just to compile the discussion in one thread, it becomes a hassle having to go through each thread individually
just to compile the discussion in one thread, it becomes a hassle having to go through each thread individually
You do this by making yet another one?
Ę̵̚x̸͎̾i̴͚̽s̵̻͐t̷͐ͅe̷̯͠n̴̤̚t̵̻̅i̵͉̿a̴̮͊l̵͍̂ ̴̹̕D̵̤̀e̸͓͂t̵̢͂e̴͕̓c̸̗̄t̴̗̿ï̶̪v̷̲̍é̵͔
Evidently the election has violated Benford's Law.
To put it simply, Benford's Law is a curve in which some frequency of numbers occur over. When a data set is contrained by the curve we say the data set follows benfords law which means it falls into the distribution.
Here's a more detailed explanation for those interested: http://assets.press.princeton.edu/chapters/s10527.pdf
Benford's Law has been used as means to detect fraud across a number of processes, one being voter fraud. The idea is that a unmanipulated voting process should obey benfords law, and when numbers are manipulated the manipulators will give preference to certain numbers that then violate the distribution.
If you're interested in how Benford's law is related to voting, here's two papers I found useful:
Examples,
I'm too ignorant to say if it's even appropriate to assume an election process should follow Benford's Law but nonetheless I think the idea is interesting and worth looking into, plus it's relevant the current discussion.
I'm too ignorant to say if it's even appropriate to assume an election process should follow Benford's Law but nonetheless I think the idea is interesting and worth looking into, plus it's relevant the current discussion.
After Monkey/LiYang's post on Benford's law, I also took a quick look. However, I similarly do not understand why elections should follow Benford's law. It seems like there's some empirical grounding.