New ABC / WaPo Poll Shows Drop In Trump Favorability Courtesy Of Aggressive "Oversamples"

Here we go again with the fake polls.

Looks like they continue with the same agenda.

We keep hearing all the negative Trump polling.

I bet they use the same over-sampling methods like before the election.

What are they trying to prove anyway?

New ABC / WaPo Poll Shows Drop In Trump Favorability Courtesy Of Aggressive "Oversamples"

In the month leaded up to the election on November 8th, we repeatedly demonstrated how the mainstream media polls from the likes of ABC/Washington Post, CNN and Reuters repeatedly manipulated their poll samples to engineer their desired results, namely a large Hillary Clinton lead (see "New Podesta Email Exposes Playbook For Rigging Polls Through 'Oversamples'" and "ABC/Wapo Effectively Admit To Poll Tampering As Hillary's "Lead" Shrinks To 2-Points"). In fact, just 16 days prior to the election an ABC/Wapo poll showed a 12-point lead for Hillary, a result that obviously turned out to be embarrassingly wrong for the pollsters.

But, proving they still got it, ABC/Washington Post and CNN are out with a pair of polls on Trump's favorability this morning that sport some of the most egregious "oversamples" we've seen. The ABC/Wapo poll showed an 8-point sampling margin for Democrats with only 23% of the results taken from Republicans...

Yeah, they did the same thing during the election. Most people aren't going to take the time to read the footnotes of polls so it is an effective form of propaganda.

The media is trotting out the exact same playbook the establishment in Britain played. First they said it had no chance of passing, then they claimed people regretted their votes. Now they're saying the same thing here. Hopefully people see through this transparent propaganda.

originally posted by: xuenchen
Here we go again with the fake polls.

Here we go again with people who don't understand statistics.

Turns out, all those national polls predicting a hillary win were pretty damned accurate, and most well within the margin of error.

Some of the state polls were wrong, particularly in the rust belt. But the quality/quantity of polling in many states was less than desireable anyway, so people relied on the national polls, assuming that the EC would follow the national vote.

It was a bad assumption, and has proven incorrect twice in recent presidential races.

However, all this talk of "fake polls" is glossing over all this.

There was no agenda among the MSM to gen up "fake polls."

Right about now is usually when someone brings up the Podesta emails and the Atlas Report, which had NOTHING to do with MSM polls.

a reply to: Greggers

Pretty damned accurate, except for the fact that most media outlets played everything up through election night that it take a miracle for Trump to win.

I remember seeing the Hillary lovers on ATS gloating about how Hillary had the election all sewed up.

Yeah, mostly accurate. They just got one thing wrong.

a reply to: xuenchen

Are this the same poll that got Hillary president?

Yeah, I guess they have gotten so low in ratings that they need to bring back the fixed polls result to justify air time this days.

a reply to: xuenchen

Yes and if the polls were favourable you'd all be lauding them.

What's the point these days anyway? Most of the electorate is too stupid to understand polling data anyway.

There's always half that are going to say they are right and the other half which are going to say it's wrong. Because of partisan nonsense.

It'll be the same thing throughout his whole Presidency, whenever a poll looks good, he'll be out on twitter taking credit and as soon as they are bad, he'll blame the media for making them up.

~Tenth

Same thing every day. How will they bash Trump this time? I didn't even vote for the guy, but for Christ sake move on and save judgment until he is at least in office for a few weeks. Will this barrage ever stop?

originally posted by: butcherguy
a reply to: Greggers

Pretty damned accurate, except for the fact that most media outlets played everything up through election night that it take a miracle for Trump to win.

I remember seeing the Hillary lovers on ATS gloating about how Hillary had the election all sewed up.

Yeah, mostly accurate. They just got one thing wrong.

Your're confusing the actual polls with the analysis. The polls were valid statistically, and generated an accurate result.

The national polls, which predicted the popular vote, predicted a Hillary win. Hillary did in fact win the popular vote. Some of those national polls got the percentage right on. Many others came remarkably close, and a huge percentage were well within the margin of error.

The polls were valid.

What was invalid was the assumption that the EC would follow the national vote. Although it's easy to see WHY that assumption was made, as that has almost always been the case.

I doubt they'll be making that mistake again.

a reply to: Greggers

Nate Silver's polls were much better at finding the true election result. Although he still had Hillary winning, it was by a much smaller margin.

~Tenth

a reply to: xuenchen

Same thing today in the UK after Theresa May's speech, apparently her strategy is "NOT" what the voters, who voted Brexit, voted for.

I voted Brexit, get us out of the organisation along with all their rules and laws ......... now i'm being told (certainly by the BBC, that this is not what I , or 90% of the Brexiteers voted for). Certain Politicians are now calling for a vote by the general populus to ratify any agreement which could take up to and over two years.

Parliament ......... we voted OUT, take us OUT. Stop with the spin and BS about looking for another vote

a reply to: xuenchen

"A total of 1,000 adults were interviewed by telephone

That is a very small poll.

originally posted by: tothetenthpower
a reply to: Greggers

Nate Silver's polls were much better at finding the true election result. Although he still had Hillary winning, it was by a much smaller margin.

~Tenth

Yes. His poll of polls gave a 30% chance for a Trump victory. His "polls plus" was pretty close, as I recall.

Since his process basically analyzes the existing polls in aggregate, the truth was out there, as they say.

originally posted by: gortex
a reply to: xuenchen

"A total of 1,000 adults were interviewed by telephone

That is a very small poll.

You'd be surprised how large a population can be accurately estimated with a random sample of 1,000 people.

Here's a calculator where you can work it out for yourself.

www.surveysystem.com...

i guess they forgot we dont care or believe what they say.....so they can make all the phony bs polls they want.


I still havent decided if i actually want them completely gone or to stick around as a living object lesson and reminder to those who know and those who are learning logic of what can happen when journalists abandon their own traditional ethics and professionalism in favor of follow the shifting sand of pop PC culture zombie nonsense.

originally posted by: Greggers

originally posted by: gortex
a reply to: xuenchen

"A total of 1,000 adults were interviewed by telephone

That is a very small poll.

You'd be surprised how large a population can be accurately estimated with a random sample of 1,000 people.

Here's a calculator where you can work it out for yourself.

Probability says you are wrong, 1000 sample size from a population of 600 million? hope you don't do capability studies for a living ..... perhaps i should put the data into Minitab and get some "P" value or other "Quality" statistics

Really?

a reply to: xuenchen

We should know by now that only Positive Polls in favor of tRump are the " Real Polls", the others "sad".

originally posted by: corblimeyguvnor


Probability says you are wrong,

No it doesn't.

1000 sample size from a population of 600 million?

The number of registered voters is about 146 million. If memory serves, a population of 146 million can be estimated with a random sample of just over 1,000 people with a 95% confidence level and a 5% margin of error.

hope you don't capability studies for a living ..... perhaps i should put the data into Minitab and get some "P" value or other "Quality" statistics

Or perhaps you should just stop talking out of your ass.

The balance of power will continue to be held by opinion polls and those who are in a position to print them

originally posted by: Greggers

originally posted by: corblimeyguvnor


Probability says you are wrong,

No it doesn't.

1000 sample size from a population of 600 million?

The number of registered voters is about 146 million. If memory serves, a population of 146 million can be estimated with a random sample of just over 1,000 people with a 95% confidence level and a 5% margin of error.

hope you don't capability studies for a living ..... perhaps i should put the data into Minitab and get some "P" value or other "Quality" statistics

Or perhaps you should just stop talking out of your ass.

OK, remind me to not use your services when deciding on a new machine or process, unless i want to waste money. Or do you not bother with ISO2859 or similar when selecting your sample size? perhaps +/- 1 standard deviation is acceptable to you, i use +/- 6

Don't go for a job in Engineering