General Discussion
In reply to the discussion: I need some debunking help, arguing against math-based claims of voter fraud in PA [View all]Pobeka
(4,999 posts)1) What is this birthday score metric, and is it being appropriately used in this situation. Often someone wanting a certain result may cherry pick through different metrics in order to find the one that give the answer they are looking for, rather than the correct metric for the analysis.
2) Can someone independently reproduce that graph, using data collected directly from the state of PA? (verify the result)
If there is a problem with either of those, the result is likely bogus.
On edit, I read the first few paragraphs at that link, and here is the *first* problem (I think), from the link:
"We construct a new metric of potential voter fraud using suspicious distributions of birthdays in Pennsylvania voter registration data. The basic idea is that people picking fake birthdays will make predictable non-random choices, like picking round numbers for days of the month, and not knowing what true birth month distributions look like."
I'm going to guess this new metric is ***bogus***. i.e. not vetted, supported by a peer reviewed community of statisticians.