Posted on 10/25/2016 10:46:41 PM PDT by Helicondelta
Gravis Marketing, a nonpartisan research firm, conducted a random survey of 2,109 registered voters across the United States. The poll was conducted from October 20th through the 23rd and has a margin of error of ±2.1% at the 95% confidence level. The total may not equal to 100% because of rounding. The polls were conducted using automated telephone calls and weighted by anticipated voting demographics.
(Excerpt) Read more at oann.com ...
Why weight polls at all?
So, if this poll oversampled Democrats by 7%; that’s a bit in the same ballpark as that one that oversamples Dems by 9%; Pat Caddel said he would not heed such a poll;
So, I don’t get the science totally but this poll would be good for Trump and would probably mean he’s leading.
Registered voters? At this late date?
Registered voters?
Folks, I think there's a tsunami coming.
The Revolution is ON!
Vote Trump!
Theoretically you can get more accurate results by having the sample reflect the overall population as closely as possible.
Because it is difficult to reach via RDD respondents that are actually representative of the voting population. If you simply polled people without setting demographic quotas and without limiting certain quota cells, you would end up with a poll that told you how older women, mostly white, in non urban areas would vote.
There are more Democrats then Republicans especially in Presidential years. Don’t assume it’s oversampling.
It’s not a 2 person race. I think the beast loses more voters to 3rd party candidates than Trump does.
Yeah, but stupid females like Megyn Kelly don’t understand the weighted sampling and are herded along by the fixed polls.
We are screwed...
Anyone who makes money continuously....is screwed.
I get it....I take it for granted...and plan for it.
“Why weight polls at all?”
CNN reported the other day that Republicans were refusing to participate in their polls.
The result was a poll with a huge democrat oversample showing Hillary ahead. Useful poll? Probably not.
I think it’s F+6, swear I saw 53/47 in the title earlier...
So many people for DJT are telling pollsters to F OFF. Rush played the NbC coverage from 1980. They were in shock
Yep, it’s F+6
This poll has some problems.
While the White Non Hispanic population in the US is 64%, 67% of the Respondents in this poll were White Non Hispanic.
However, Jews and Muslims are over represents in the poll.
Seriesly! The reason is twofold 1. To depress R voters 2. To account for expected fraudulent votes cast - a CYA move.
Mrs Clinton will top out at 42% Any poll we see with her above 42% is a Megan Kelly handpicked troll poll. Trump Movement needs enough votes to cover the illegal votes that the political / media establishment is going to cast.
A pollster can oversample on age, race, gender and income as each of these typically stays the same throughout a person’s life and can be pegged to census data.
However, political party affiliation changes from year to to year and even from day to day. With that, political affiliation cannot in itself be a quota cell and cannot be an instance of over sampling.
“Why weight polls at all?”
If done properly it improves their reliability. The big if there is “IF DONE PROPERLY”
If the pollster makes a mistake the weighting can give really weird results.
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.