3 ms·

I'll give one of the study author's some credit for chiming in - following media wind up, mind you - regarding interpretation and pointing out that the deviatio

I'll give one of the study author's some credit for chiming in - following media wind up, mind you - regarding interpretation and pointing out that the deviation is completely reasonable when considering erroneous poll numbers being used for comparison. Which, I think, is a very worthwhile position to take because of, ahem, how generally off the polls were this time around. Those swing state numbers weren't even close to reflective of the majority of polls, IIRC, by Nov. 9.

hackuser 10y ago
Do you have any data on how well the polls did? According to FiveThirtyEight, national polls did pretty well, while state polls had more problems - but he doesn't say how many or how bad the problems were. https://fivethirtyeight.com/features/what-a-difference-2-percentage-points-makes/ https://fivethirtyeight.com/features/what-a-difference-2-per... ... Theoretically, if a scientific survey says one thing and a reported vote count says another, how do you know which is wrong (or are both wrong)?