Pants-of-dog wrote:Can you provide a reasonable standard that would make you trust the system?
I don't really distrust the system. But clearly, others do. So how about simply following the same standards in place in places like Canada or Western Europe? Do you trust their elections?
Pants-of-dog wrote:You misunderstood. I will repeat myself more clearly.
There are many studies that show that these restrictions have a negative impact on BIPOC voters.
You have not shown that the many studies are incorrect.
Yes I did. I included you a much more credible one that finds no effect.
Pants-of-dog wrote:And since this is the only support for your argument that these laws are necessary, your argument falls apart.
Maybe, OTOH if you accept that paper than you have to accept that the laws also don't affect electoral results, turnout or registration. Do you accept that finding?
If so, then I don't see the point of fighting those laws either. And again, they can and will be used the next time the GOP tries to question electoral results, which is also important.
Pants-of-dog wrote:No. I have provided several studies supporting my claim.
You provided two of them, both based on surveys and not administrative data. This is a problem because survey nonresponse is not random. Also, one is behind a paywall so it's impossible to assess its quality, the second one is shit and doesn't prove anything.
Quantity is irrelevant if all the papers you use to support your claims are shit.
Pants-of-dog wrote:Your speculation is not an argument.
Unless you can provide evidence for your speculations, this is just a way of rationalising your refusal to accept evidence.
Identification is an old concept in statistics, and indeed the second paper you cited would have been rejected in any decent economics journal.https://en.wikipedia.org/wiki/Parameter ... on_problem
The parameter in the second paper you cited is not identified for the reasons I mentioned earlier: The voter ID laws were passed in Republican-leaning states where Democrats and minorities already have low turnouts compared to Democrat-leaning states. The model used by the authors cannot separate the effects of the laws from those previous trends.
The paper that does in fact allow to control for those previous trends, finds no effects at all.
Pants-of-dog wrote:The only places where electoral results are being called into question is battleground states where Democrats won, and they are only being contested by Republicans.
Here is an obvious vested interest and bias.
For 2020 perhaps. For 2016, the result in general was called into question by Democrats alleging Russian interference.
This issue of questioning election results in the US is not a new thing, at all. And no party has a monopoly on this, indeed, both will whine when they don't get voted.