10-30-2017, 01:23 PM
(10-30-2017, 12:19 PM)StLucieBengal Wrote: Numerous reports the past six months or so.
Edit: here is one http://m.washingtontimes.com/news/2017/jun/19/noncitizen-illegal-vote-number-higher-than-estimat/
If you are referring to that, then I have to disappoint you. The methodology used in the ODU findings, which this "independent think tank" used for their evidence was flawed. Extrapolating the numbers that they did from the data was just bad statistical analysis. On top of that, this is information about prior elections. His claim was that is happened during his election. Using that information is an attempt to move the goal posts, even if it wasn't deeply flawed.
(10-30-2017, 12:29 PM)GMDino Wrote: I get the feeling that Matt might want to see an actual study and methodology not a report on a study where the people behind it say "it's all very technical" and don't provide anything except their "findings".
I already knew about that one, at least. It made the rounds with political stat nerds a while back, and was widely disputed based upon flawed methodology to reach some of the conclusions. They neglected to account for response errors, and to adjust the margins based upon the small size of the sample of the population, especially when looking at non-citizen respondents. When you take those issues with methodology into account and combine it with actual data on non-citizen voting, you can tell that the calculation they came up with was laughable, to say the least.
