Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: haircutter

Good grief. What are you drinking? I’m 1000% for Trump-Pence.

But this is a meaningless poll. You just can’t poll a random selection of 1000 people from each of the 50 states and use that data to predict an election.

Most polls are equally bad on the Hillary end. But not all. The USC poll which has Trump and Hillary basically tied looks pretty legit. Corrected for a known bias has Trump up by about a point.

But this poll tells us nothing. Sorry about that. I wish it was true.


23 posted on 08/05/2016 8:47:46 PM PDT by InterceptPoint (Ted, you should have endorsed. Big mistake.)
[ Post Reply | Private Reply | To 21 | View Replies ]


To: InterceptPoint

drink a root beer...old fashioned one..

to each of us our opinion..

stop trying to be so perfect..

Sure hope your telling the truth that your 1000% TRUMP~PENCE living in California, must be difficult for you.


25 posted on 08/05/2016 8:54:28 PM PDT by haircutter
[ Post Reply | Private Reply | To 23 | View Replies ]

To: InterceptPoint
I will agree there are several ways this poll methodology skews. For one thing it over samples less populated states. They could improve by making the calls proportional to the states population (or electoral college size). For another the voter rolls they used might not be complete...or maybe they are, not sure. The other skewing would be who was able and willing to answer the phone of course.

Some of the sting of these sins is made up by making sure they had 1 third of each of R, D, and I, but that has its own problems too.

But even with all these problems, the discrepancy in the results is significant. Unless they are making this up (which I am suspecting they might have, or they might have fudged data, 50,000 calls is a lot!), it has interesting implications as to the methodology of the professional polls, and they may want to do some sanity checking.

28 posted on 08/05/2016 9:05:03 PM PDT by AndyTheBear
[ Post Reply | Private Reply | To 23 | View Replies ]

To: InterceptPoint
"...you just can’t poll a random selection of 1000 people...

Ooops! You must have missed the part where they explain that they selected 34% Dem, 33% Rep, 33% Independent (or some voter-similar specific breakdown). Hardly a "random selection" of just any 1000 people.

Why do you feel the need to misrepresent the issue?

36 posted on 08/05/2016 9:53:54 PM PDT by Gargantua ("President Trump... nice ring to it..." ;^)
[ Post Reply | Private Reply | To 23 | View Replies ]

To: InterceptPoint

The thing that leaps out at me re this poll is that some 35 40 percent of the people who voted WILL vote Dem no matter what. If Trump wins by 6 I’d consider that pretty big.


47 posted on 08/06/2016 4:14:12 AM PDT by TalBlack (Evil doesn't have a day job....)
[ Post Reply | Private Reply | To 23 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson