Bias indicates that we *overestimated* the Republican by X (fivethirtyeight house effect, interestingly, seemed to assess that we were overestimating Dems)
Show this thread
Upshot/Siena results (preliminary) Finished 11/4: avg error 1.9 pts, bias R+0.4 Final 30 polls: 2.9 pts, R+0.5 "Final wave": 3.4 pts, R+0.7 All polls: 4.2 points, R+0.9
-
-
-
Our polling shows pretty clear evidence of a Kavanaugh effect, now that there are real results for comparison. Also a rapid reduction in error as the election approached, which makes sense (both bias/error are by margin)pic.twitter.com/Rp5N2F79ny
Show this thread -
One interesting thing is that our polls rapidly became more accurate over the final couple of weeks, which didn't really materialize for other polls (which were still reasonably accurate)pic.twitter.com/TyfNugLucQ
Show this thread
End of conversation
New conversation -
-
-
What do you think happened in TX-23 - feels like everyone was expecting a blowout that didn't materialize
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Nice work you guys.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Does the bias mean the polls were more bullish on republicans by that margin compared to the actual results?
- Show replies
New conversation -
-
-
Good job and much appreciated!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I thought there was an R bias
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You guys did amazing
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.