4 ms·

What are those "some measures" and where is the 1% coming from?

by 101404 7y ago

joshuamorton 7y ago
"We provided $9.7 million in adjustments to a total of 10,677 Googlers." -> just under $1000 per adjustment, on average. Note that since the comment here is that the affect group (L4 men) was large and led to the majority of the changes, we can generally discount ideas like "oh there were a small number of massive adjustments and a large number of tiny ones", or we can believe that's the case, but then the L4 men adjustments would be smaller on average, not larger. The average L4 SWE at Google is taking home more than $200K annually, and has a salary well over 100K, so these changes amounted to less than half a percent of total comp, and less than one percent of salary. As for "some measures", the short answer is that when you control for performance rating and level, as Google did here, things generally look reasonable. The problem is that there are good reasons to believe that historically equally performing women were less likely to be promoted (for various structural reasons that I won't get into, weren't malicious, and have been at least somewhat addressed). The result then is that high performing women may still be under-leveled or under-compensated in relation to their actual performance. So a completely reasonable interpretation of this sequence of events is that women were as a group, somewhat under-leveled. Their managers, individually, noticed this and did (as Google leadership suggests) what they could to use manager-discretionary compensation to address this underpayment. Again, individually and stochastically. The overall analysis notices this, and because one stochastic process (promotion) previously had a bias against women, and because the recent analysis didn't control for that, I get an extra $1K. Which is nice, but it's not anything particularly special.
- 101404 7y ago
  > equally performing women were less likely to be promoted So they disproved the "wage gap" lie with hard data, and, as if nothing happened, you pull out the next unfounded myth? Got any hard data?
  - joshuamorton 7y ago
    No, they showed that Google doesn't have a gender wage Gap when controlling for performance. This doesn't disprove anything in general. Nor did it address preexisting concerns that aren't myths. And yes, but not publicly.

What are those "some measures" and where is the 1% coming from?

by 101404 7y ago

What are those "some measures" and where is the 1% coming from?

"We provided $9.7 million in adjustments to a total of 10,677 Googlers." -> just under $1000 per adjustment, on average. Note that since the comment here is that the affect group (L4 men) was large and led to the majority of the changes, we can generally discount ideas like "oh there were a small number of massive adjustments and a large number of tiny ones", or we can believe that's the case, but then the L4 men adjustments would be smaller on average, not larger. The average L4 SWE at Google is taking home more than $200K annually, and has a salary well over 100K, so these changes amounted to less than half a percent of total comp, and less than one percent of salary. As for "some measures", the short answer is that when you control for performance rating and level, as Google did here, things generally look reasonable. The problem is that there are good reasons to believe that historically equally performing women were less likely to be promoted (for various structural reasons that I won't get into, weren't malicious, and have been at least somewhat addressed). The result then is that high performing women may still be under-leveled or under-compensated in relation to their actual performance. So a completely reasonable interpretation of this sequence of events is that women were as a group, somewhat under-leveled. Their managers, individually, noticed this and did (as Google leadership suggests) what they could to use manager-discretionary compensation to address this underpayment. Again, individually and stochastically. The overall analysis notices this, and because one stochastic process (promotion) previously had a bias against women, and because the recent analysis didn't control for that, I get an extra $1K. Which is nice, but it's not anything particularly special.

101404 7y ago
> equally performing women were less likely to be promoted So they disproved the "wage gap" lie with hard data, and, as if nothing happened, you pull out the next unfounded myth? Got any hard data?
- joshuamorton 7y ago
  No, they showed that Google doesn't have a gender wage Gap when controlling for performance. This doesn't disprove anything in general. Nor did it address preexisting concerns that aren't myths. And yes, but not publicly.