Skip to comments.Barone: When it comes to polls, readers beware
Posted on 09/30/2012 6:57:26 AM PDT by SMGFan
As a recovering pollster (I worked for Democratic pollster Peter Hart from 1974 to 1981), let me weigh in on the controversy over whether the polls are accurate. Many conservatives are claiming that multiple polls have overly Democratic samples, and some charge that media pollsters are trying to discourage Republican voters.
First, some points about the limits of polls. Random-sample polling is an imprecise instrument. There's an error margin of 3 or 4 percent and polling theory tells us that one out of 20 polls is wrong, with results outside the margin of error. Sometimes it's easy to spot such an outlier; sometimes not.
In addition, it's getting much harder for pollsters to get people to respond to interviews. The Pew Research Center reports that it's getting only 9 percent of the people it contacts to respond to its questions. That's compared with 36 percent in 1997.
(Excerpt) Read more at washingtonexaminer.com ...
At this point in time if someone is still undecided about voting against Obama, after the 4 years we have been through with this clown and his administration then there is no hope of changing them and we are in real trouble as a nation.
From the full article:
“Part of that is interviewer error: Exit poll pioneer Warren Mitofsky found the biggest discrepancies between exit polls and actual results were in precincts where the interviewers were female graduate students.
But he also found that Democrats were simply more willing to fill out the exit poll. Which raises the question: Are we seeing the same thing in this month’s polls?”
The MSM chickens are coming home to roost. They’ve been BSing us, hating us, mischaracterizing us for decades. So we just don’t bother talking to them. Serves ‘em right.
Part of that is interviewer error: Exit poll pioneer Warren Mitofsky found the biggest discrepancies between exit polls and actual results were in precincts where the interviewers were female graduate students.
polls: jewish voters down to 65% from 78% commie, blacks down to 88% from 96% commie, etc. Anyone registered repub or Indy, why would they now vote communist?
Surfing Youtube for oldies but goodies like The Seekers, Carpenters, Jay and the Americans, Gary Lewis and the Playboys and Pet Clark is a good diversion from ObamaCRAP.
Try this heartening exercise. Go read some articles and opinions on left-wing websites (not the “media”, but the “Nation”-types). There ain’t a whole lot of confidence in the comments. The far Left is sitting this one out at the top of the ticket.
We get an average of three calls a night right now - I don’t know how many we get during the day.
Some are likely pollsters.
We don’t answer any of them.
“why would they now vote communist”
It sounds good until you’re starved to death or shot. It is always sounds good in the beginning...we now have a dumber population as a whole, not because of the union education establishment which is a small part of the education system, but because of popular culture thrust on us by corporate advertising and the media and more importantly the nanny state who is in bed with the MSM. Kids these days don’t come to Kindergarten prepared to learn and then go home in the evening to parents that only see them as a check and who fail to reinforce the lessons they have learned during the day. Its easier to complain that the teachers aren’t doing their job. So why wouldn’t they go for the check? Excuse me communism. Many only attend school so they can get a SSI check for ADHD. If you don’t get the SSI check surely you’ll get into the NBA or the NFL just ask the voters who approve of new stadiums every day from HS to Pro...the Coliseum was built by Roman voters after all. Oh that’s right they gave the government the power.
Interesting that only in Rasmussen this morning, only 42% of voters feel certain they will vote for Obama, while Romney gets 43%. There's room for victory there, even a big victory.
same here in Hudson County, friends father (A Dem) used to answer the calls from pollsters, etc. now he is deceased. But pollster keep calling.
Believe it or not, millions of voters are just now starting to pay attention. The key is for Romney to tell the American public why Obama is a disaster and why his policies will never work. Strong rather than weak. Forget the “he’s (Obama) is a nice guy, but I’m a better technocrat” garbage. Tell them “Obamas’s socialist-inspired plans have failed and they’ll never work!!” Victory through strength.
And if it's 9 %, it's past time to ask, "who is more likely to answer? could there be a political bias in who is answering?"
Dems are huge on new media and old fashioned snail mailings. GOP loves the ophone calls, which means, after 200+ phone calls this month, I immediately hang up if I get a whiff that it may be political.
I have become "un-pollable." I'll bet others have, too.
You are spot on. There is a significant number of persuadable democrat votes out there. If Romney could persuade 1 out of every 20 democrats or even 1 of every 40, he turns the election. I agree with you.
A nine percent response rate is unacceptable for social science survey work. If you don’t have at least a 30% response, you couldn’t get a paper based on the survey published in any reputable journal in the social or behavioral sciences.
The problem is that with a low response rate the likelihood that there is some systematic cause for non-response is much higher than for higher response rates, so the sample, no matter how thoroughly random the selection of people to survey was, may not be a ransom sample of the population, but only a random sample of the subpopulation that lacks the characteristic that systematically causes non-response.
The obvious test for this would be to have the exact percentages of voter registrations by party in a state, do a random survey from the population (registered voters in the state) for which the percentages are known, and do a chi-squared test on the percentages of party IDs among the respondents with the null hypothesis that the percentages are the same as in the voter registrations. If the null hypothesis is rejected, then you’d have (statistical) proof that polling in that state has a systematic cause of non-response that is skewing the party ID in polls.
Catholics up to 51% for GOP this time.. from 47% in 2008 and almost to the 52% Bush got in 2004
Catholics here tout this ...not sure myself but it can’t hurt and seems like a trend overall but will it be enough
IMHO the pollsters are calling the same people over and over again. Some people claim to be getting numerous calls.
I have a home phone and a cell phone. I live in an average size area. I had a steady job for 30 plus years. I am registered and vote in every election regardless of the size.I have a valid DL. I get Jury Notices every three years like clockwork. I have never received a call from any Pollster in my life. I have never heard of anyone who has heard from a pollster. My family is alive and well and has the same home phone for over 55 years. Never rec’d a call from a pollster. Every telemarketer in the known Planet finds the number but NO POLLSTERS. Do the math. They are polling off a list of known people to get the result they want.
Why was that? Was it a bunch of Republican males pretending they voted for Obama in the hopes of getting laid with the female graduate student?
Several reasons. First, female graduate students are OVERWHELMINGLY liberal. This point is not remotely worth arguing. Even the best pollsters have trouble keeping their own feelings and preferences hidden. Female graduate students, otoh, are conditioned not to even try to hide their feelings and preferences. They have been misled into believing that their opinions, feelings and moods represent nothing less than Truth, Justice and the American Way. Superman would envy such iron-clad, bullet-proof, corrosion and rust resistant self esteem.
Second, everyone knows and understands point number one. When they encounter a FGS pollster, they assume rule one applies, and unless they want to get hassled or get into a debate, they will, deliberately or subconsciously, tell her what they think she wants to hear.
Third, as P.J. O’Rouke observed, when he was in college a lot of guys feigned liberal attitudes, because that’s were the chicks were, and liberal chicks are easy. So guys will adopt liberal attitudes and cliches, and sometimes, you know, half convince themselves that they believe that crap. They really just want to get into your pants.
Indeed a fine tool.
What I remember about it is the null has to be that the factors are NOT related [independent] ie null could be the percentages are NOT the same as in the voter registrations
Then if the null is rejected, accept the alternate hypothesis and hence percentages of poll respondents IS related [dependent on] party ID... there IS a lurking variable.
I always liked stat. Math but real world.
How ANY catholic could vote for The Kenyan after what he is doing to their doctrine is beyond me.
Rush just reads Drudge on the air. I can read that myself. Sean knows less about conservatism than my 11-year old - a "Cliffs Notes" conservative who means well but isn't terribly bright.
Mark Levin is the only one of the three worth your time.
After Wed night’s debate I imagine you will see some changes in the polls. Mitt is going to look very presidential and very sharp.
Other than Ann Romney, no one is coming out to vote for Romney. They are coming out to vote against the other guy.Romney’s strength in this race is the other guy’s failure. He needs to capitalize on it. To date, he has not.The only message Romney should have 24/7 is the other guy is a failure. Eye on the ball, please.
Romney needs to take the tack of “It’s not that I’m rooting for Obama’s policies to fail. It’s that I understand that the government cannot spend its way out of debt and into prosperity any more than you can. Eventually, you have to pay your bills and despite the fancy talk and complicated money policies, so does the government.”
A nine percent response rate is unacceptable for social science survey work. If you dont have at least a 30% response, you couldnt get a paper based on the survey published in any reputable journal in the social or behavioral sciences.
Agree. Very similar in the corporate world as it relates to surveying employees.
Rush just reads Drudge on the air.
And other articles, which you would be knowledgeable on if you surf the net before his show.
Most of the time I find him covering topics that FR has posted, and commented on .
Do enjoy his take, so he’s worth listening too.
Agree that Levin , is the take no prisoners guy with very precise questions, and demands for equal responses.
Sean if informative, and can be a bulldog, but needs to avoid the crosstalk on his panels, and focus on 1 or 2 points.
You don’t have to prove yourself on each and every comment.
Rush just reads Drudge on the air. I can read that myself.
Yes but he pretends that he “discovered” each news item by only mentioning the linked news outlet and never Drudge. Rush needs to pack it in and buy a nice Island... he’s been worthless since about 2006/shamnesty etc.... he’s been threatened into silence over eligibility and other hot-button topics... maybe he should let one of his fill-ins take over... He’s mentioned dozens of times that he’s bad with money and that he has “safe” holdings (which I’ve taken to mean bonds or bond funds) ,,, he could be bankrupt shortly when rates increase and he can’t unwind fast enough with all his fixed costs... Wouldn’t bother me in the least as he is so damn arrogant regarding people that have lost their life savings due to world changes...
The simplest chi-square test is a test for independence, but there is a fancier version that tests the likelihood that a set of responses with any number of discrete values (say D, R and I) was drawn at random from an population with given frequencies, which is what I’m advocating using here.
(Actually the last time I did a chi-square test it was to see whether the random number generator used on the online hex-grid wargaming site hexwar.com simulated a fair die. It rejected the null hypothesis, but the folks who run the site didn’t take notice to my objection. I’m a pure mathematician not a statistician, though I keep up my stats since my wife does psychological research and I end up helping out when she works at home and can’t consult a proper statistician in a timely manner.)
Hannity has gotten so stump-hole-stupid that I will listen to my tinnitus rather than listen to Hannity.
Well, there are Catholics, and then there are Catholics. These purport to call themselves Catholics, yet are four-square behind the Kenyan:
Sorry, don’t SEE him quoted here much.
I attend a couple different Catholic Churches. Recently at the one, a small congregation of maybe 250, the priest discussed the healthcare mandate and how there will be civil disobedience. The entire congregation stood and applauded him. At another church I also attend, the congregation is ten times the small church. When the priest discussed the horrors of abortion, 2 people stood and walked out. The remaining 2498 congregants sat quietly and listened intently.
I have hopes for my religion of choice.
>> “I have hopes for my religion of choice.” <<
Christ is not “religion,” he is The Way; religion is all else.
And forget about Sean; he works for Murdoch& Co.
>> “Surfing Youtube for oldies but goodies like The Seekers, Carpenters, Jay and the Americans, Gary Lewis and the Playboys and Pet Clark is a good diversion from ObamaCRAP.” <<
The above are all socialist dupes; how would they be a good diversion?
>> “Sean if informative...” <<
Give him a clue, he hasn’t found one on his own yet.
Sean is there because he is too stupid to know that he is being used... ...simply because he is stupid.
“Safe holdings” means prescious metal bullion coins, and ‘junk’ silver.
No honest investment councellor would call anything else safe.
At my parish, the message from the pulpit is: Holy Mother Church is under attack. Resist we much.
>> “Hannity has gotten so stump-hole-stupid that I will listen to my tinnitus rather than listen to Hannity.” <<
Also young women are often uncomfortable approaching people. As a result they seek out people to interview whom they feel most comfortable with. That’s probably not they 55 year old guy with the John Deere cap.
Safe holdings means prescious metal bullion coins, and junk silver.
No honest investment councellor would call anything else safe.
I agree with you but Rush has shown an unrealistic “faith” in the markets and the people that manipulate them, taking the side of the banksters on the air repeatedly... he could be wiped out if he holds debt instruments.
Rush may be sufficiently in with the elite that he would be notified before the Fed or Euro bank pull out of the markets.
Well said. I couldn't agree more.
On the other hand I know several people who aren't lazy, food stamp collecting types who are going to vote for Obama. Even though some agree wholeheartedly that everything sucks right now they just can't bring themselves to vote for any republican. I swear sometimes people won't change until they hit rock bottom face first.