recovering_rationaleist's profile

recovering_rationaleist 4mo ago

I'm currently adjacent to an R1 research university, and here's what's been happening:

After the administration cancelled grants which contained diversity-speak, faculty who were supporting DEI due to social pressure have quietly stopped promoting it, while the true believers have started communicating by FOIA-untraceable means. Outreach and education programs which had been set up to give special scientific mentoring opportunities to black students have suddenly dried up due to a lack of faculty support; the minority of faculty who still want to support these programs are still running them, but at much reduced headcount. The programs are only discussed in meetings and not by email.
Everyone is concerned about funding. NIH restructuring earlier this year will result in at least one full unfunded "gap year" between grants, if not more. Furthermore, the federal government has not been paying out on its existing grant obligations, so the institution has been covering staff salaries out of its endowment. This will continue during the government shutdown, but at some point the institution may need to dip into its investments to pay salaries (instead of funding salaries off the interest on those investments), and at that point layoffs are on the table.
The foci of scientific research are shifting with the political winds. A faculty member who last year gave a talk about "the ethics of whole-genome studies on minority groups", which heavily implied that Native Americans are due some special degree of genetic privacy, this year talked about an actual study design and how they support study participants for long-term follow-up. A staff member whose poster last year was about diversity in science gave a poster this year about the opioid epidemic.
"Diversity, equity, and inclusion" as a term has disappeared. Its institutional replacement is "inclusivity".
Mandatory training was nice this year; it focused on harrassment and publication ethics, and HR lady removed the snide remarks about "white men".
Racial and gender interest groups still exist, but they are no longer advertising their meetings in traceable ways. The departmental "women's cafe" is now advertised on posters in the corridors, rather than by (more FOIA-able) email and intranet.
Humanities students remain absolutely woke-brainwashed; during class discussions, the younger and humanities students will try to shoehorn any and all concepts into DEI jargon, or derail the discussion to be about minorities. A discussion about "what makes teaching effective" ends in "Student learning depends on their identity and positionality."
In contrast, older students (whose day jobs involve work in radiology or physical therapy) and engineers are more likely to remove woke jargon and project woke claims into a concept-space where things actually make sense and testable claims can actually be made.
I'm editing an effort-post on this, but the actual scholarship for the DEI position is incredibly intellectually weak. Every intervention on behalf of diversity is claimed to have huge positive effects in multiple dimensions; the actual citations go to junk studies (small N, self-reported results, no control group) which only support a fraction of the broad claims made. I think this is due to the same effect which results in woke movies sucking: the focus on diversity is a shield against criticism which would otherwise improve the scholarship. There is also a survivorship bias for early-career faculty whose research supports woke positions.

96

Context

recovering_rationaleist 7mo ago

It seems like "e-verify" is just filing a simplified form I-9 (which has a field for social security number/A-number/passport number) online, and then getting a result saying that "records match" from the SSA or DHS. On Form I-9, the employee attests under penalty of perjury that they have the right to work until some date, and provides IDs to back up that claim. The employer attests under penalty of perjury that they have verified the worker's official documents which establish identity and employment authorization. This is already required for all employers. I imagine a ton of employers are not actually verifying documents and perjuring themselves, but nobody cares, because it is employers who sponsor political candidates.

According to USCIS, E-verify is different from Form I-9 in that it requires a social security number and photo identity documents, and tells the employer whether the employee is eligible to work within three to five seconds. Watching one of their tutorials for employers, the employer fills out "Name", "Date of Birth", "Social Security Number", "Employee's email address", "Citizenship status", and which documents the employee provided. (With the exception of the email address, this all duplicates information on the I-9). The employer is then prompted to upload scans of the documents provided.

So how does the program not work? How does an employer who uses it end up hiring ineligible workers? News outlets are saying that the Glenn Valley Foods CEO "explained that federal officials said his company was a victim of unauthorized workers using stolen identities or fake IDs to get around the E-Verify system."

We have a couple options here. None of them leave the employer, the immigrant, or DHS/SSA/USCIS looking very good:

Employees were signing up to work with scans of other people's documents. Literal identity theft, and the employer didn't catch it because they aren't actually comparing the photo on the ID to the physical person at the worksite (as they are testifying they did under penalty of perjury). This probably shows up as tax fraud later, too.
It's all a performative show. There is no photo recognition on the back end, or USCIS is failing to actually assess eligibility to work. Perhaps people on refugee status with scheduled court dates are automatically waived through. This one is on the US government.
"Photo ID" is not what you think. Form I-9 instructions and E-verify instructions both link a list of "Documents that establish identity," List B from Form I-9. These include a "school ID card with photograph" or, for minors, a "school record or report card". I can't imagine that a "school ID" is a challenging document to fake, given the number and variety of schools of higher education (includes tech schools!) around the country and how there are no standards for what constitutes a school ID. (Not to mention the number of minors who are able to buy alcohol in the US.) I also can't imagine a "report card" incorporating a photo ID, and again, they are printed by every school in the nation, and nowadays report cards are probably html files which can be modified by anyone with technical savvy before being printed off at home. So potentially all that an immigrant has to do is claim to be an under-18 refugee or college student, provide a fake ID or report card, and they can pass document inspections at a lower level of scrutiny. Again, literal identity theft, with the US government complicit. The employer must be wondering how 30% of their balding day laborers are children and students.

I think it is likely that the answer is (2). The reason this business was raided was because DHS already knew there were a bunch of people working there who didn't have the right to work. Perhaps they were previously permitted to work, and perhaps the Biden admin was letting anyone work.

In summary, I am appalled by the low standards of quality the US holds itself to for ID verification (one can also use a student ID to vote!), and I think it's possible anyone involved - immigrant, employer, or government - could be telling "motivated truths".

9

Context

recovering_rationaleist 8mo ago

Korean romance dramas aren't exactly realistic romances. If you watch just a few of them you can start to see the formula: Episode 1 introduces high-status guy and average girl who hates everything he stands for, Episode 2 we meet their friends, Episode 3 she befriends his best friend who has a crush on her, Episode 4 high-status guy has physical contact with main character in a plausibly deniable way, ... Episode 10 they kiss, Episode 11 something happens to estrange them, ... Episode 16 they marry and live happily ever after. I'm sure the writers and producers spend enough time watching dramas that they know the tropes, know the formula, and have an instinct for the progression of a good drama.

Also, I'm sure that there is a selection bias. We hear about every Marvel and Disney production even when it sucks because there is a large marketing budget targeted at English speakers; we only hear about the Korean dramas when they are actually good. (Counterexample which demonstrates the rule: Squid Game 2 sucked and had a large marketing budget, and I heard about it "organically" before it came out).

13

Context

recovering_rationaleist 8mo ago

Yup. Thank you.

2

Context

recovering_rationaleist 8mo ago

Thank you. Should be fixed, but might be paywalled. The title of the article is "If The New England Journal Of Medicine Doesn’t Correct This Error, You Cannot Trust Anything It Publishes", written on Mar 11, 2025.

4

Context

recovering_rationaleist 8mo ago

Ironically, reversibility was among the conditions that were being studied among the eight "transgender mouse" studies which the Trump administration cancelled funding for.

1

Context

recovering_rationaleist 8mo ago · Edited 8mo ago

I certainly don't think that it's a given that they need puberty blockers.

The ultimate pro-puberty-blocker argument is that if treatment is not provided children will commit suicide. Last time I looked at this (mid-pandemic), there were no randomly controlled trials on suicide rates in trans children under different treatment regimens. If you looked at the effect sizes of the few existing small Scandinavian studies about the effectiveness of different transitioning methods on suicide rates, it looked like social transition had about the same effect size as medical transition.

I'm generally in favor of doing more RCTs on children whose parents consent (and on pregnant women). There are so many medical questions that we don't have answers to because medical ethics has raised the standards for informed consent higher than is reasonable.

then when I point out the responses to it

This response to the Cass review was particularly hilarious: a paper written by two lawyers attempting to dispute the "evidentiary standards" of the Cass review, which manages to misinterpret the Cass review as well as misquote two of the scientific meta-analyses used by the Cass review. If that's the highest quality of argument they can put forward (in NEJM of all places!), then I'm going to guess that the actual "evidentiary standards" in support of their position are quite weak.

(And indeed, the Cass review is up front about there being no RCTs available for use.)

14

Context