Hacker News Reader

Comment by 0xy | Hacker News Reader

>From the start, she alleges, she was "de-leveled"

Oh, so she was treated like literally everyone applying there? This is not uncommon.

>During the 2017-2019 time frame, no Black employees were hired at or promoted to the director level ("L8") within Amazon's public policy division, the suit alleges

This statement is meaningless without knowing how many black employees were eligible or applied.

>While working for Amazon, Newman alleges that she faced multiple instances of colleagues overtly engaging in racial stereotypes to discuss her or her work, describing her as "just scary," "too direct," or looking "like a gorilla."

Only one of those is a racial stereotype, and a single incident does not imply systemic racism.

This looks exaggerated at best to me. Everyone experiences downleveling regardless of race.

View on HN

> This statement is meaningless without knowing how many black employees were eligible or applied.

I would like to know the denominator as well. It’s hard to really mean anything without the number of L8s in total, how many L8s were hired, etc

It’s actually odd that the press release doesn’t include this info. I expect that it’s hard to get and they have to sue to get it, or the numbers aren’t in their favor so weren’t reported.

Race is really hard to “eyeball” since it’s not necessarily clear. For example, I was working on a team studying this and we surveyed a group of a few hundred people and asked what a fair distribution by race should be. It was interesting that many people overestimated how much their race should be represented, overestimated how much of the population their race makes up, and underestimated their race within coworkers.

It kind of makes sense as not many people know even basic demographic data. It made communicating with data hard because when sharing that 25% of positions were held by African Americans and 20% of leadership, the most common feedback was that these percentages needed to go up to match population distribution. Which is odd because AA is like 12-15% of population.

So even with these numerator and denominator numbers it’s hard to understand what’s ok. Hopefully the lawsuit will result in an investigation and if there is discrimination, to have it corrected.

Man I'd love to see the data from the outcome of that study you mention. That seems really interesting. Reminds me of that anecdote about something like 80% of engineers consider themselves to be exceptional compared to their peers.

This was our internal data and internal questionnaires in a group of about 1000 with 200 responses.

I wish there was more research in this area. I’ve found a hard time with HR type data as it’s all either too general or so specific it’s never published or shared.

I’ve wondered about programmer efficiency for decades now and wish there was some objective way to measure. I’ve never been part of a programmer labor lawsuit but I wonder how that would go as it’s really common for people with similar titles to make very different pay and it’s usually based on completely subjective measures of impact. But my basic understanding of this stuff from business school is that the onus is on the employee to prove a pattern and just “my boss subjectively pays me less” is perfectly legal.

The cases I know of are ones like WalMart’s sex discrimination case [0] that was eventually won (or lost if you’re Walmart) based on the ease of measuring gender (compared to race) and the size of the organization.

I think part of the issue is measuring race and ethnicity is hard. Many people’s race and ethnicity aren’t necessarily widely known so informal methods of looking at the company prof pics isn’t definitive. It’s also hard because race is self-selecting so if someone responds to a survey there’s not really anything that can be done to qualify responses unless they are specifically claiming tribal nation status.

[0] https://en.wikipedia.org/wiki/Wal-Mart_Stores,_Inc._v._Dukes

This is really the crux I find in these situations.

A pure statistic doesn't tell the whole story. But for pointing it out you usually get labelled a racist.

If there are less people of color applying or objectively eligible for a job the statistics will "paint a damning picture". But maybe the picture isn't damning at all and if the same people had been white they also wouldn't have been chosen. The smaller the numbers involved the less we can just assume that because of the sample size something like this has 'worked itself out'. It's like looking at my coding statistics for the last week. That says nothing as there's so much variability week to week depending on circumstances. But looking at a full year any outliers should have been reduced enough to be able to make decisions based on the numbers.

That said looking at the sibling post as well, there seem to be quite a few positions open in that department. So now one could probably start deducing.

EDIT: and there comes the down voting for 'being a racist' for pointing out that statistics are statistics and that it looks like statistics might actual favor the article protagonists odds. Awesome!

> Only one of those is a racial stereotype

Which one?

Reply on news.ycombinator.com