Fisher's exact probability test
[Fisher's exact probability test - Wikipedia https://ja.wikipedia.org/wiki/%E3%83%95%E3%82%A3%E3%83%83%E3%82%B7%E3%83%A3%E3%83%BC%E3%81%AE%E6%AD%A3%E7% A2%BA%E7%A2%BA%E7%8E%87%E6%A4%9C%E5%AE%9A] Fisher's exact probability test
[Fisher
Polis goes halfway through the process of testing for Fisher's exact probability that the probability of a vote v for opinion (comment) c is significantly different between groups g and g, and uses the higher p-value as the degree to which a vote v for opinion c characterizes group g.
This usage is not appropriate as a test, but I think it is useful as a measure. What do you think?
The p-value of a statistical inspection indicates "whether it could happen by chance" and is originally intended to determine "significant or non-significant" by comparison with a pre-determined significance level. However, the p-value obtained in the process may reflect to some extent the strength or weakness of the association between voting for comments and the group. Although this diminishes its validity as a test, it may be practically useful to use it as an indicator for relative comparison of "the degree to which a group is characterized".
This page is auto-translated from /nishio/Fisherの正確確率検定 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.