Lefty Math

I am looking specifically for some math that would show what the probability is of online polls being flawed or correct in regards to the current situation where almost every online poll shows a decisive victory for Trump in the recent debate.

I am NOT looking for lefties to jump in here and blurt out their emotional convictions of why Hillary won the debate.

I am not well educated in math and statistics, but am interested in knowing how likely it is that the almost unanimous online polls showing Trump's decisive victory are as wrong as the media and lefties who have an interest in seeing Trump as the loser are implying. I am interested in mathematical speculation, not political.

I am not interested in comparing the accuracy of online polls vs scientific polls, I fully understand that various methods of polling have various degrees of accuracy. I am interested specifically in the contrast of an overwhelming majority of online polls showing Trump the winner vs the media claiming Hillary the winner. What is the probability that the overwhelming number of polls are all wrong, and that the lefty controlled media is correct?

For those who did not understand that I am seeking math here, and not political assertions, I will remind you here in the last section of the OP that you misunderstood what I am after.

 

Its rather straightforward.

In on-line polls, a person can "vote" any number of times. The polls can be very susceptible to robot voting as well.

In person polling, a person can only "vote" once. not to mention getting more accurate demographic info, which is used in weighting.

Doesn't really matter what the math is when the source data is so susceptible to massive manipulation by a very few.

 

I suspected that this kind of discussion would emerge instead of the actual math, and am quite aware that it is possible to manipulate polls. I am interested here in finding out some kind of mathematical probability that the huge number of independent online polls have been compromised vs a situation where they really do reflect something closer to reality. It does not seem probable to me that such a broad number of independent online polls could come up with such similar results, regardless of potential ways where these polls could be compromised, if the results are allegedly flawed. I am no math or probability expert, but I would think that the broader the number of these unscientific polls who come up with the same independent results is, the more credibility their results have as a whole.

The large number of poll results that show Trump winning vs claims of the folks who have an interest in seeing Trump as the loser seems like an elephant in the room. Do we have any math people out here in the community who can shine some light on this in mathematical terms?

 

An interesting question, for which there is no mathematical answer. Scientific polling is done with the object of composing a representative sample of the population. The readership of the Drudge Report is not a representative sample. There is no mathematics that can recover any information about the feelings of the American public about the debate from such polls except that the number of people who thought Trump won the debate is not zero.

There is no way of knowing just how unrepresentative, and in what way, the sample is without conducting a scientific poll, in which case you might as well just use those results directly.

 

Perhaps the folks at Fox et.al. can illuminate this issue for you:

http://www.businessinsider.com/fox-news-online-debate-polls-trump-drudge-2016-9

 

"While mainstream media and pundits quickly crowned Hillary Clinton the winner, Trump touted several online polls showing that he triumphed.
“Such a great honor. Final debate polls are in — and the MOVEMENT wins,” Trump tweeted.
Online polls aren’t given much credence because they aren’t scientifically representative samples of the electorate.
But The Daily Dot suggested online results from about 70 debate polls were particularly egregious because Trump supporters “artificially manipulated” the results “to create the false narrative” that he won.
Trump supporters using Reddit and 4chan message boards bombarded the online polls and spread the effort to Twitter to catapult the hashtag “#Trumpwon” to the No. 1 trending topic."
http://nypost.com/2016/09/27/online-polls-showing-trump-won-the-debate-were-rigged/

 

It's simple, the public/online polling didn't come out in Hillary's favor so, the media has to immediately discount those polls in favor of selected individuals being polled. OR more simply, when the polls are against your candidate there MUST be something wrong with them.

 

I can see what you are after..You want to confirm your convictions that online snap polls are an accurate depiction of reality. So ask Trumpeizdas and Trump Turddettes to respond to you and ignore all other people who are rational and honest.

 

If you constipated conspiratal view had any reality why did the big bad MSM report that Trumpeizdas has closed to a virtual ties in virtually all of the battleground states. If the big bad MSM was as corrupt and constipated with conspiracybas you believe why have they reported that Trumpeizda has caught up with thecClinton Womam ?

 

You seem to have misnamed your thread. It should have been titled, "Another Conservative Refuses to Accept Reality".

 

any internet polls that says Hillary only has 10% support, is either a very localized poll on a right-wing website, or right-wingers are flooding the poll or voting multiple times using proxy servers.

- - - Updated - - -

any good national poll will show that Hillary and Trump are basically neck and neck, give or take 3-5 points

 

Why do you feel the need to get nasty?

 

Online polls do not accurately measure anything beyond the views of the poeple responding. There is no way to correlate the responses in an online poll wothany cross section of the general population without knowing a lot about the people who respond.

 

Here is a paper on the subject: http://math.arizona.edu/~jwatkins/505d/Lesson_12.pdf

 

and they can VERY easily be manipulated by the responders.

one person, if patient enough, could vote thousands of times.

 

I have had coursework in statistics to econometrics, and I know of NO probabilities for such a situation.

But I think you answered your own question. It is less likely that several online polls in full agreement with each other than liberal media colluding to manipulate so-called "scientific polls." Those are easier because it only takes a small group, a single representative from top pollsters, and then following up to manipulate said polls.

One happened recently where Trump was ahead after the FBI announcement and a leading pollster announced in a press release they were changing their methodology to exclude "not decided" from their choices, and that supposedly flipped Trump's lead to Clinton leading. Go figure.

Steve

 

that's quite a conspiracy theory.

do you have any evidence that NBC, CNN, ABC, CBS, have conspired to fake the results of their scientific random polls?

did they also collude with Fox News, who also often agree with their polls?

 

You're still having reading comprehension problems.

Steve

 

and you are now on Ignore.

bye

 

Online polling isn't random. It comes from people who choose to take the poll, not people chosen at random. That means that a few enthusiastic people can skew the poll. If online polls were accurate, Ron Paul would have been Republican nominee in 2008 and 2012, and he would be President now. Online polls are not scientific. To be scientific, polls need to be random samples of the population being surveyed. These days most polls are problematic, as it's harder to get a data source to pull random samples from. In the days of the landline, it was easy--just use directory listings. Without random samples, polls are useless.

This isn't a math thing, it's an assumptions of random sampling thing. With random sampling, less than a thousand people can pretty accurately reflect the U.S. population as a whole. Without random sampling, you'd have to measure all of them.

 

I've taken two years of graduate research classes. It's not a math question, it's a sampling question. Online polling doesn't have random sampling, hence it can't accurately reflect the views of a population.

 

If you believe that you must think Ron Paul was the Republican nominee in 2008 and 2012. Ron Paul dominated online polls, despite being horrible at the real thing. The race is a tossup, and will be close until the end is my prediction. The scientific polls pretty much bear that out. Neither Trump nor Hillary get much out of the low 40% range.

 

Ouch! You seem a little upset.

 

I think I made myself very clear what I am after. Below is a quote from my OP. You were unable to process or respond to the OP correctly. Below is what you were wanting to divert attention away from when you derailed the thread.

 

Please compare your response to this quote from my OP, and tell me how your response is related to my OP: