= 27d567bc2d48d9f40e9ccc20ea370b15)
Jayson_Virissimo comments on LW systemic bias: US centrism - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (58)
Presuming the US when no country is named is statistical discrimination (not a bias).
Most Less Wrong users are from the US.
X is a Less Wrong user.
Probably, X is from the US.
What, if anything, is biased about this pattern of reasoning?
Wouldn't it be more likely that, since the majority of LessWrong users are from the US, most posts are US-centric because that's what the poster himself is familiar with?
I mean, certainly we could pose a line of reasoning to create a post hoc justification for the practice, but what seems more likely is that US-centric posts are reflective of poster, not audience, knowledge. Unless you think we'd have reason to suppose that posters would readily be plucking examples, etc. from their immense knowledge of the UK or Nepal.
Presuming a poster is male when no gender is given is statistical discrimination (not a bias).
Most Less Wrong users are male.
X is a Less Wrong user.
Probably, X is male.
What, if anything, is biased about this pattern of reasoning?
Correct, that is another instance of the same reasoning pattern with high inductive probability. I see no evidence of cognitive bias in either case.
My point is that the fact that the probabilistic inference is valid does not imply that you should e.g. use examples that assume the user is that way, which was the reason you were making that point to begin with. I can safely assume that an unknown user is male. Doesn't mean I should use male-experience specific examples for elucidation.
See here (minus the part about the robe). My response was about whether a certain presumption was biased, not whether it was an optimal social norm.
(Picture me saying this in dramatic tones, standing on a podium wearing robes and frequently howling "Fools!")
For a perfect Bayesian, it works. For humans, not so much. Just having a category exist makes us develop silly beliefs around it. If they're categories of people, we start loving our category and hating others - the ingroup/outgroup dichotomy. We treat ourselves as default and other, er, others, increasing the status differential. If a power structure already exists on top on that, forget it. It's really not innocent.
It seems to me that whether or not something is good social practice is distinct from whether or not it involves cognitive bias. BTW, I like the robe; it is everything I imagined it would be.
Upvoted for the flavor text and the anvilicious necessity.
The point is that even though the majority of the audience is American, it (often) still isn't optimal to use US-centric terms and ideas.
Agreed, but I was responding specifically to the assertion of bias I quoted in my comment, not the underlying point of the post.