Posted on 10/17/2020 11:33:07 AM PDT by 11th_VA
Thanks for posting the interview.
The Q guy isn’t just wrong, he is insane.
Real pollsters who want accurate information qualify their respondents in a bunch of ways, and validate their data against _known_ _historical_ _data_.
This is so obvious it should not need to be stated.
An accurate poll must accurately reflect:
—A sample that voted in the last relevant election (which needs to be _verified_ against that election’s actual result, and then _adjusted accordingly). That must be apples to apples, so one Presidential election to another Presidential election, not mixing and matching with Congressional elections.
—Party registration should _not_ determine who you poll—it is just a data point you collect so you know who you _have_ polled.
—What _should_ determine who you poll is matching them to known results, regardless of party affiliation!
So, let us say you interview one hundred Florida voters.
You ask them who they voted for in 2016 and 60% say they voted for Hillary.
Then you would have to _adjust_ your sample to make sure that the number of Hillary voters was equal to her actual Florida percentage in 2016 = 48%.
No more than 48% of your sample can be 2016 Hillary voters.
This is common sense!
I am thankful my post spurred you to make yours, which adds to our knowledge of the sitution as it really stands. Sorry if it upset you.
Everything you wrote should be in our arsenal of facts to use as arguments against the Left. Thank you.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.