WTF?

Sorry, feeling ornery tonight. I used to work with real scientific surveys and this stuff gets my goat -- long before this election.

Chris Matthews did an entire segment on "Clinton's demise" as though he can't judge the accuracy of various polling techniques. But a month ago, he shot down a guest for citing Rasmussen:

MATTHEWS: You really see him in double digits in a national poll?

Which poll is that?

BACON: I think it was a Rasmussen survey.

MATTHEWS: Rasmussen? Let‘s MoveOn.com here. We‘re not doing Rasmussen polls here. Come on. The polling average in Iowa shows Huckabee trending up. What do you think?

Planet B @ 36:

Really... since when is Zogby so disreputable? It's no winger poll outfit, that's for sure. So, is it just disreputable because it's harshing on Hillary's buzz?

A quick google search hasn't enlightened me to why you think Zogby is unreliable? Maybe you can offer some evidence...?

maybe you should read zogby's questions before typing that. and yeah, they are reliable if you ask the question on a tuesday in front of a fish market with four clams singing the somali national anthem a capella while sitting on top of an iranian IED.

I saw Chris Matthews on MSNBC and Allen Colmes on Faux News bring up this Zogby poll and neither one of them mentioned that it was an INTERACTIVE, ONLINE poll where people who wanted to express their opinion signed up to be contacted for polling. Why didn't they mention it? That is a very different animal from the kind of polling that is usually cited. And from the kind of polling that far more accurately reflects actual election results.

By the way, despite what some here may wish were the case, RANDOM polling methodology of the kind that is used by most major polling services is remarkably accurate. It has been for decades.

The polling results change over time because opinions change over time. But actual national election results mirror the major polling results leading up to election day so often and so consistantly that a discrepancy beyond a normal margin of error in more than a scant few polls (which has not happened in a very, very long time) would be a HUGE upset and headline news for several days after an election.

See, if Hillary Clinton was a lock (which poll after poll says she is...although she still has to be officially nominated and elected), it wouldnt be good for news reporters (i guess). Its like some ridiculous reality TV show.

Is it possible the media will let us vote? This is insanity... polls and the press pick the candidates before the votes are even cast.

These are "two polls" are reflective of our two party (Dem/Repub) presidential race. The poll numbers all depends on which party paid for the polls. Both will show the best poll showing for a price. They (polls and party) share several things in common .

1. They are paid for profit poll corporation.
2. They are supported by political party.
3. Too much hype are put in that; they seem to think they already know who will win the election.

Only rely on your objective observation when making a vote. Don't let the media or polls tell you who will be the winner. They are out there to manipulate your natural way of thinking.

Johnny2Bad @ 31:

What's so wrong with the "methodology"?

Ok, they didn't call people on their land line phones or stop them in the mall. They e-mailed their respondents from a random list of demographics and had them take an online poll. (Uh, oh...Thems usin them new fangled compooters.) Oooo. "questionable" It wasn't self initiated...like the Ron Paul polls.

In the study you cite (Actually I had to go through three sites just to get to it...here...from those "responsible journalists" over at the WSJ, no less.) Ok...Zogby's Interactive missed by 8.3% while the other polling firms missed by 3.4, 4.8 and 4.4. in one gubernatorial race. They all picked the winner but missed the numbers.

Look, polls are polls...they are transient, changable and not scientific. Paleeze. If you're going to slam one...slam em all...not just the ones that piss you off.

I was wondering when someone was going to even mention the methodology. It took 31 posts? Come on C&L, be more specific in your attacks.

CoIntelPro @ 50:

Planet B @ 36:

Really... since when is Zogby so disreputable? It's no winger poll outfit, that's for sure. So, is it just disreputable because it's harshing on Hillary's buzz?

A quick google search hasn't enlightened me to why you think Zogby is unreliable? Maybe you can offer some evidence...?

maybe you should read zogby's questions before typing that. and yeah, they are reliable if you ask the question on a tuesday in front of a fish market with four clams singing the somali national anthem a capella while sitting on top of an iranian IED.

Nice snarky comment. Still no answer.

1) you provide no proof for your assertion that Zogby is bad, (if you go 2 links deep into your post you get some guy in the WSJ claiming that Zogby is right but not that right most of the time . . . whatever that means.)

2) Here's what the Zogby folks had to say:

no other campaign has made as many requests for Zogby polling data over the years than Penn has made on behalf of Clinton.

I tend to think that there is no way Hillary would be polling that badly against the 5 worthless republican losers, but put the flame thrower away and bust out some, you know, facts . . .

I find it interesting that people question the methodology of Zogby's Interactive poll without bothering to read the details on Zogby's website. Yes, people do indicate their interest in signing up for the poll, but a small fraction of those who indicate such an interest are actually contacted for the poll. Every person who signs up is contacted over the phone and is registered on the panel only after being authenticated. All standard criteria for ensuring representativeness are followed. Yes, the sample tends to skew towards the more partisan, but that should be reflected in both the Democratic- and Republican-leaning groups. It is also likely more representative of people who actually vote (the uninvolved rarely vote). No one here seems to mention that the same methodology that is being questioned also seems to show that Obama and Edwards are beating all the Republican candidates. If the sample were skewed Republican (resulting in people like Huckabee getting ahead of Clinton), wouldn;t this also show up for Obama and Edwards? Oh, and by the way, the Interactive survey has a sample size of 9150 voters, compared to 1000 or so for most telephone surveys. Yet, Steve Benen is willing to place greater faith in the Gallup poll which has 1/10 the sample size of Zogby. OK, so let's assume that Zogby is not as "reliable" as Gallup. How much would a larger sample (Zogby's Margin of error is +/- 1% compared to Gallup's +/- 4%) compensate for errors due to skewed samples? We don't know, but I don;t think the Zogby sample skews so much in one direction as to bias the results to the extent of 3-4%.
I do not hold any brief for Zogby nor am I employed by them, but it is amusing that people just jump at the slightest chance to question polls when it does not suit their interest and cite the very same polls when the numbers are rosier for them.

I just wanted to add to the discussion that I am on the Zogby Interactive Survey list, and got the poll on 11/23/07.
It is less open than other polls because people have to have already self chosen to get polls periodically from Zogby.
I do it because I want my liberal, atheist, -8.00, -6.31 on the political compass self reflected on the occasional poll.
I cannot go back to the survey, because it blanks-out after being submitted. But I wanted to share the basics of it.

The poll presented a match-up of Hillary, Obama and Edwards with I think the top 5 Repub candidates.

I am one of the people who didn't help out Hillary. For every match-up between Obama or Edwards and a Repub, I picked Obama or Edwards. For every match-up between Hillary and a Repub, I chose "Other".

I hope the idea catches on though. Because I ain't voting for Hillary in the primary or the general.

Suresh @ 58:

I find it interesting that people question the methodology of Zogby's Interactive poll without bothering to read the details on Zogby's website. Yes, people do indicate their interest in signing up for the poll, but a small fraction of those who indicate such an interest are actually contacted for the poll. Every person who signs up is contacted over the phone and is registered on the panel only after being authenticated. All standard criteria for ensuring representativeness are followed. Yes, the sample tends to skew towards the more partisan, but that should be reflected in both the Democratic- and Republican-leaning groups. It is also likely more representative of people who actually vote (the uninvolved rarely vote). No one here seems to mention that the same methodology that is being questioned also seems to show that Obama and Edwards are beating all the Republican candidates. If the sample were skewed Republican (resulting in people like Huckabee getting ahead of Clinton), wouldn;t this also show up for Obama and Edwards? Oh, and by the way, the Interactive survey has a sample size of 9150 voters, compared to 1000 or so for most telephone surveys. Yet, Steve Benen is willing to place greater faith in the Gallup poll which has 1/10 the sample size of Zogby. OK, so let's assume that Zogby is not as "reliable" as Gallup. How much would a larger sample (Zogby's Margin of error is +/- 1% compared to Gallup's +/- 4%) compensate for errors due to skewed samples? We don't know, but I don;t think the Zogby sample skews so much in one direction as to bias the results to the extent of 3-4%.
I do not hold any brief for Zogby nor am I employed by them, but it is amusing that people just jump at the slightest chance to question polls when it does not suit their interest and cite the very same polls when the numbers are rosier for them.

The problem with interactive polls is self-selection. Only those who have said that they want to be polled can be polled. It's much like the self-selection you see on political blogs. Notice that on very few blogs is Senator Clinton the favorite candidate of the posters. Yet the traditional polls show her as the leader in most states, some by huge margins. The likely reason is the political blogs only attract the most rabid political junkies. Among that population, Senator Clinton is clearly not favored. An interactive poll like Zogby's also attracts the same most rabid political junkies to sign up to be polled to a greater degree than traditional (more random) selection methods because those political junkies have a high desire to be polled and will sign up with Zogby. But that overabundance of political junkies in the Zogby pool versus a random pool of registered voters makes the Zogby results match what you see on the political blogs more closely than a random sample of registered voters would. Zogby recognizes that he has a self-selection problem with his methodology and he tries to mitigate it by doing larger samples but it is far from clear whether larger sample sizes really helps with the self-selection problem. If everyone who registers to vote were made available to be polled interactively, then interactive polls would likely fare much better.

One should take all polls with a grain of salt because it is easy for errors to creep into them, but interactive polls have a built-in self-selection error so even if the rest of their methodology is perfect, the results are still dubious, in my mind.

'Responsible Journalism?'

Where do you think you are, Doha, Qatar?

'Responsible Journalism' is much too high a bar for the U.S. MSM.

'Babbling Gossipers' fits. So does 'Corrupt Corporate Connivers'.

The MSM is NOT interested in news or informing the public. They're all about entertaining and profit. And frankly, the broadcast news media is hurting the country. Americans could do their country a ginormous favor by tuning them out. They corrupt the whole election process, and ultimately have sway over the elections. Is it any wonder we end up with an idiot like Bush? It's all misinformation and propaganda all the time. They choose and spin every story for maximum hype and mileage. It's gotten ridiculous. I refuse to watch it anymore. You're better off sticking to written material or even the evening news, which although watered down, pretty much sticks with a brief synopsis of the facts. Unfortunately, a majority of Americans aren't smart enough to sort out fact from fiction, and are easily manipulated by propaganda. Who can ever tell what the truth is anymore? It's really a form of brainwashing, which is kind of frightening. The right wingers figured that out a long time ago. If you control the media, you control the message, and therefore you control everything. People are easy to fool.

Cowboy Bob in Austin @ 29:

Chris Matthews made these questionable poll results his centerpiece Big Number of the Day. "FIVE GOP candidates are leading Hillary in a national mach-up... EVEN HUCKABY, by a small margin..." WHAT A BUNCH OF BUNK!

When I saw that this evening, I thought to myself... "That is f...ed up! Can't be right." Guess I was right and Tweety was wrong... AGAIN."

Chris Matthews shows his obvious Republican bias over and over again, like his question for Buchanon tonight: (to paraphrase:) "If you were on a mid-evil RACK, writhing in pain, with your arms being pulled off... being forced to vote for Obama or Hillary...(laugh, laugh, laugh, big joke...) which one would it be?" I was insulted. What a bullshit way to carry on a discussion about politicians and issues ON NATIONAL TV. Here's my response:

If someone held a gun to my head, telling me I had to vote for Guiliana or Romney... they were my only two choices and I have to pick one or the other... Which will it be?

I would say: "PULL THE TRIGGER."

What a question, huh? The stuff Tweety comes up with that is so crazed & foaming-at-the-mouth misogynistic is unbelievable! If Tweety & his lead-in Tucker didn't have "reliable data" like Zogby, what would they use for their hour-long nightly HRC bashing session? Tonite, Tweety's HRC bashing session is being promoted with a clip from Jimi Hendrix (?!!!) "Are you experienced?" which of course will bash her for having any experience outside the bedroom (which is really the only experience he's interested in "examining"). Morning Joe isn't any better especially since they (Mika in particular) are actively campaigning for Obama & to a lesser extent, Edwards. Without KO, MSNBC isn't much more "fair & balanced" than Fox Noise.

rob (formerly) in toronto @ 15:

Responsible Journalism... hmmm .... Don't think this is the week for that

Don't think this is the CENTURY for that.

Hey! As a "responsible journalist" I'm a little offended by the notion that we are not taking our jobs seriously... we're not irresponsible... just lazy. It's better to just repeat whatever the talking points are than to actually do work and research.

It seems nowadays that most journalists just don't let facts get in the way of a good story.

(For the record... I cover sports for a small-town newspaper, but I still see it every once and while at that level as well.)

Strange, as a professed liberal/progressive, the Zogby poll wasn't offered to me though I am regularly polled and a member of their data base. My partner, identified as moderate, got his usual poll invitation.

I seem to recall that Zogby polls showed Kerry in a strong lead in 2004 when other polls didn't, and therefore Zogby was put forth as the most reliable. And now it's not? I'm confuzzled.

The Media is probably instructed to make the races look like they are closer than they are or tied, so that the election system administrators can throw the election one way or the other. Are these elections all really 51% or 53%??

Tshober @ 60:

The problem with interactive polls is self-selection. Only those who have said that they want to be polled can be polled. It's much like the self-selection you see on political blogs. Notice that on very few blogs is Senator Clinton the favorite candidate of the posters. Yet the traditional polls show her as the leader in most states, some by huge margins. The likely reason is the political blogs only attract the most rabid political junkies. Among that population, Senator Clinton is clearly not favored. An interactive poll like Zogby's also attracts the same most rabid political junkies to sign up to be polled to a greater degree than traditional (more random) selection methods because those political junkies have a high desire to be polled and will sign up with Zogby. But that overabundance of political junkies in the Zogby pool versus a random pool of registered voters makes the Zogby results match what you see on the political blogs more closely than a random sample of registered voters would. Zogby recognizes that he has a self-selection problem with his methodology and he tries to mitigate it by doing larger samples but it is far from clear whether larger sample sizes really helps with the self-selection problem. If everyone who registers to vote were made available to be polled interactively, then interactive polls would likely fare much better.

One should take all polls with a grain of salt because it is easy for errors to creep into them, but interactive polls have a built-in self-selection error so even if the rest of their methodology is perfect, the results are still dubious, in my mind.

Granted that the Zogby Interactive attracts more rabid political "junkies". The bias is however in both directions. So, what this establishes is that Hillary's support is weaker among people who are strongly left or right of center. We know that to be a fact, so the poll does accurately represent this point. It is not inaccurate in that sense. The second point is whether those that are moderate or not as motivated as these "rabid junkies" will come out and vote. I guess the primaries may see more of them compared to the general election. To that extent, I suspect that the Interactive poll may be a little more valid for comparisons within party (that is, for the primaries) than for the general election.

Your point about self-selection is well-taken but do pause to think about whether the telephone survey draws a truly representative sample. People who answer such surveys are also more likely to be those with strong political views. To that extent, a telephone survey also does not draw a truly random sample.

Polls...Schmolls!
I could give 2 shits about polls.

liberalNmoderation @ 70:

Polls...Schmolls!
I could give 2 shits about polls.

That's why they don't put outhouses near polls, to save on the confusion.

You can put a lot of the blame on HuffPo as well.

Check Ariannas blog every day, and you'll see this same type of anti-HRC spin puked all over her front page. It's gotten to the Drudge levels over there, and I don't see why anyone gives her credibility any longer.

Ordinarily shady polling would bother the hell out of me. But honestly, at this point anything that weakens Hillary Clinton is good. The White House is not a frisbee for the Bush and Clinton families to toss back and forth. It is so very far from the way America is supposed to function. Anything, and you bet your ass I mean ANYTHING, is better than continuing that particular farce. New World Order indeed. Throw the bums out.

sadly in many people minds. polls like that add to the credibility of what is a poor lineup in the GOP.
people think..."maybe there is more to the GOP lineup then I thought"....

brainwashing 101

Personally, I hope the poll is right and people get scared in the Dem primary and vote for anyone but Hillary.

Another point, the Gallup pole was taken some 2-3 weeks prior to the Zogby poll. A lot has happened in the interim with the polling in Iowa for example and Hillary having a few slip ups. That can certainly account for some difference. To really use the Gallup poll as a point of comparison for criticizing the Zogby poll, it would have to have been taken over the same time frame.

msm still trying to bring down clinton. For the most part obama gets a free ride. But its not just the media that is biased.
many so called liberal or progressives what ever they want to call themselves bash clinton constantly. Clinton is getting alot of flack for CONSIDERING USING POWELL. OBAMA HAS ALREADY USED HIM. BUT LITTLE WAS SAID IN THE BLOGS ESPECIALLY DAILY KOS.

WHO CARES ALL THESE POLLS ARE WRONG THE REAL WINNER IN THIS PRESIDENTIAL RACE WILL BE THE AMERICAN CITIZENS WHEN AMERICA GETS OFF ITS LAZY ARSE, DOES A LITTLE RESEARCH AND VOTES RON PAUL TO BE OUR NEW PRESIDENT

78 comments

Login or Register to post comments.