And this. I've no idea what any of it means but it seems I have stumbled into an argument which makes statisticians very angry & I'm on the side of the ones who aren't rude to me, speak plainly & accept I can talk abt gender issues sans statistical models. https://twitter.com/prshearer/status/1001152411417874433 …
-
This Tweet is unavailable.
-
I think it's not necessary to be knowledgeable in statistics to understand their point about this. It's basically, if we're saying the earnings gap between men and women is 0 when we look at men and women in the same jobs, hours, etc, that doesn't mean there's no discrimination
2 replies 1 retweet 9 likes -
Replying to @salonium @HPluckrose and
because men and women might be choosing different jobs and hours as a result of being forced into them/discriminated against.
2 replies 1 retweet 7 likes -
Replying to @salonium @HPluckrose and
Respectfully, men and women may be choosing different jobs and hours for reasons that have nothing to do with force or discrimination.
1 reply 0 retweets 10 likes -
Replying to @mjaeckel @HPluckrose and
Yes, of course. I'm saying that even if the earnings gap is 0 after controlling for those differences, that doesn't *necessarily* mean there's no discrimination. There might be, or maybe there isn't.
3 replies 0 retweets 13 likes -
Replying to @salonium @HPluckrose and
And my point is that if the earnings gap is in fact 0 after controlling for differences, discrimination or lack thereof clearly do not seem to have a statistically significant impact on an alleged wage gap.
3 replies 0 retweets 8 likes -
Replying to @mjaeckel @HPluckrose and
No that's not right. The point is that discrimination and choice can both influence the differences that are controlled for, so just controlling for them doesn't tell us much either way about how much of it is choice-driven vs discrimination-driven.
5 replies 0 retweets 13 likes -
Replying to @salonium @HPluckrose and
Either they have a measurable impact on an earnings gap or they don’t. If the gap is 0 after controlling for differences, we’d only be speculating as to whether discrimination plays a significant role.
1 reply 0 retweets 1 like -
Replying to @mjaeckel @HPluckrose and
I think you're missing the point here - discrimination could have a measurable impact on the earnings gap through occupational choice, hours, etc. and other things that are being controlled for. But, yes, if you controlled for it and the gap was 0, you would only be speculating,
2 replies 0 retweets 5 likes -
Replying to @salonium @HPluckrose and
With all due respect, I doubt that I’m the one missing the point. I think that the people who find it “illogical” that there doesn’t seem to be a wage gap when we control for differences are those who are not willing to accept that discrimination may not be that significant.
2 replies 0 retweets 5 likes
It could show that and probably does but starting with that assumption means you are looking at statistics to find evidence of discrimination but starting with the premise that there is no discrimination which could account for the difference further up the chain than gender.
-
-
Replying to @HPluckrose @mjaeckel and
It doesn't work logically from a mathematical point of view even tho it is a perfectly reasonable & highly explanation for the disappearance of difference when accounting for occupational differences. I conceded this when explained to me but it just means statistics don't help us
1 reply 0 retweets 1 like -
Replying to @HPluckrose @mjaeckel and
no it does not! that paper I linked to outlines a potential approach on the problem! I also linked you to Sarsons' papers which outline a different approach! there is a lot of work on this and much more to be done too. stats is not useless here.
0 replies 0 retweets 1 like
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.