Posted on 12/22/2007 3:52:38 PM PST by Yaelle
Survey Methodology [Zogby America Likely Voters] 12/12/07 thru 12/14/07
This is a telephone survey of [likely voters] conducted by Zogby International.
The target sample is [1000 likely voter] interviews with approximately [96] questions asked. Samples are randomly drawn from telephone cd’s of national listed sample. Zogby International surveys employ sampling strategies in which selection probabilities are proportional to population size within area codes and exchanges. Up to six calls are made to reach a sampled phone number. Cooperation rates are calculated using one of AAPOR’s approved methodologies[1] and are comparable to other professional public-opinion surveys conducted using similar sampling strategies.[2] Weighting by [region, party, age, race, religion, gender] is used to adjust for non-response. The margin of error is +/- 3.2 percentage points. Margins of error are higher in sub-groups.
The target sample is [436 Democratic Likely Primary Voter] interviews with approximately [96] questions asked. Samples are randomly drawn from telephone cd’s of national listed sample. Zogby International surveys employ sampling strategies in which selection probabilities are proportional to population size within area codes and exchanges. Up to six calls are made to reach a sampled phone number. Cooperation rates are calculated using one of AAPOR’s approved methodologies and are comparable to other professional public-opinion surveys conducted using similar sampling strategies. Weighting by [party, age, race, religion, gender] is used to adjust for non-response. The margin of error is +/- 4.8 percentage points. Margins of error are higher in sub-groups.
The target sample is [432Republican Likely Primary Voter] interviews with approximately [96] questions asked. Samples are randomly drawn from telephone cd’s of national listed sample. Zogby International surveys employ sampling strategies in which selection probabilities are proportional to population size within area codes and exchanges. Up to six calls are made to reach a sampled phone number. Cooperation rates are calculated using one of AAPOR’s approved methodologies and are comparable to other professional public-opinion surveys conducted using similar sampling strategies. Weighting by [region, party, age, race, religion, gender] is used to adjust for non-response. The margin of error is +/- 4.8 percentage points. Margins of error are higher in sub-groups.
[1] See COOP4 (p.38) in Standard Definitions: Final Dispositions of Case Codes and Outcome Rates of Surveys. The American Association for Public Opinion Research, (2000).
[2] Cooperation Tracking Study: April 2003 Update, Jane M. Sheppard and Shelly Haas. The Council for Marketing & Opinion Research (CMOR). Cincinnati, Ohio (2003).
(12/19/2007)
Is there some kind of automatic screening going on, simply by who chooses to answer the phone and continues with a very long polling session? And how does that correspond to Actual American Voters, some of whom are busy and don't answer polls?
Also, the sampling strategy for lower population sizes: does this really work? Does it end up meaning that if only two Asian men were sampled, it would be extrapolated to signify some absurd percentage of Asian men prefer one candidate to the other?
I know there is some science behind statistics and surveys, and pollsters like to show how close they were AFTER an election, but I keep coming back to little things like Caller ID. Is anyone explaining that the samples are heavily weighed toward either people who do not yet have caller ID or toward people who love speaking with pollsters?
Ping to explain polling methodology!
Ping to help explain why we should believe polls, mathematically, please?
Are you saying that polls are flawed from the start due to the “push-polling” that goes on with the questions pushing the poll “victim” into one direction?
We’ve all had polls like that, in which the “neutral party” on the line is asking questions like “If you knew that Joe Opponent burnt kittens in his fireplace, would you still be likely to vote for him?”
However, I am assuming that Zogby’s questions are more neutral and not “push polling.” Am I right?
Everybody has an opinion, and lots vote in actual elections, but few vote in primaries.
Zogby targets actual past primary voters.
If we really “live and die” by the poll numbers around here, I’d like to know the basis of the science of calling and bugging people, and how it relates to real voters actually voting!
So the ones he includes are very likely voters on the basis that they DID vote in the last primary. OK, good. But how many of them decide not to go ahead with the entire poll because they are busy? Is his poll top heavy with people who have a lot of time on their hands?
I think they gather the samples and then they get demographics and apply a statistic test which include error
error = sqrt ((p X (1-p))/(n-1))
p (candidate A) = 0.48
q (candidate B) = 0.52
n = 502
error = (sqrt(0.52 X 0.48)/(501)) =0.02
Candidate A = 48%
Candidate B = 52%
Margin of error = +/- 2%
If n > 33, then n-1 is not considered statistically significant.
I love your math but I am just not completely understanding it. What is the variable for the “demographics” and how does that work in? How is significance determined?
Please pretend you are speaking to an absolute idiot (me) with blonde hair.
In my view only an idiot would respond to a poll of any kind. I never answer local census forms when they arrive, but the national census is something that I feel bound to respond to even though I resent it. If I ever get one of those comprehensive national census forms I’ll burn it.
The problem for me is that somebody is responding to pollsters somewhere, everyday, and in large numbers. In the case of Zogby polls, they are usually better than most.
I do believe that Zogby is biased and has been doing some tweaking during the last two presidential elections.
They were wrong both of those years.
In the lesser publicized elections, the Zogby polls are quite good, and the hacks use them....a lot.
“Math ExpertsPlease explain”
Zogby, unlike say Rasmussen, just makes it up.
I have heard that his brother is the head of the American Arab Institute. I do not know if that makes Zogby biased. His polls seem to be very trusted on both sides; why, if his family is so un-neutral?
How would bias be hidden by a poll? How to bias your poll and yet after the election be the one who had the closest results? AND WHY ARE POLLS TAKEN SO SERIOUSLY, by basically everyone (FReepers included)?
Is there really only a tiny MOE (margin of error)?
Do we talk poll numbers all day and all night just to have SOMETHING to talk about?
Why should we believe polls?
The demographics would be separated out into sub categories and the same statistically test applied.
If you live in a district where one party is significantly more represented on the voter roles, then the number of that party may be sampled more than other parties on a pro rata bases.
Eg, if I have a district in Utah that is 75% republican, I may want to have a sample size of 75% republicans. You may perform another survey, lets say 55% republicans and then compare the samples.
This of course, is not cheap.
OK, then please explain Rasmussen. Explain the methodology of ANY telephone poll you actually TRUST. And why should we put faith in it?
I mean, if your father is in surgery and the doc comes out to give you an update midway through, of course you hang onto every raised eyebrow or sigh in his delivery. Is that what we are doing when the polls come out?
This makes sense to me only if it's a general election.
So Zogby tweaks and can’t be trusted in big elections, but is thought to be trustable in the smaller races?
I too think that many people would either NOT pick up the phone for a pollster (unknown caller ID) or even refuse to go further if they’d got him on the line. Isn’t that a screening right there? How do we trust polls, when the headline never says “ROMNEY IS WINNING 2 to 1 OF LIKELY VOTERS WHO CHOOSE TO TALK TO POLLSTERS”?
Zogby predicts Kerry electoral landslide! by daninoah Tue Nov 02, 2004 at 02:23:55 PM PST He says that Kerry will will 311-213, but barely lose the popular vote. Kerry wins Ohio, Florida, and most of the other battlegrounds.
Even without talking to people with Caller ID one can impute what answers they would give.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.