Let’s say you’re polling 1000 people to reflect the U.S. population.
Hispanics would be about 15% of that, or 150 people… #oversampling (1)
Oversampling is standard procedure for accurate polling. Before publishing, the sample is weighted back down. http://pewrsr.ch/2erheaZ pic.twitter.com/gjbHd6rhuN
-
-
-
…A 150-person sample is not statistically significant enough to yield an accurate picture, so you'd oversample to 300-350.
#oversampling (2) -
Think of oversampling like asking 10 people in your neighborhood instead of just 1 or 2.
#oversampling (3) -
After polling the oversample, you adjust its size back down to 15% of the overall poll sample, so it reflects U.S. demos.
#oversampling (4) - 8 more replies
New conversation -
-
-
So oversample R's then. They overvoted in primary.
- 1 more reply
New conversation -
-
-
But why would the polls change it so out of control to +8? Democrats will not have turnout of 2008.
- 1 more reply
New conversation -
-
-
Jesus, you people truly don't understand what he is trying to tell u. Not rocket science.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You're going to be jobless if Trump wins. Your focus groups are a scam
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How do you explain the disparity between legacy media (Hillary wins big) and online media polls (Trump wins big)?
-
That's easy to explain: online polls are meaningless.
End of conversation
New conversation -
-
-
look at Hillary vs. Trump in swing state primaries -- Acutals vs. RCP....big difference thoughpic.twitter.com/eYCNEHgyH2
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.