There is so much wrong with this post, but first I will talk about what I think is your main point, which may actually be correct. To stick with your 1st example, what I think you are saying (sorry if I have your point wrong) is that we have insufficient data to allow statistics to give a confident answer to if the collecting and submitting of absentee or mail-in voter ballots by volunteers or workers causes more people who should vote to go from not voting to voting than people who should not vote to go from not voting to voting. If you were to display the slightest sign of actually having looked at the data and assessed the question quantitatively then I would certainly accept your word, as I am not going to look that deeply. If someone was to I would certainly read their conclusions. From a very quick look, the search 'statistics "ballot harvesting" non-citizens' brings up no peer reviewed literature with replicable statistics demonstrated. This supports the point I think you are making.My point here was that there are logical problems with the statistics that you, and anyone relying on a statistic to substantiate a predetermined viewpoint, simply overlook. Your racially-linked study and your wiki edit war invite a couple of safer conclusions:
But on the general subject of whether we can gather good data about voter fraud, not even the higher-quality source has anything to say. In fact if you follow the link you posted, you will see other studies in the roster commenting on the hurdles I mentioned: the extreme difficulty of gathering voter ID info and actual data about fraud in the US. I think some of these difficulties have been created on purpose. You make these stupid pretenses of empiricism. One need look no further than the census questionnaire, and the spirit in which, "are you a US citizen," was struck. There's the real level of empiricism.
- Production of identity politics viewpoints is amply-financed by the university network.
- Wikipedia is not.
Now to some other points in your post:
- I am well aware of the problems inherent in this sort of statistical modelling, and am generally pretty critical of statistical modelling, including my own.
- What racially linked study? I choose the first peer reviewed article that had real statistics, that was about the effect of rain on the result. I thought that would be the least controversial thing I could think of. In the UK the weather is famously the safest subject to bring up in small talk, as it is inherently non-controversial (we are famous for talking about it for this reason).
- I had nothing to do with that "wiki edit war". I linked to that page because I had to look up what ballot harvesting was, and wanted to ensure that we are on the same page about what we are talking about.
- I have never seen a "wiki edit war", does one inappropriate comment and its removal in a month count as a war? I suspect not.
- My university was so not productive of identity politics viewpoints, there were more young earth creationists than openly gay and black people combined.
- I have not expressed a predetermined viewpoint. The only viewpoint I have expressed is that there are statistics about the factors that affect voter turnout, and I did that after looking to see that there were.
- You accepting that there are "other studies in the roster commenting on the hurdles [of statistically analyzing voter turnout]" shows that there are statistics on voter turnout. That is what statistics are.
- You accepting that the census could inform this question indicates that not collecting citizenship information at the point of voting does not preclude building models around this question, just that you have to look around a bit more for the data for your models.
Last edited: