crackercrews

Gives new meaning to "show hn"

I wouldn't assume this if UC didn't make such a big deal about how they are boosting their Pell and first-gen admits. I think first-gen is now over 40% of UC freshman, which seems disproportionately high.

Time will tell. Let's check back in a few years and see if policies like these have aged well. It didn't take long for the SAT-blind policies to be reversed.

That's one way of looking at it. But it's not just about alumni of UCSD. This is about privileging kids whose parents didn't go to college over kids whose parents went to any college.

More important is the impact. The GPA cutoff is very low, just 3.0. Who gets into UCSD with less than a 3.0? And the vast majority (75%) [1] of UCSD students are in-state. So the only factors that actually matter are Pell and first-gen college. So CS and other majors will be enriched with students who check these boxes. Will they succeed? Some will. But based on existing research we know that students who are admitted with lesser credentials transfer out of STEM majors at a higher rate. This applies to both AA admits and legacy admits. It likely also applies to Pell/first-gen students, whom the UC system prizes and privileges.

I see two possibilities. Either students will wash out in years 2 and 3, or so much of the majors will be made up of these students that there will be pressure to dumb down the courses so that students can pass. Either way, it's a bad outcome for these kids. And it's a bad outcome for the kids who were rejected from the major and who can't shift into it after slots open up later.

1: https://ir.ucsd.edu/stats/publications/21_22_StudentProfiles...

> Yeah, 4.9 drivers explicitly ask for a good rating.

Literally have never experienced this.

Original document is here. [1]

1: https://students.ucsd.edu/academics/advising/majors-minors/s...

How do you know they could have dragged him? If what he said was fair then they wouldn't have been able to land any punches.

Coleman Hughes shares a good perspective on this in his new book. He cites surveys showing that people changed their opinions for the worse regarding racism in america around 2010. This could be explained by an increase in racism or by increased awareness of racism. But he shows it is not based on these things because surveys also show that people significantly overestimate the number of black people killed by police. The rise in media coverage has led them to think that certain events are much more common than they are.

WAMU had layoffs recently.

No. They can't bury a story after it's been in the NYT.

> they shy away from being based if it could potentially alienate the right wing audience.

Funny typo, considering that "based" means something completely different in the political realm. NPR is most certainly not based...

This isn't laudable, it's defensive. The snafu has been covered in the NYT so there's no burying it. They are making sure their side is told by covering it themselves.

Local NPR affiliates also carry shows from other national organizations like PRI. It's not just ATC versus local programming. Many people think such shows are NPR even though they aren't.

Any McKinsey client should insist that the consultants who worked on this report never work on their engagements. These consultants are either too stupid to realize that they got the causation backwards or they realized it but wrote up a garbage report anyway.

> Green and Hand not only were unable to replicate the studies’ findings; they also found that each of the three studies had analyzed the data backward. Instead of looking at a firm’s diversity policies in the years leading up to a given year’s financial performance, McKinsey had reviewed each firm’s financial performance in the four or five years leading up to the year in which its researchers snapshotted their executive demographics. In other words, according to Green and Hand, the positive correlations that McKinsey researchers observed may have reflected “better firm financial performance causing companies to diversify the racial/ethnic composition of their executives, not the reverse.”

> Now, a new paper raises questions about McKinsey’s methodology and suggests that its advertised findings may have gotten the causation backward: financial success may lead corporations to embrace diversity efforts, rather than the other way around.

Completely unsurprising.

> Since McKinsey refused to turn over its numbers, Green and Hand had to reverse-engineer the firm’s 2015, 2018, and 2020 datasets.

So it's still not completely settled who is right. But the fact McKinsey wouldn't turn over their dataset is suspicious. When one researcher works in the open and another does not, conflicting results are likely to favor the open researcher.

Quite some re-framing there. I'm all in favor of viewpoint diversity. But suggesting that this is achieved by hiring based on skin color is not the right way.

Also it is very easy to show how a team is hurt when normal hiring standards (GPA, other numerical prerequisites) are ignored in order to hire someone based on sex or race. I first experienced this when I sat in a classroom and worked on group projects with people who were clearly admitted via AA preference. AA was not functioning as a tiebreaker in close cases. It was used as an enormous boost, and everyone could tell.

> I've been on teams where I'm the only white person. I've also been on teams where women outnumber the men (yes, even in tech).

This isn't surprising at all. There are tons of asians in tech. And it's not uncommon for a couple devs to be on a team with lots of product or design people. I'm on one such team right now. The vast majority of this team is non-white women.

This type of case study could be valuable but it could also just be cherry picking. Did they look at all changes suggested by minority employees? Or did they just look back and reflect on the changes that worked out?

I have seen posts on social media by journalists looking to write about DEI. Instead of asking open-ended questions they were just soliciting success stories. That doesn't lead to the truth. It leads to the preferred narrative.

How did you organization create its case studies? Would it have actually written up a case study showing that the net effect was negative?

Only certain majors would affect tech job hiring. Are men now lagging women in those majors? It's not especially relevant that there are tons more women than men getting degrees in social sciences etc.

The studies that claim benefits from diversity are almost always correlation, not causation. It's easy to see how reverse causation could be at play. Companies that are more profitable can afford to have diversity programs. This fact can be reframed to make it look like the diversity came first and caused the higher profitability.

When DEI studies were first being touted I used to ask to read them. I thought that perhaps I would eventually find one that actually had any rigor behind it. After reading dozens, I stopped asking. It seems that rigor is not required. Just a headline that resonates with key constituencies.

Yesterday I posted a link that was flagged off the front page. I can see that your tool is driving traffic to it and appreciate that.

Not everyone wants to discuss political topics on HN. They say there are other places to discuss such topics. I like to hear the opinions of the HN audience on a wide variety of topics. Maybe this tool will help those of us who value the HN community in this way by facilitating discussions on topics deemed inappropriate for the official front page.

The test and controversy have been heavily covered in national media for years.

I don't know. But it's well known that this test is based on academic achievement and that low-income Asians do very well.

Yes it depends on where you set the threshold. If you try to keep as close to your academic standards as possible, but also admit some kids from the bottom 25% of income then you'll get mostly Asians. If you set a much lower threshold and do a lottery as you propose it depends much more on the composition of the applicant pool.

I'm not taking issue with the outcome described. I'm pointing out that targeting class diversity will not always result in racial diversity as claimed.

Not too tricky to navigate after you've got tenure.

> Targeting class diversity naturally gets you racial diversity anyways.

Not always true. Look at the SHSAT. If you looked at the poor kids and took the top scores, it would be mostly Asians.

The same thing applies for admissions to the Ivies. If they set aside 25% of their incoming class to be from the bottom 25% of the income distribution they'd fill most of those slots with Asians.

Last I checked (not super recently) there was a login wall that popped up after you’ve seen X articles that the authors have chosen to put behind the login wall.

Also, if an author wants an article to be eligible for inclusion in aggregated publications, they have to put it behind the login wall. It’s quite possible my info is outdated; happy to learn what the current state of affairs is if things have changed.

> I began to do freelance consulting in 2019 during my second postdoc, to test the waters outside academia

Testing the waters outside your specialty when you're in your second postdoc is not a bad idea. Presumably that means you've not been able to get a permanent job at least a couple times.

> My blood pressure even dropped five points.

Is a drop this small reliably measurable? I thought variances this small are very common.

The point is that even in a society where everyone has an equal shot at academic success there will still be a noticeable correlation between parental income level and student achievement.

Therefore the fact that a correlation exists doesn't prove that the system is flawed.

NYT commenters point out that people who have extremely high income are more likely to have very smart kids. Even the article admits that academic ability is heritable.

How can we know what is the appropriate amount of disparity between the lowest-income and highest-income? Headlines like this one don't help.