5 ms·

Nate Silver commented (debunked, I would say) this issue on Twitter. It seems that the "abnormality" detected or does not make sense (Pennsylvania, Michigan) or

by cplanas 10y ago

Nate Silver commented (debunked, I would say) this issue on Twitter. It seems that the "abnormality" detected or does not make sense (Pennsylvania, Michigan) or disappears when controlling for race and educations levels (Wisconsin). https://twitter.com/NateSilver538/status/801220813890277376 https://twitter.com/NateSilver538/status/801220813890277376

kordless 10y ago
> debunked, I would say That's actually NOT for you to say. Rationalizing why it's not without doing the analysis yourself doesn't buy you that right, even if you read it somewhere and it sounds good.
- kbenson 10y ago
  I think "I would say" is meant as a stand-in for "in my opinion", which is entirely within their right. It's easy to say you have to go a level deeper to get the truth, but at some level you have to accept the legitimacy of a source. If you don't trust Nate Silver, you can run the analysis yourself. You then have to choose whether to trust the source of the data. If you don't trust that, and you have the ability to to go the pre-aggregated data, you can go to that. Do you trust the people reporting that? At some point,it's not feasible to follow the chain of custody all the way back, so you have to make a call as to what level you are willing to go back before trusting the source. If someone trusts the integrity and competency of Nate Silver, then they should be able to apply some level of assurance to his results based on that. If you don't trust Nate Silver's results because you doubt his integrity or competency, or even if you think it's just too important to trust to a single person, then speak to that, not to some simplistic and often impossible to satisfy ideal of determining everything yourself from pure sources.
  - throwaway5752 10y ago
    On the other side of Nate Silver, you have J. Alex Halderman and Ron Rivest. I respect Nate Silver, but I think everyone can speak to the need to not trust a single person in this case.
    - kbenson 10y ago
      Which would have been an entirely acceptable and valid reply to the original comment, while also providing some additional information (who is disagreeing) in context. :)
    - lets-tryagain 10y ago
      Reputation really isn't relevant here...the election is over, Clinton already delivered a concession without raising criticism herself...the bar here is incredibly high and their evidence seems mostly hypothetical. To satisfy the criticism suggested, you would have to re-run the election...it's just not happening
tnecniv 10y ago
What does he use to make those reports?
- trosi 10y ago
  It looks like Stata: https://en.wikipedia.org/wiki/Stata https://en.wikipedia.org/wiki/Stata
- Terribledactyl 10y ago
  Edit: nope, definitely Stata. Also a tool he's mentioned using. In the past he mentioned using excel with various addons, and these printouts look very, grid like.
thaw13579 10y ago
I don't think Nate's twitter post and associated language are the right way to have a conversation as serious as this one. The people making the claims are not just slinging "BS" to be "debunked", they are experts who have credibility to lose by making claims without evidence, and I think it's worth hearing more from them before writing them off.
gerty 10y ago
Judging by this Twitter feed and Stata output, it looks that we are looking at a classic case of "correlation does not imply causality". Clinton shift might be correlated with paper vs. electronic ballot but it looks like districts that opted for electronic ballot have different population groups compared to the ones with paper ballot. Once you control for that, e.g. race and college degree, paper/electronic ballot correlation disappears. To be fair, data sample is rather small so small effects might be difficult to pick up.
- dillondoyle 10y ago
  Agreed. Many reports claim that e voting machines are in mostly poor voting locations, which could basically explain the 'discrepancy': http://www.reuters.com/article/us-usa-election-machines-idUSKCN11Q0EU http://www.reuters.com/article/us-usa-election-machines-idUS... But controlling for everything is almost impossible to be perfect: designated location vs county wide, income, race, education, Trump/Clinton modeled support. You'd have to pull actual voting returns by individual for those voting locations and states, and then have a good vote file. Then you can start to control for those variables and I would bet almost everything I have that doing so would reduce any irregularity these scholars are using as an argument that the election was somehow hacked.

by cplanas 10y ago

kordless 10y ago

> debunked, I would say That's actually NOT for you to say. Rationalizing why it's not without doing the analysis yourself doesn't buy you that right, even if you read it somewhere and it sounds good.

kbenson 10y ago
I think "I would say" is meant as a stand-in for "in my opinion", which is entirely within their right. It's easy to say you have to go a level deeper to get the truth, but at some level you have to accept the legitimacy of a source. If you don't trust Nate Silver, you can run the analysis yourself. You then have to choose whether to trust the source of the data. If you don't trust that, and you have the ability to to go the pre-aggregated data, you can go to that. Do you trust the people reporting that? At some point,it's not feasible to follow the chain of custody all the way back, so you have to make a call as to what level you are willing to go back before trusting the source. If someone trusts the integrity and competency of Nate Silver, then they should be able to apply some level of assurance to his results based on that. If you don't trust Nate Silver's results because you doubt his integrity or competency, or even if you think it's just too important to trust to a single person, then speak to that, not to some simplistic and often impossible to satisfy ideal of determining everything yourself from pure sources.
- throwaway5752 10y ago
  On the other side of Nate Silver, you have J. Alex Halderman and Ron Rivest. I respect Nate Silver, but I think everyone can speak to the need to not trust a single person in this case.
  - kbenson 10y ago
    Which would have been an entirely acceptable and valid reply to the original comment, while also providing some additional information (who is disagreeing) in context. :)
  - lets-tryagain 10y ago
    Reputation really isn't relevant here...the election is over, Clinton already delivered a concession without raising criticism herself...the bar here is incredibly high and their evidence seems mostly hypothetical. To satisfy the criticism suggested, you would have to re-run the election...it's just not happening

tnecniv 10y ago

What does he use to make those reports?

trosi 10y ago
It looks like Stata: https://en.wikipedia.org/wiki/Stata https://en.wikipedia.org/wiki/Stata
Terribledactyl 10y ago
Edit: nope, definitely Stata. Also a tool he's mentioned using. In the past he mentioned using excel with various addons, and these printouts look very, grid like.

thaw13579 10y ago

I don't think Nate's twitter post and associated language are the right way to have a conversation as serious as this one. The people making the claims are not just slinging "BS" to be "debunked", they are experts who have credibility to lose by making claims without evidence, and I think it's worth hearing more from them before writing them off.

gerty 10y ago

Judging by this Twitter feed and Stata output, it looks that we are looking at a classic case of "correlation does not imply causality". Clinton shift might be correlated with paper vs. electronic ballot but it looks like districts that opted for electronic ballot have different population groups compared to the ones with paper ballot. Once you control for that, e.g. race and college degree, paper/electronic ballot correlation disappears. To be fair, data sample is rather small so small effects might be difficult to pick up.

dillondoyle 10y ago
Agreed. Many reports claim that e voting machines are in mostly poor voting locations, which could basically explain the 'discrepancy': http://www.reuters.com/article/us-usa-election-machines-idUSKCN11Q0EU http://www.reuters.com/article/us-usa-election-machines-idUS... But controlling for everything is almost impossible to be perfect: designated location vs county wide, income, race, education, Trump/Clinton modeled support. You'd have to pull actual voting returns by individual for those voting locations and states, and then have a good vote file. Then you can start to control for those variables and I would bet almost everything I have that doing so would reduce any irregularity these scholars are using as an argument that the election was somehow hacked.