Trump vs Clinton Polling - Are they skewed?

I have been taking a look at the polling over the last month or so - and the make up of those numbers.

RealClearPolitics shows Clinton with a current 4.8 point lead over Trump, however the variances are higher between polls than they were a few months back. For example here are two extremes on either side.

Rasmussen - Trump +4
Reuters - Clinton +10 It's the Reuters poll that has gained a lot of press covering this huge margin.

However, here is the breakdown of that poll:
With Republicans : Trump +63
With Democrats : Clinton +68
With Independents : Trump +11

The numbers here look odd when you consider that Clinton is up 10 points overall in the Reuters poll, so I looked at party affliations. The poll was made up of about 1500 voters.

44% identified themselves as Democrat or lean Democrat
30% identified themselves as Republican or lean Republican
26% independent or other

That seems like a large skew towards Democrats... from Gallup for the USA Gallup Party Affiliation as at June 23rd 2016
Democrats 31%
Republican 28%
Independents 39%

This polling by Reuters looks a long way out and they cant control the sample as it's an online poll.
If it's weighted according to the Gallup data then Trump would actually be leading, or about level.

Or is it that people are switching their affiliation in their answer along with a voting intent for Clinton?
I would also note that the Rasmussen poll is skewed towards Trump voters and with weighting, Trumps 4 point lead would become a 1 point lead for Clinton.

Polling JMO is a waste of time.

First you have to FINALLY get someone on the phone who doesn't hang up on you.
Then since the questions are formatted to render the answers wanted by the pollster....well you aren't getting anything near the answers needed by the public.

Having been polled in the past I was appalled just how biased the questions were.
Nothing as simple as "which one are you voting for" was asked.

I don't think the polling is that much in favor of the Democrats right now, and I'll tell you why:

It isn't that much in the media. During both 2008 and 2012, the polling was very central to the media campaign and it was in the Demcorats' favor much more than many wanted to admit or believe.

This year so far we have very few polling stories and a lot more stories about Trump's high unfavorables as if they are the equivalent of a poll. That tells me there isn't that much favorable polling data, and if you look at the internals on even those polls that do show a high percentage for Clinton, the numbers breakdowns often show the Democrats getting beat on all issues which really isn't a good thing at all for them and ominous going into a major election cycle.

Every single poll bar none said that Britain would stay in the EU

So yes, polls are skewed

And no, the intentional and blatant skewing does not always affect the outcome.

Stay tuned

a reply to: UKTruth

The numbers here look odd when you consider that Clinton is up 10 points overall in the Reuters poll, so I looked at party affliations. The poll was made up of about 1500 voters. 44% identified themselves as Democrat or lean Democrat 30% identified themselves as Republican or lean Republican 26% independent or other

It seems weird for me.... when every Democrat that I know personally says that they will not vote for Hillary Clinton. These Democrats that I speak of are aware that I do not like any of the candidates that were available to voters during the primary election cycle, so I think they are being honest with me.

Any old person can tell you they remember a time when polling was illegal, because it 'skewed' elections before the actual voting.

But then so is waging unjust aggressive war, detaining people indefinitely without charge, friggin gunning them down in the streets for 'not complying'.

Don't complain because your kids are in the jail, hospital, morgue or cemetery, you completely forgot what your rights are about 'Murica.

One poll study I read a while back by a university (can't remember) found that people who supported Trump would often say they were undecided but when polls were conducted anonymously Trump's numbers went up. The conclusion from the data was many would vote for him in a booth but they do not want anyone knowing in a poll. Makes sense with the media labels on him. Another report showed how pollsters game the polls, throw out data they don't like. All I know is Trump has a lot of support. GOP tried to oust him and he is just too strong. They tried everything. DNC is trying everything and he is still very strong. Hundreds of millions spent on the effort and almost no effect.

Are they skewed? No, more like we're screwed with either one.

a reply to: UKTruth
British polls have been tending to underestimate the Conservative vote for decades (when Heath won, when John Major won, when Cameron won), and I'm wondering if this is the same phenomenon.
My theory goes like this; the left wing try to present a Conservative vote as something to be ashamed of.
This propaganda does nothing to change the voting intentions of potential Conservative voters, but it may induce them to conceal their voting intentions even from the pollsters. They say brightly "Labour" or "Don't Know" and then go out and vote Conservative. In the referendum, the Remain campaign was trying the same "shaming" tactic on the Leave campaign, with the same effect of causing the polls to underestimate the Leave vote.

As far as I can see, the Trump vote is supposed to be the "shameful" option in this election, so the phenomenon might work in the same way. Trump voters would keep their intentions to themselves until they reached the polling booths.

originally posted by: DISRAELI
a reply to: UKTruth
British polls have been tending to underestimate the Conservative vote for decades (when Heath won, when John Major won, when Cameron won), and I'm wondering if this is the same phenomenon.
My theory goes like this; the left wing try to present a Conservative vote as something to be ashamed of.
This propaganda does nothing to change the voting intentions of potential Conservative voters, but it may induce them to conceal their voting intentions even from the pollsters. They say brightly "Labour" or "Don't Know" and then go out and vote Conservative. In the referendum, the Remain campaign was trying the same "shaming" tactic on the Leave campaign, with the same effect of causing the polls to underestimate the Leave vote.

As far as I can see, the Trump vote is supposed to be the "shameful" option in this election, so the phenomenon might work in the same way. Trump voters would keep their intentions to themselves until they reached the polling booths.

Very good point. I think there is some truth to it.
It's all looking pretty close in the swing states and I am not sure Trump has much more downside...

The hardest part is Trump is so off road no way to really gauge it. He could in fact be way down or he could be way up. I think this really worries the DNC. They run these very industrial style campaigns with not much ability to get off course. If they do they will lose voters. Trump can say or do just about anything and won't lose any support. He likely gains small chunks with all the rallies.

originally posted by: LifeMode
The hardest part is Trump is so off road no way to really gauge it. He could in fact be way down or he could be way up. I think this really worries the DNC. They run these very industrial style campaigns with not much ability to get off course. If they do they will lose voters. Trump can say or do just about anything and won't lose any support. He likely gains small chunks with all the rallies.

Indeed, I thought it was a crazy decision for Clinton to go on the attack yesterday in NJ. The whole country is told she is actually incompetent in even the most simple task of keeping emails secure and she then she hits out. In this case attack was not the best form of defence at all. She is not being well advised.

originally posted by: UKTruth

originally posted by: LifeMode
The hardest part is Trump is so off road no way to really gauge it. He could in fact be way down or he could be way up. I think this really worries the DNC. They run these very industrial style campaigns with not much ability to get off course. If they do they will lose voters. Trump can say or do just about anything and won't lose any support. He likely gains small chunks with all the rallies.

Indeed, I thought it was a crazy decision for Clinton to go on the attack yesterday in NJ. The whole country is told she is actually incompetent in even the most simple task of keeping emails secure and she then she hits out. In this case attack was not the best form of defence at all. She is not being well advised.

Agree, then the Air Force One power move. Serious mistake. It was an effrontery to the hard work of the FBI and taxpayers.

a reply to: UKTruth
Yes they are and here is an example which while they provide their methodology, do you believe most people look beyond the numbers or that the methodology is told on television? Of course not.

This is from June.

abcnews.go.com...

Their methodology.

This ABC News/Washington Post poll was conducted by landline and cellular telephone June 20-23, 2016, in English and Spanish, among a random national sample of 1,001 adults, including 836 registered voters. Results have a margin of sampling error of 3.5 points, including the design effect, for the full sample, and 4 points for registered voters. Partisan divisions are 36-24-33 percent, Democrats-Republicans-independents, in the full sample, 37-27-30 among registered voters.

originally posted by: ksiezyc
a reply to: UKTruth
Yes they are and here is an example which while they provide their methodology, do you believe most people look beyond the numbers or that the methodology is told on television? Of course not.

This is from June.

abcnews.go.com...

Their methodology.

This ABC News/Washington Post poll was conducted by landline and cellular telephone June 20-23, 2016, in English and Spanish, among a random national sample of 1,001 adults, including 836 registered voters. Results have a margin of sampling error of 3.5 points, including the design effect, for the full sample, and 4 points for registered voters. Partisan divisions are 36-24-33 percent, Democrats-Republicans-independents, in the full sample, 37-27-30 among registered voters.

Ah so the WaPo poll has a big DEM population in it too... higher than the estimated party affiliations

The only polls worth listening to are the ones RIGHT before an election. They are the most accurate. Other than that they are practically useless.

I would suggest ignore rasmussen. they actively skew their polls for reasons that are beyond my understanding.

here they are explaining their calling that Romney will beat Obama
R money

This link may be helpful in trying to figure out which polling is most accurate

2012 polling results

originally posted by: SaturnFX
I would suggest ignore rasmussen. they actively skew their polls for reasons that are beyond my understanding.

here they are explaining their calling that Romney will beat Obama
R money

This link may be helpful in trying to figure out which polling is most accurate

2012 polling results

Useful, thanks - IBD were the best last time round.. this year they have it Clinton +4

I'm not a Trump supporter but you can be sure I will be voting for him...because he is the better option BY FAR than Secretary Clinton...

So, people like me could easily be under reported...

Nixon coined a brilliant phrase in regard to draft dodging hippie protesters: "A silent majority exists" that diaagrees with radicals....or something to that effect...

Anyhow, I believe there are MILLIONS of people like me who are horrified at the thought of another Clinton President..

-Christosterone

originally posted by: UKTruth
Useful, thanks - IBD were the best last time round.. this year they have it Clinton +4

and that sounds about right. Still early though. If Trump ever becomes more statesmen, he may close that gap..but I am thinking that is like asking a fish to act a bit more like a bird.