1) I'm pretty sure it would be possible for a study to show that females make more than 100% of male salaries, as adjusted. i.e., high school grads make 50% of college grads - if females were predominately HS grads and males were predominately college grads and females made more than the predicted ... (let's round down to) 50%, then that might be above 100% of expected wages.
2 & 3) I'm not sure what other proof could be produced. There are real-life studies that show the total effect, lab studies that demonstrate that some bias exists at all stages of the hire/negotiation/promotion life cycle, and a real-life catalog of complaints of bias. I'm inclined to believe that a corporation that is willing to pay $29MM to settle a claim of gender bias probably doesn't have a great defense. [Granted, there are other factors involved - corporations never want to go to jury trial since juries root for the underdog. But to assign a weighting of zero to the successful complaints...? That doesn't seem right.]
6) There's some work on this. Example: Dweck's articles on mindset. Abbreviating, girls and boys perform on par at math until 5th grade, then girls fall off (which she attributes to gender roles - not without some justification - i.e., there have been studies that show how teachers react differently to male/female students). She found that girls had a tougher time being called smart or clever - i.e., when they struggled with math (as everyone does eventually at some point), the girls figured they weren't smart at math and accordingly performance fell. In response, they demonstrated an intervention where they showed the girls how one can become better with practice; the group who received that intervention tracked with the male math scores through HS.
I'm mostly self-interested - by the time my 4-year old daughter is making career choices, I hope she doesn't have to choose between friendly/nonlucrative and unfriendly/lucrative (wild simplification of available choices).
7) The public is starting to hold the local tech firms (I'm in SF) accountable for having a somewhat homogeneous work force. This is requiring that they go upstream to encourage talent development - in total, the firms can't hire more than 20% females if only 80% of grads are male.
In other words, each little piece of the puzzle needs to be researched for cause, hypotheses tested, then effective hypotheses implemented over time.
I certainly hope that there are also folks researching why 10% fewer male HS grads go to college (61% vs. 71%) and are more likely to drop out (6-year graduation rate of 56% for males vs. 62% for females) and working on a fix. Counter-intuitively, females may benefit from inequality - i.e., females w/o college degree have worse wage prospects than male counterparts, so are incentivized to enroll and graduate in higher proportion.
@TerrorKingA - I liked the essay. Was the rumor of toy discontinuance ever substantiated?
"In other words, each little piece of the puzzle needs to be researched for cause, hypotheses tested, then effective hypotheses implemented over time."
While a beautiful thought, this is exactly what not to do. Issues include but are not limited to:
1) Inflated type 1 error rate
2) Introduction of confounders, distorters, and missed mediators and moderators.
3) Reifying p-values over effect sizes.
When studying a complex model such as A-> B -> C -> D -> E it is not enough to simply measure a->b, b->c, c->d, and d->e in isolation and if they are all significant call it a day. The entire model has to be run all at once. Among other issues, this is probably one of the biggest ones.
Was the rumor of toy discontinuance ever substantiated?
I haven't looked into it since I heard it on Kevin Smith's podcast. Despite Star Wars being my second favorite franchise, I've been sort of burned out on it ever since Disney bought it.
This doesn't have anything to do with the direct topic on hand, but I think it's a good look into the general phenomena being questioned here- Specifically "Do you think there is a tendency for groups to shut down dialogue?"
I am no big fan of people who claim that "scientific/rational/objective" thinking frees them from biased thinking, and this article give good examples of why I believe this.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
1) I'm pretty sure it would be possible for a study to show that females make more than 100% of male salaries, as adjusted. i.e., high school grads make 50% of college grads - if females were predominately HS grads and males were predominately college grads and females made more than the predicted ... (let's round down to) 50%, then that might be above 100% of expected wages.
2 & 3) I'm not sure what other proof could be produced. There are real-life studies that show the total effect, lab studies that demonstrate that some bias exists at all stages of the hire/negotiation/promotion life cycle, and a real-life catalog of complaints of bias. I'm inclined to believe that a corporation that is willing to pay $29MM to settle a claim of gender bias probably doesn't have a great defense. [Granted, there are other factors involved - corporations never want to go to jury trial since juries root for the underdog. But to assign a weighting of zero to the successful complaints...? That doesn't seem right.]
6) There's some work on this. Example: Dweck's articles on mindset. Abbreviating, girls and boys perform on par at math until 5th grade, then girls fall off (which she attributes to gender roles - not without some justification - i.e., there have been studies that show how teachers react differently to male/female students). She found that girls had a tougher time being called smart or clever - i.e., when they struggled with math (as everyone does eventually at some point), the girls figured they weren't smart at math and accordingly performance fell. In response, they demonstrated an intervention where they showed the girls how one can become better with practice; the group who received that intervention tracked with the male math scores through HS.
I'm mostly self-interested - by the time my 4-year old daughter is making career choices, I hope she doesn't have to choose between friendly/nonlucrative and unfriendly/lucrative (wild simplification of available choices).
7) The public is starting to hold the local tech firms (I'm in SF) accountable for having a somewhat homogeneous work force. This is requiring that they go upstream to encourage talent development - in total, the firms can't hire more than 20% females if only 80% of grads are male.
In other words, each little piece of the puzzle needs to be researched for cause, hypotheses tested, then effective hypotheses implemented over time.
I certainly hope that there are also folks researching why 10% fewer male HS grads go to college (61% vs. 71%) and are more likely to drop out (6-year graduation rate of 56% for males vs. 62% for females) and working on a fix. Counter-intuitively, females may benefit from inequality - i.e., females w/o college degree have worse wage prospects than male counterparts, so are incentivized to enroll and graduate in higher proportion.
@TerrorKingA - I liked the essay. Was the rumor of toy discontinuance ever substantiated?
While a beautiful thought, this is exactly what not to do. Issues include but are not limited to:
1) Inflated type 1 error rate
2) Introduction of confounders, distorters, and missed mediators and moderators.
3) Reifying p-values over effect sizes.
When studying a complex model such as A-> B -> C -> D -> E it is not enough to simply measure a->b, b->c, c->d, and d->e in isolation and if they are all significant call it a day. The entire model has to be run all at once. Among other issues, this is probably one of the biggest ones.
I haven't looked into it since I heard it on Kevin Smith's podcast. Despite Star Wars being my second favorite franchise, I've been sort of burned out on it ever since Disney bought it.
Your mods are terrified of me.
Do they make more money than you?
w/r/t Slave Leia - y'all were probably aware of this, but there's a wide range of toys.
http://www.amazon.com/Star-Wars-Princess-Outfit-Figure/dp/B00CFELUZS (creepy!)
http://www.amazon.com/Princess-Outfit-Figure-Vintage-Collection/dp/B0085XPF5O
http://www.amazon.com/Kotobukiya-Princess-Leia-Slave-Return/dp/B000FQ9X6G
http://www.amazon.com/Star-Wars-Princess-Jabbas-Deluxe/dp/B000IO6P2K (I don't recall this pose in the movie)
http://www.vox.com/science-and-health/2015/12/15/10219330/elite-scientists-hold-back-progress
I am no big fan of people who claim that "scientific/rational/objective" thinking frees them from biased thinking, and this article give good examples of why I believe this.