Polling Perils

[Rexx Taylor]

my latest poll taken amongst just normal people who work for a living. Ted Cruz 101%,,,Hillary...0.0%...Dead People...32%.

 

[Old Rocks]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

my final conklusion of what happened in 2008 and 2012 are both the same. Obama won because "People Are Stupid". (the infamous catch phrase). Only the dumbest Americans vote to collapse the economy, allow terrorists/illegals to cross the border and dont see a problem with our enemies building nuclear weapons. {And will the loony lefties stop pointing out my speelling errors! I do on poropse!}

Wow, just wow!

The economy was already collapsed. In case you hadn't noticed

In January, 2009, the unemployment rate stood at 7.8, and was heading rapidly south, thanks to the Bush Administration. By October, it was down to 10.0. However, by Nov.2012, it was back up to 7.7, in spite of GOP efforts to prevent any improvements. And today, it stands at 5.4.

Bureau of Labor Statistics Data

Now, after 9-11, Bush could have closed the southern border, and all would have supported him. He did nothing, and we do not know who or what came across the border from then until 2009. Cheney's company, Halliburton, sold much of the equipment needed for the construction of nuclear plants to Iran.

Cheney Pushed for More Trade With Iran Fox News

2. Halliburton Charged with Selling Nuclear Technologies to Iran Top 25 of 2007

 

[Mertex]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

That must have been what happened in 2008 and 2012, right........

 

[Mertex]

amongst just normal people

Herein lies the problem.....you aren't able to recognize normal people anymore.

 

[Mertex]

Most people predicted a Romney win. They did not count on the huge turnout of black and illegal voters in key states.

What planet do you reside in? And "most people" is not the same as "most polls" - you're comparing "most people" meaning your circle of friends and GOP groups and people that call in to Rush Lamebaugh and Faux News.....of course, Romney was going to win among them. Megan Kelly thought the same thing....right to the end...no, way past the end....

 

[The Rabbi]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

What is "mean" about a British conservative? Why would they have that perception? Do you think American conservatives are perceived to be "mean"? Does degree of conservatism indicate degree of meanness?

You're kidding, right? How many times have conservatives been vilified on this site as "mean", as waging "war on women/black/minorities/the poor etc etc ad naus."?
Basically any proposal to scale back the disaster that social spending has been is greeted with a chorus of "meanies."

 

[NYcarbineer]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

What is "mean" about a British conservative? Why would they have that perception? Do you think American conservatives are perceived to be "mean"? Does degree of conservatism indicate degree of meanness?

You're kidding, right? How many times have conservatives been vilified on this site as "mean", as waging "war on women/black/minorities/the poor etc etc ad naus."?
Basically any proposal to scale back the disaster that social spending has been is greeted with a chorus of "meanies."

This from a guy who's said that blacks as a race are unfit for leadership positions.

 

[Statistikhengst]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

There may be a bit of truth in that in the US, but not much. For example, when asked in telephone polls, 40% of Americans have generally identified themselves as conservatives. However, in exit polls, conservatives are around 35% of the electorate. This means that, if true, only 10%-15% of conservatives, and 1 in 20 Americans, fibbed about being a conservative in the exit poll. That Republicans have lost the popular vote in 5 of the past 6 Presidential elections suggests this isn't an issue in the United States.

The last time I remember watching a British election where the exit polls were wrong was in 1992, when the BBC, based on exit polls, called a majority government for Neil Kinnock when in fact, John Major wound up winning. When the BBC did an analysis after the election, they found that a sizable minority voted for the Tories but then said they voted for Labour when asked by an exit pollster after they left the polling booth.

Polls in the last Canadian election were also wrong. All of the polling firms had the Tories around 34% going into election voting, which probably would have meant a Conservative minority government, but the Tories wound up winning 38% and a majority government.

But people misunderstand polling. Polling isn't infallible. A poll is an attempt to get a snapshot of the electorate at large. It is a probability assessment, not a certainty.

The polls were pretty much spot on during the last two Presidential elections. In fact, in 2012, the polls underestimated Obama's strength. It was conservatives who were wrong in 2012, and embarrassingly so, with all these "skewed polls" and such nonsense. For example, Republicans genuinely believed that they would win Pennsylvania. But in the 45 state polls prior to the election, Obama led in 44 and Romney was tied with Obama in 1. Clear-headed analysis would have suggested the Republicans would have had a very long shot in PA, but they were so blinded by their hatred for Obama that they created all these fantasies that they were going to win.

Yepp.

Every single poll here, and every single end-poll compared to the actual results.

Statistikhengst s ELECTORAL POLITICS - 2015 and beyond The moment of truth how did the pollsters do

Overall, on the aggregate, there was a slight to medium Conservative mathematical bias in the polling, sometimes for very specific reasons in specific regions of the country.

In 2014, the aggregate polling was off to the LEFT, and in some cases, extremely to the left. Reason: voter turnout that went through the floor. Midterms are notoriously hard to gauge and it would be a mistake to compare mid-term polling to GE polling or visa-versa, also to compare polling for US elections to polling elsewhere on the planet.

As far as I can tell, the only firm to have polled both the US 2012 GE and the British elections from last week was YouGov, a british polling company, and they do internet only surveys - a process still in development.

 

[NYcarbineer]

?

No, they did not. Not even close.

An Exhaustive Collection of Pundit Predictions -- NYMag

 

[Luddly Neddite]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

There may be a bit of truth in that in the US, but not much. For example, when asked in telephone polls, 40% of Americans have generally identified themselves as conservatives. However, in exit polls, conservatives are around 35% of the electorate. This means that, if true, only 10%-15% of conservatives, and 1 in 20 Americans, fibbed about being a conservative in the exit poll. That Republicans have lost the popular vote in 5 of the past 6 Presidential elections suggests this isn't an issue in the United States.

The last time I remember watching a British election where the exit polls were wrong was in 1992, when the BBC, based on exit polls, called a majority government for Neil Kinnock when in fact, John Major wound up winning. When the BBC did an analysis after the election, they found that a sizable minority voted for the Tories but then said they voted for Labour when asked by an exit pollster after they left the polling booth.

Polls in the last Canadian election were also wrong. All of the polling firms had the Tories around 34% going into election voting, which probably would have meant a Conservative minority government, but the Tories wound up winning 38% and a majority government.

But people misunderstand polling. Polling isn't infallible. A poll is an attempt to get a snapshot of the electorate at large. It is a probability assessment, not a certainty.

The polls were pretty much spot on during the last two Presidential elections. In fact, in 2012, the polls underestimated Obama's strength. It was conservatives who were wrong in 2012, and embarrassingly so, with all these "skewed polls" and such nonsense. For example, Republicans genuinely believed that they would win Pennsylvania. But in the 45 state polls prior to the election, Obama led in 44 and Romney was tied with Obama in 1. Clear-headed analysis would have suggested the Republicans would have had a very long shot in PA, but they were so blinded by their hatred for Obama that they created all these fantasies that they were going to win.

Most people predicted a Romney win. They did not count on the huge turnout of black and illegal voters in key states.

LOLOLOLOL

Nobody but a couple of RW polls said he'd win.

Oh and Fox.

 

[Statistikhengst]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

There may be a bit of truth in that in the US, but not much. For example, when asked in telephone polls, 40% of Americans have generally identified themselves as conservatives. However, in exit polls, conservatives are around 35% of the electorate. This means that, if true, only 10%-15% of conservatives, and 1 in 20 Americans, fibbed about being a conservative in the exit poll. That Republicans have lost the popular vote in 5 of the past 6 Presidential elections suggests this isn't an issue in the United States.

The last time I remember watching a British election where the exit polls were wrong was in 1992, when the BBC, based on exit polls, called a majority government for Neil Kinnock when in fact, John Major wound up winning. When the BBC did an analysis after the election, they found that a sizable minority voted for the Tories but then said they voted for Labour when asked by an exit pollster after they left the polling booth.

Polls in the last Canadian election were also wrong. All of the polling firms had the Tories around 34% going into election voting, which probably would have meant a Conservative minority government, but the Tories wound up winning 38% and a majority government.

But people misunderstand polling. Polling isn't infallible. A poll is an attempt to get a snapshot of the electorate at large. It is a probability assessment, not a certainty.

The polls were pretty much spot on during the last two Presidential elections. In fact, in 2012, the polls underestimated Obama's strength. It was conservatives who were wrong in 2012, and embarrassingly so, with all these "skewed polls" and such nonsense. For example, Republicans genuinely believed that they would win Pennsylvania. But in the 45 state polls prior to the election, Obama led in 44 and Romney was tied with Obama in 1. Clear-headed analysis would have suggested the Republicans would have had a very long shot in PA, but they were so blinded by their hatred for Obama that they created all these fantasies that they were going to win.

Most people predicted a Romney win. They did not count on the huge turnout of black and illegal voters in key states.

No. They were wrong. The VAST, VAST, VAST majority of polls in every battleground where it counted showed Obama ahead.

You just lied out your ass, and I can prove it, too:

Statistikhengst s ELECTORAL POLITICS - 2015 and beyond The moment of truth how did the pollsters do

 

[Statistikhengst]

Dick Morrison?

You realize there are a lot of pollsters out there and you stated that 'most' of them were predicting a Romney win. I distinctly remember the polls in general predicting an Obama win. What Dick Morrison predicted is irrelevant.

The aggregate pointed to a definite Obama 303 EV, Romney 206 EV, 29 EV tossup of tossups.

And Obama won: 332 EV to Romney's 206.

And the ONLY polling firms to get the national margin right were Democratic firms.

 

[Statistikhengst]

Most people predicted a Romney win. They did not count on the huge turnout of black and illegal voters in key states.

If by "most people," you mean "Fox News" or "conservative bloggers," you would be correct. But not everyone else. It was the biggest example of mass confirmation bias in American political history.

All one had to do was to look at the polls and take them at face value to know that Obama was going to win since Obama was leading in most of the polls.

RealClearPolitics - Election 2012 - General Election Romney vs. Obama

Not only that, but Obama was leading in most of the swing states going into the election.

These were the swing states.

Swing state - Wikipedia the free encyclopedia

And if you tracked the polls in the swing states prior to the election, you would have seen Obama was leading in most of them.

All you had to do was believe the polls and you would have seen that Obama was going to win. But conservatives couldn't accept that, and spun fantasies to convince themselves otherwise, as they still do to this day.

There were 117 polls taken in Ohio, Obama vs. Romney, from 2010 through November 5th, 2012. You could "drop the needle" at any two-week span, or four-week span and take the aggregate and it was ALWAYS at least Obama +3, more often than not, Obama +4. The aggregate of the polls released within 36 hours of election day showed Obama +3.06. He won Ohio by +2.97, so the aggregate was right on the mark, as a variance of 0.09 is nothing more than statistical noise.

For almost three years, we saw North Carolina as a bitter battleground, from between Obama +2 to Romney +2. On election night, Romney picked up NC, by +2.04. The aggregate was right on the money.

The polling for Florida showed a pig-fuck the entire time. The end aggregate was Romney +0.61. But Obama won Florida by +0.88. So, the aggregate was off to the Right by 1.5 points, well within the MoE.

The aggregates in Missouri showed Romney +10. He won by +10.

The aggregates in WI, MN, OR, NV, CO and PA were all depressed to to shitty (meaning, wrong) numbers from Conservative pollsters. Rasmussen showed a perfect tie in Wisconsin, which Obama won by +7, so RAS was off to teh Right by 7 complete points.

Yepp, the state aggregates clearly showed an Obama win of at least 303 EV.

And the networks were all playing into it. Romney won GA by +7.8 (+8), and Georgia was called within 1/2 hour of poll-closing time on most stations for Romney. Obama, on the other hand, won Wisconsin by +7 and Minnesota by +8 and yet, CNN waited around 2 hours to call the states. Here, FOX was actually faster: it called Wisconsin at 9:36 PM EDT on election night. CNN called it at 11:02 EDT. The stations took about 90 minutes to call New Mexico, althought Obama won it by +10. Romney won Mississippi by +11, and yet, the state was called immediately. Obama actually improved on his New Jersey statistic, won the state by +18, but because of the storm, the state was first called one hour later.

Any one individual poll can be wrong. In fact, most all of them are a tad-bit wrong almost all the time. But the aggregate usually comes very close to reality and when you see 110 out of 117 polls for a state going for one person, it is pretty likely that he is going to win that state.

 

[Statistikhengst]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

Guy, 60% of Brits voted against the Tories. That they couldn't agree on an oppossition doesn't mean anyone thinks Cameron has done a good job.

The scary thing is that Scotland has pretty much decided they've given up on the United Kingdom. So Conservatives will have accomplished something- the End of the Country.

Essentially, yes.

With this kind of unheard of hypermajority in Scotland, it is only a matter of time before secession comes onto the table.

 

[longknife]

Polling is a joke. The only ones who benefit are the ones conducting the polls. And the media's probably getting payoffs for doing it.

 

[Statistikhengst]

How do you square that with the incredible accuracy the polls have for predicting the elections here? They are very accurate right before the election occurs.

Perhaps it is because they ask party affiliation in the exit polls ... And not who the person actually voted for.
I know my grandparents were registered Democrats ... And didn't vote for Democrats half the time.

.

The answer is even easier than that.

Exit polls poll A LOT more people than a standard measurement poll before the election, and exit polls must not factors in undecideds, for those who went to vote definitely "decided".

In the vast majority of states where exit polls were taken, the exit polls were absolutely on the money.

The exit polls were off in Colorado. And they were reduced in margin for Obama in Oregon.

Statistikhengst s ELECTORAL POLITICS - 2015 and beyond The moment of truth how did the pollsters do

 

[Statistikhengst]

Most people predicted a Romney win. They did not count on the huge turnout of black and illegal voters in key states.

No, they really didn't.

In fact, only two polls predicted that the Weird Mormon Robot would win. - Gallup and Ratmuffin. Gallup already had a bias against Obama because he sued them for defrauding the government on contracts and Ratmuffin is of course, Ratmuffin. He lives in his own reality. Four other polls had Obama leading and two had a tie.

RealClearPolitics - Election 2012 - General Election Romney vs. Obama

National polling, 2012, end-polls highlighted in yellow:

Google Sheets - create and edit spreadsheets online for free.

The two best end-pollsters, nationally:

Democracy Corps (D), predicted Obama +4, he won by +3.9
(actually, when you go deep into the poll, it predicted Obama +3.8, he won with +3.86, so the poll really WAS dead on).

Rand Poll, Obama +4.32, 0.53 points to the LEFT of the actual margin. But Rand did an entirely different type of polling. It did internet-only and exactly the same group of people over 365 days, 3 day rolling polling totals. It was a fascinating experiment to watch.

Angus Reid (a Canadian firm Toro ), ABC/WAPO and Pew all predicted Obama +3, so they were essentially off to the Right by 1 point. Pretty decent.

National Journal predicted Obama +5, so it was off to the LEFT by +1, also ok, and a rarity in that election year. The rest were pretty far off.

 

[Statistikhengst]

What? You see everyone who shows up to vote while you are there? That's really amazing. Thanks for the info.

Yeah ... And there is usually a line ... Which indicates there are a lot more people voting than in the exit polls somewhere else.
And ... Since our precinct/district is heavily weighted Conservative ... Then a poll somewhere else certainly wouldn't include that influence.

Now ... You have the information ... Let's see what you can do with it.
Big problem with Democrats ... Always thinking about stupid shit ... Never trying to make sense of anything.

You get points for being a stupid assed clown though ... Troll on peanut.

.

You think I'm a democrat?

You have really hit on something here. Exit polls......being something other than an accurate tabulation of actual votes....are often unreliable. It's a brilliant theory. You should be on TV or something.

The exit polls in both 2008 and 2012 were by and large extremely accurate.
And even the majority of 2004 exit polling still showed the correct winner.
Ohio's exit poll showed Kerry by 1,Bush 43 won by 2, so the exit poll was off to the left by 3, still better than the national polling from 2012....

 

[AceRothstein]

More Republicans/Conservatives showing their ignorance when it comes to math and statistics.

 

[Mertex]

Results from the British election show one clear loser: POllsters. Every pollster virtually had the parties in a dead heat. But in the end it was a slaughter for the Torys. Even Nate Silver, favorite of the libs here, was way off, offering an explanation
What We Got Wrong In Our 2015 U.K. General Election Model FiveThirtyEight

But the truth, which no oe has brought out, is that people tend to lie to pollsters because to admit being for the conservatives is to be perceived as "mean." The same thing probably operates here: people claim to like Hillary or whatever Dem but then vote GOP. This will play out on the coming election for sure.

There may be a bit of truth in that in the US, but not much. For example, when asked in telephone polls, 40% of Americans have generally identified themselves as conservatives. However, in exit polls, conservatives are around 35% of the electorate. This means that, if true, only 10%-15% of conservatives, and 1 in 20 Americans, fibbed about being a conservative in the exit poll. That Republicans have lost the popular vote in 5 of the past 6 Presidential elections suggests this isn't an issue in the United States.

The last time I remember watching a British election where the exit polls were wrong was in 1992, when the BBC, based on exit polls, called a majority government for Neil Kinnock when in fact, John Major wound up winning. When the BBC did an analysis after the election, they found that a sizable minority voted for the Tories but then said they voted for Labour when asked by an exit pollster after they left the polling booth.

Polls in the last Canadian election were also wrong. All of the polling firms had the Tories around 34% going into election voting, which probably would have meant a Conservative minority government, but the Tories wound up winning 38% and a majority government.

But people misunderstand polling. Polling isn't infallible. A poll is an attempt to get a snapshot of the electorate at large. It is a probability assessment, not a certainty.

The polls were pretty much spot on during the last two Presidential elections. In fact, in 2012, the polls underestimated Obama's strength. It was conservatives who were wrong in 2012, and embarrassingly so, with all these "skewed polls" and such nonsense. For example, Republicans genuinely believed that they would win Pennsylvania. But in the 45 state polls prior to the election, Obama led in 44 and Romney was tied with Obama in 1. Clear-headed analysis would have suggested the Republicans would have had a very long shot in PA, but they were so blinded by their hatred for Obama that they created all these fantasies that they were going to win.

Most people predicted a Romney win. They did not count on the huge turnout of black and illegal voters in key states.

LOLOLOLOL

Nobody but a couple of RW polls said he'd win.

Oh and Fox.

You misspelled Faux......