Real time Facebook presidential poll shows a completely different story than mainstream media polls

a reply to: Byrd

I tried to explain this about 8 pages ago.

I hope you have better luck than I did.

a reply to: Hazardous1408

Seeing as how democrats vote twice, and felons and illegals vote democrat- how are any numbers reliable?

a reply to: kaylaluv

A womens group. Lol, basically five hundred feminists hating on men. Just lol.

a reply to: Byrd

Weird speculation there. I have a good mix of friends on FB. I watch my FB friends banter about politics with their own friends, whether they agree or not.

In the adult world, people don't defriend everyone on their FB friend list who has an opposing political view unless a person is completely obnoxious about their politics.

Also, absentee ballot requests are not an unvarnished view when you consider that the HAVA mandates every state register voters who don't provide a DL# or SS#. The laws are so weak and open to exploitation, I have no faith in them. You can register fraudulently in many swing states and cast fraudulent votes -- but no red flags would even be raised in the system.

originally posted by: Byrd
Heck... look at YOUR feed (if you have one) - how many people on yours are a different race, different sexual orientation, different philosophy/religion, different socioeconomic category.

BTW, I have Atheists, Agnostics, Christians, Muslims, and even a Buddhist on my FB friend list. I have a drag queen, multiple lesbians, gay men, and straight people on my FB friend list. I have multiple black men, women, white people, and other nationalities and races on my FB friend, list, too. I even have a few mixed couples. The socioeconomic range is as broad as you can imagine, although I don't believe I have any multi-millionaire FB friends.

Wow.

I guess it's safe to assume your FB friend list is pretty stale and limited to make you feel safe? Good god...I am so happy I don't live life like that.

My neighborhood is diverse, too. I like it that way.

a reply to: MotherMayEye

No disrespect intended, but perhaps you are the exception rather than the rule.

Your experiences are not the measure of all things.

Besides that, why the low-grade snideness directed at another member who is obviously trying to discover the truth?

While you admit that this "facebook poll" is scientifically worthless ... you're still "fascinated" by it.

Fair enough. But why do Byrd's observations (which are, by the way, right on point and logical as opposed to most of what is transpiring in this thread) seem to threaten you?

Surely, we're all looking for the best evidence we can find ... right?

There are some counties in some states that are 100% for Trump, Trump signs everywhere and not one Hillary sign.
Anecdotal maybe, but you can be sure that if those counties come out 49 Trump 51 Clinton those voters are going to scream rigged. And you can't have that level of visual obvious support on the ground and lose. That is not some rigged internet poll it's real life.

originally posted by: BlackProject

originally posted by: Greggers

When Trump loses, will you come back and acknowledge that scientifically conducted polls use reliable methodology, while Facebook polls have no statistical relevance?

Wow. You really must love Hilary. A war propagating, corrupt lawyer and wife to a pervert. She is a cracking gal!

Any poll is a poll. What a joke.

I actually think Hillary is unlikable, untrustworthy, and generally shady. So... No.

And while any poll is indeed a poll, only scientific polls produce statistically valid results.

originally posted by: Greggers

originally posted by: BlackProject

originally posted by: Greggers

When Trump loses, will you come back and acknowledge that scientifically conducted polls use reliable methodology, while Facebook polls have no statistical relevance?

Wow. You really must love Hilary. A war propagating, corrupt lawyer and wife to a pervert. She is a cracking gal!

Any poll is a poll. What a joke.

I actually think Hillary is unlikable, untrustworthy, and generally shady. So... No.

But... you're still going to vote for her.

originally posted by: Rezlooper

originally posted by: Greggers

originally posted by: BlackProject

originally posted by: Greggers

When Trump loses, will you come back and acknowledge that scientifically conducted polls use reliable methodology, while Facebook polls have no statistical relevance?

Wow. You really must love Hilary. A war propagating, corrupt lawyer and wife to a pervert. She is a cracking gal!

Any poll is a poll. What a joke.

I actually think Hillary is unlikable, untrustworthy, and generally shady. So... No.

But... you're still going to vote for her.

No, I'm not.

Where do you clowns come up with this nonsense?

a reply to: Greggers

Greggers are you in a Swing state ?

originally posted by: Stormdancer777
a reply to: Informer1958
according to wikileaks it is rigged

No, it's not.

originally posted by: Blue_Jay33
a reply to: Greggers

Greggers are you in a Swing state ?

Nope.

And even if I were, I wouldn't vote for Hillary. I'm trying to get Gary Johnson the percentage he needs to qualify for federal funding next go round.

originally posted by: Byrd

originally posted by: MotherMayEye
I have a hard time believing a huge sample size, like this poll, would yield a margin of error of +/- 45 percentage points.

I find this FB poll interesting because of the enormity of the sample size -- even with all the possible errors it encompasses.

The problem with the poll (which hasn't shown up on MY Facebook feed or that of any of my friends (a fairly large sample) is that it just reaches his friends....

...and the friends of his friends...

...and the friends of those friends.

I will wager you that they've blocked or don't have many liberal SJW friends. Heck... look at YOUR feed (if you have one) - how many people on yours are a different race, different sexual orientation, different philosophy/religion, different socioeconomic Catergory

Exactly. This FB poll suffers from self-selection bias - you're characterization of who the poll reaches is a perfect example of self-selection bias for any of you future econometricians/statisticians/etc.

originally posted by: Rezlooper

originally posted by: imjack
a reply to: StallionDuck

originally posted by: CrapAsUsual
What I thought was that the FB public was in general younger and more pro democrats.

It is. You clearly don't understand scale as Facebook has 1.7 billion users and this 'poll' has 19,000 'votes'.

Unless the Republican Party is less than 19,000 people, or the entire party does not use Facebook, almost no part of such a small number being compared to such a large number is surprising that you can deliberately draft that many votes in anyway. It makes you look sort of inept at math and technology by constantly drawing this comparison.

Using that logic, how does a "scientific" sampling poll of 1,000 likely voters mean anything when America has over 300,000,000 citizens?

Here's some math. Knock yourself out.

www.qualtrics.com...

a reply to: Greggers

Have you not read the wikileaks emails? Podesta sent the rigged poll criteria to his shills in the media to do. lol

originally posted by: amfirst1
a reply to: Greggers

Have you not read the wikileaks emails? Podesta sent the rigged poll criteria to his shills in the media to do. lol

Yes, unlike you, I actually have read that ENTIRE email thread.

I've even read most of the attached booklet from Atlas.

Would you like me to walk through it with you?

a reply to: Greggers

It's not scientific when they poll 25% more Democrats, a lot more women, and hispanics. The polls have already been debunked and a fraud and coordination by Podesta and the media.

originally posted by: amfirst1
a reply to: Greggers

It's not scientific when they poll 25% more Democrats, a lot more women, and hispanics. The polls have already been debunked and a fraud and coordination by Podesta and the media.

I presume you are referring to the email in which Podesta asked Atlas to recommend oversamples.

Some important points:

1) Both Oversampling and Undersampling are techniques that can be used to make a dataset more accurate.
2) The Oversampling in question was used (as is commonly the case) to provide in depth analysis of particular demographics
3) The polls in question were internal Democrat polls designed to improve the efficiency of their messaging to potential voters, not public polls. If you had read the Atlas attachment, you would know this.
4) Every single oversample recommendation from Atlas was in the attached booklet, along with the reasons why. Go read them for yourself. They do not match up with the oversamples you're citing in ANY WAY, and in all cases the goal was greater clarity for voter outreach.

With regard to your other assertions, I think you need to ask yourself why fewer people would identify as Republican in a statistically random sample.

originally posted by: Greggers
Legitimate polling organizations poll RANDOMLY, with sample sizes that give reliable margin of error.

Facebook polls measure whoever voluntarily shows up to take the poll.

I don't really use Facebook, but as I understand it, the platform uses algorithms which result in social media bubbles where people typically hear opinions they agree with, so chances are this guy is connected to a lot of right leaning people and that will skew the results. The exact same thing happens on ATS and on the internet in general. I remember during the last elections, if you were to guess who was going to win based purely on ATS comments and even comments all around the internet, you would have guessed that Ron Paul was the favourite, but he didn't even make it into the debates. That taught me never to trust one single sample group, even if that group is the entire internet, because actually there's a lot of people who don't use the internet or just don't go sharing their opinion online. At this point I'd say it's fairly clear Hillary will win the election, no one really cares about her email scandals compared to the recent Trump shenanigans.