Unskewing megathread - Latest: Trump will win in landslide because Twitter

[The Other Castro]

Even without getting into the methodology problems, it's not possible for a group of 13 people to get 50,000 responses in less than 2 weeks. Let's assume an extremely favorable timing of 1 response every 10 minutes (factoring in time it takes to make calls, low response rate, and time to complete a response). That's 500,000 minutes, or 641 hours per person. There's only 336 total hours in 2 weeks.

[Adam Griffin]

Even without getting into the methodology problems, it's not possible for a group of 13 people to get 50,000 responses in less than 2 weeks. Let's assume an extremely favorable timing of 1 response every 10 minutes (factoring in time it takes to make calls, low response rate, and time to complete a response). That's 500,000 minutes, or 641 hours per person. There's only 336 total hours in 2 weeks.

I posted this earlier to that stupid page:

You must be logged in to read this quote.

[Ronnie]

Maker of the app "Zip Q & A" says Trump will win the election.

You must be logged in to read this quote.

You must be logged in to read this quote.

http://www.usatoday.com/story/tech/2016/08/13/app-maker---trump-win-election/88640044/

[The Other Castro]

Even without getting into the methodology problems, it's not possible for a group of 13 people to get 50,000 responses in less than 2 weeks. Let's assume an extremely favorable timing of 1 response every 10 minutes (factoring in time it takes to make calls, low response rate, and time to complete a response). That's 500,000 minutes, or 641 hours per person. There's only 336 total hours in 2 weeks.

I posted this earlier to that stupid page:

You must be logged in to read this quote.

Did you get any responses?

[Adam Griffin]

Did you get any responses?

Nope.

[Desroko]

You must be logged in to read this quote.

I was pretty skeptical right up until HASHTAG LANDSLIDE.

[Wells]

POTENTIAL BREAKTHROUGH: 92% Of Trump Supporters Say That They Have NEVER Been Contacted To Participate In An Official Election Poll. Not. A. Single. Call.

Came here to post that lol. It's just amazing how stupid some of these people are. Reddit is absolutely terrible.

[Kingpoleon]

I unskewed the polls for their anti-Clinton bias:

482 - 56

[Since I'm the mad scientist proclaimed by myself]

I unskewed the polls for their anti-Clinton bias:

482 - 56

That "unskew" is the dumbest thing I've ever seen. First of all, you have Clinton losing her HOME STATE of Arkansas. Secondly Clinton is a horrible fit for Colorado. She isn't winning it. Lastly I doubt Clinton can win Vermont. Its one of the whitest states in the union and it's mostly rural. It's a perfect fit for Trump.

[GeorgiaModerate]

I'm not sure this is the right thread for this, but someone has posted a fake "leaked" internal PPP poll of Florida at https://www.scribd.com/document/321417522/Ppp-Internal-Raw. It shows...

...wait for it...

74% Donald Trump
9% Hillary Clinton
3% Jill Stein
1% Gary Johnson
6% Undecided

Read the whole thing.  It's hilarious.

[Wells]

I'm not sure this is the right thread for this, but someone has posted a fake "leaked" internal PPP poll of Florida at https://www.scribd.com/document/321417522/Ppp-Internal-Raw. It shows...

...wait for it...

74% Donald Trump
9% Hillary Clinton
3% Jill Stein
1% Gary Johnson
6% Undecided

Read the whole thing.  It's hilarious.

You must be logged in to read this quote.

XXXD

[Badger]

That's stupid, because they should really weigh their results by population. It's not like people in Alaska are going to vote at the same numbers as people in California. And what's their phone bill after 50,000 phone calls?

I have questions.

1. Did they use a random-digit dialer? Live interviewer or IVR, and were their known effects controlled?
2. Did they call cell phones or provide Spanish-language options? Blended-frame?
3. What were the timeframe and calling hours? How many attempts per number?
4. How did they randomize within households?
5. What demographic data was collected?
6. Were questions and answers randomized in any way?
7. Was the questionnaire tested for wording and order effects?
8. Was a likely voter screen applied? How hard were leaners pushed?
9. Was the sample weighted, and how? ACS, voter reg, both? Was aging out/in a factor? Was the process simultaneous, iterative, or both?
10. Do they know how to calculate the margin of error for each proportion? What confidence level was used?

More important question: Are they just lying out their asses?

Both of you are missing a VERY important point. They gave each state the same impact in their poll i.e  the poll acts as if all states are the same size.

Wyoming is not as big as California.

of course Vermont isn't big as Texas either, nor Hawaii as large as Tennessee. but still, that type of smallbstate overbalance, plus a 1/3 even division amog dems, repubs and indies gives a VERY bad shift from reality.

[Badger]

Even without getting into the methodology problems, it's not possible for a group of 13 people to get 50,000 responses in less than 2 weeks. Let's assume an extremely favorable timing of 1 response every 10 minutes (factoring in time it takes to make calls, low response rate, and time to complete a response). That's 500,000 minutes, or 641 hours per person. There's only 336 total hours in 2 weeks.

I posted this earlier to that stupid page:

You must be logged in to read this quote.

Adam: when did you make your post? i just scowered the comments section looking for it to see if anyone responded and didn't see it. Maybe it was deleted for excessive factual analysis?

[Darthpi – Anti-Florida Activist]

I'm not sure this is the right thread for this, but someone has posted a fake "leaked" internal PPP poll of Florida at https://www.scribd.com/document/321417522/Ppp-Internal-Raw. It shows...

...wait for it...

74% Donald Trump
9% Hillary Clinton
3% Jill Stein
1% Gary Johnson
6% Undecided

Read the whole thing.  It's hilarious.

To be fair, it might be a valid result if the 15th and 19th Amendments had never passed, which is probably the America the people who posted that wish they lived in.

[Adam Griffin]

Adam: when did you make your post? i just scowered the comments section looking for it to see if anyone responded and didn't see it. Maybe it was deleted for excessive factual analysis?

August 14 @ 1:00 AM. It's still showing up for me, but it is perfectly possible to hide comments so that only the person who posted it and their friends can see it...at least that's possible on a page. I'm not sure about their specific comments format allowing that, but if so, then that's probably what's going on.

[Desroko]

https://twitter.com/DylanByers/status/766021779361968128

This popped up in another thread, but...just watch it.

[Sorenroy]

I'm not sure this is the right thread for this, but someone has posted a fake "leaked" internal PPP poll of Florida at https://www.scribd.com/document/321417522/Ppp-Internal-Raw. It shows...

Read the whole thing.  It's hilarious.

I find it hard to believe this was done as anything but a joke:

[For Hillary Voters] What most concerns you about Hillary Clinton?

Possible answers include:

Probably Killed Some Guys

Health. She’s super sick!


Also if you look at the crosstabs, none of them add up.

Example:

"Indited by FBI for Murder"

Obama voters are more likely to vote for that candidate 90-3, Romney voters 90-0 less likely, however the overall is 24-73 and, according to the numbers above, Obama won people in the sample 47-42.


I find it hard to think that anyone was trying to pass this off as legit. Far more likely it was just a bunch of people goofing around.