Jump to content

Question about polls


Recommended Posts

Since I was listening from another room, I don't know who said it or the channel...but since my mother-in-law had the remote it was either CNN or MSNBC. Regardless, the woman had a "pundit" from both sides and they were talking about the polls. Naturally the republican pundit was trying to dismiss them, but she said something that the democrat pundit agreed with...to an extent.

 

The contention was that all the polls generally skew liberal. She cited four polls where the stated preference (democrat, republican or independent) showed democrats made up about between 42 and 48% of the respondents...liberals made up about a third...republicans the rest. Now obviously the democrat guest felt that wasn't the reason Clinton had such a large lead but said "she has a point in that it is a factor".

 

That made me wonder, why don't ALL pollsters keep their respondents even? I know the calls are random, but if you know you are going to use 1,000 responses and you are asking about party affiliation, why not use the first 333 responses from each group? Seems to me it would give you a more accurate representation of the general electorate.

 

I know, I am trying to interject logic into politics...just a thought.

Link to comment
Share on other sites

Kellyanne Conway admitted Trump is behind in the polls yesterday in Meet the Press.

 

You cons screamed in 2012 about the polls were skewed to favor liberals.

Obama won with the margin predicted in most polls.

Imagine that.

Do you ever do anything except dodge questions and change the subject?

 

Everyone knows Trump is behind. If they admit it publically is irrelevant and has nothing to do with the question. I said weeks ago I had resigned myself to the fact that Hitlery would win.

 

Now...care to address the question or are you simply going to pull your same old shit?

Link to comment
Share on other sites

What's your question?

Why don't pollsters, all of them, keep their respondents balanced to get a more accurate projection of the electorate as a whole. Having more respondents, from any group, skews a poll. If 40% of respondents identify as republican it is just as skewed as it would be if 40% identify as democrat.

 

Historically, independents and undecided break along the same percentages as "decided" voters within a point or two either way (hence the margin of error). That means if the respondents of a poll are even 36% democrat, 34% independent/undecided and 31% republican the results should breakdown, in a two candidate race, 54% for the democrat and 46% for the republican (36/67)...thus a built-in bias of 8 points.

 

I realize a poll is nothing more than a sophisticated wild-ass guess, but virtually every poll in a POTUS race can tell you the political leanings of those responding. If you know that, it is easy to eliminate any built-in bias.

Link to comment
Share on other sites

Since I was listening from another room, I don't know who said it or the channel...but since my mother-in-law had the remote it was either CNN or MSNBC. Regardless, the woman had a "pundit" from both sides and they were talking about the polls. Naturally the republican pundit was trying to dismiss them, but she said something that the democrat pundit agreed with...to an extent.

 

The contention was that all the polls generally skew liberal. She cited four polls where the stated preference (democrat, republican or independent) showed democrats made up about between 42 and 48% of the respondents...liberals made up about a third...republicans the rest. Now obviously the democrat guest felt that wasn't the reason Clinton had such a large lead but said "she has a point in that it is a factor".

 

That made me wonder, why don't ALL pollsters keep their respondents even? I know the calls are random, but if you know you are going to use 1,000 responses and you are asking about party affiliation, why not use the first 333 responses from each group? Seems to me it would give you a more accurate representation of the general electorate.

 

I know, I am trying to interject logic into politics...just a thought.

the reason they over sample polls is to project the libs ahead in all polls so when they cheat it works out to that factor,,, they have the electronic voting machines rigged to that factor

 

theres the truth and ill be cussed for saying it

Link to comment
Share on other sites

Do you ever do anything except dodge questions and change the subject?

 

Everyone knows Trump is behind. If they admit it publically is irrelevant and has nothing to do with the question. I said weeks ago I had resigned myself to the fact that Hitlery would win.

 

Now...care to address the question or are you simply going to pull your same old shit?

Every polling firm DOES NOT do what you insinuated.

There are over 500 polls in this country.

You got a lot of work to do to prove all of them favor democrats. Hop to it , Sally.

Link to comment
Share on other sites

Every polling firm DOES NOT do what you insinuated.

There are over 500 polls in this country.

You got a lot of work to do to prove all of them favor democrats. Hop to it , Sally.

Hey dicknose...where did I say all of them did that?

 

I asked, why don't ALL of them MAKE SURE they don't skew their results.

Link to comment
Share on other sites

Hey dicknose...where did I say all of them did that?

 

I asked, why don't ALL of them MAKE SURE they don't skew their results.

They're the professionals ..and you're not.

They get it right most of the time the EXACT way they perform their polling.

They did it the same way in 2012 and got it EXACTLY right.

That's what you're afraid of. Lol

 

Your question is unanswerable unless you contact all 500 plus polling outfits and ask them.

Hop to it, deadbeat.

Link to comment
Share on other sites

Actually, when you compare the last poll before election day with the actual results, the final percentages were outside the margin of error on 4 of the most widely cited polls...and one other had the wrong winner.

Moron thinks there were only 4 polls in 2012. Lol

 

Let's see how things are going today for your serial sex offender.

 

http://projects.fivethirtyeight.com/2016-election-forecast/?ex_cid=2016-senate-forecast

Link to comment
Share on other sites

Moron thinks there were only 4 polls in 2012. Lol

 

Let's see how things are going today for your serial sex offender.

 

http://projects.fivethirtyeight.com/2016-election-forecast/?ex_cid=2016-senate-forecast

Moron can't read.

 

Did I say there were only 4 polls? By the way...the answer to that question is "no".

 

I could retype what I posted for you...but I cannot comprehend it for you...you are going to have to manage that on your own.

Link to comment
Share on other sites

Moron can't read.Did I say there were only 4 polls? By the way...the answer to that question is "no".I could retype what I posted for you...but I cannot comprehend it for you...you are going to have to manage that on your own.

Your whole thread is based on a lie. You ASSuME that all 500 plus pollls use the same methods as you described in the OP.

The purpose of this thread is very clear. You think the polls are wrong and your serial sex offender still has a chance.

That's so sweet.

Link to comment
Share on other sites

Your whole thread is based on a lie. You ASSuME that all 500 plus pollls use the same methods as you described in the OP.

The purpose of this thread is very clear. You think the polls are wrong and your serial sex offender still has a chance.

That's so sweet.

Still having trouble with comprehension I see.

 

I never said every poll was wrong. I never put forth an assumption. I could not care less if the current polls are accurate of totally wrong.

 

What generated my question was two pundits, from different political schools of thought, agreeing that there is built in bias, to varying degrees, in the polls being discussed on THAT show.

 

If you don't want to offer (or are incapable of formulating) an opinion as to why pollster don't take the bias out of polls, that is your choice.

 

By the way...nothing was put forth as fact...so there can be no lie.

Link to comment
Share on other sites

Still having trouble with comprehension I see.I never said every poll was wrong. I never put forth an assumption. I could not care less if the current polls are accurate of totally wrong.What generated my question was two pundits, from different political schools of thought, agreeing that there is built in bias, to varying degrees, in the polls being discussed on THAT show.If you don't want to offer (or are incapable of formulating) an opinion as to why pollster don't take the bias out of polls, that is your choice.By the way...nothing was put forth as fact...so there can be no lie.

It's your assumption there is inherent bias in polls.

Most do not have bias.

The end.

Link to comment
Share on other sites

Join the conversation

You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...