News for hackers. Stuff that matters.
Re:California Businesses Leave the State by the Thousands by jacobolus on Sunday September 13, 2020 @08:54PM (#24465477)
The Hoover Institution is really a stain on Stanford’s reputation. Full of pseudo-scholars who start with their conclusions and then hunt around for cherry-picked evidence, but don’t spend too much effort worrying about the details, and never bother to admit when they were wrong. It’s fitting that Condoleeza “we don't want the smoking gun to be a mushroom cloud” Rice is now the director.

The example of the day is Scott Atlas, Trump’s new darling “anti-Fauci” who (without any particular expertise in epidemiology or public health, and without any kind of convincing evidence or reasoning) has concluded that we should just give up against Covid19, and disclaim any responsibility for the results.

Re:California Businesses Leave the State by the Thousands by guscost on Sunday September 13, 2020 @09:12PM (#24465602)
You make it sound like it is uncommon for a think tank to publish research that helps its backers and allies. “The example of the day” (provided by NPR?) is just one small feature of a much larger landscape.
Re:California Businesses Leave the State by the Thousands by jacobolus on Sunday September 13, 2020 @09:16PM (#24465632)
No, it’s not uncommon at all. There is a whole industry of billionaire-funded GOP “think tanks” whose business is laundering bullshit. It still reflects very poorly on Stanford though.
Re:California Businesses Leave the State by the Thousands by guscost on Sunday September 13, 2020 @09:22PM (#24465669)
You really think that this phenomenon only happens on one side of the political aisle? I can’t relate to any worldview where that situation is possible.
Re:California Businesses Leave the State by the Thousands by jacobolus on Sunday September 13, 2020 @09:26PM (#24465705)
It’s not really comparable. The difference is that scholars who aren’t GOP operatives generally live in what one senior Bush-administration official derisively called the “reality-based community.” https://en.wikipedia.org/wiki/Reality-based_community

Of the people who work for CAP, Brookings, or the like, which ones do you think are hack pseudo-scholars?

Re:California Businesses Leave the State by the Thousands by guscost on Sunday September 13, 2020 @09:28PM (#24465726)
This world of yours where there are presumably only “real intellectuals” in one camp, and a plague of “hack pseudo-scholars” in the other, is simply absurd. It makes no sense to me so I cannot comment on it.

I could go into how I feel about the whole cult of NPR news-announcers going after their enemies day in and day out, but you probably wouldn’t want to hear it.

Re:California Businesses Leave the State by the Thousands by jjeaff on Sunday September 13, 2020 @11:21PM (#24466433)
I have seen very little partisanship from npr reporters. Much less "going after" specific people on the regular.
Re:California Businesses Leave the State by the Thousands by thu2111 on Monday September 14, 2020 @04:12AM (#24467923)
There's a lot of hack pseudo-scholarship going on in universities, almost all on the political left. Just yesterday there was this article on HN:

https://fantasticanachronism.com/2020/09/11/whats-wrong-with...

And that doesn't include the various grievance studies, nor, epidemiology, which turns out to be absolutely full of hack papers that cherry pick their sources, abuse citations, make absurd assumptions and then bury them in appendices, etc etc. If Dr Fauci has been replaced (didn't know that) then that's certainly just fine given the overall level of intellectual rigour and integrity in the field. You could use a bunch of laymen and get more trustworthy output.

The problem you're decrying is the proliferation of pseudo-science in general, it's not a political problem.

Re:California Businesses Leave the State by the Thousands by jacobolus on Monday September 14, 2020 @05:20AM (#24468257)
Your argument is: using a university’s reputation to lend credibility to abject bullshit written by career GOP political operatives in return for large donations from billionaires is okay, because grad students at universities often publish piles of methodologically unsound studies in their hurry to get their credentials and get out the door, especially in fields like evolutionary psychology, marketing, management, and criminology.

Seems like a complete non sequitur to me. We were talking about poorly researched and argued advocacy/opinion, not nominally peer-reviewed scholarly papers.

Your link specifically points out that political activism is not directly related to the problems of studies that won’t replicate: “Limited Political Hackery: Some of the most highly publicized social science controversies of the last decade happened at the intersection between political activism and low scientific standards: the implicit association test, stereotype threat, racial resentment, etc. I thought these were representative of a wider phenomenon, but in reality they are exceptions. The vast majority of work is done in good faith.”

But I won’t disagree that there are many problems with academic publishing and the for-profit journal industry; the publish-or-perish career system; the exploitation of grad students, postdocs, and untenured teachers; people throughout academia using statistical tools who lack basic training/competence in data analysis and statistical inference; etc.

Re:California Businesses Leave the State by the Thousands by thu2111 on Monday September 14, 2020 @02:56PM (#24473753)
It's pretty much the same. I'm sure the Hoover Institution uses some level of peer review for their papers, except those peers probably get paid to review it whereas academics don't. Peer review doesn't do much, it's very much getting you to first base quality wise, often not even that. It can be effective if peers are able and willing to really go deep, burn bridges and object to the core thesis of a paper. Clearly not going to be true at the Hoover Institution but equally untrue within most academic fields.

Many bad papers are bad because they're actually political papers dressed up to look like science, thus (ab)using a university's reputation to lend credibility to abject bullshit. I know that link claims most of them aren't. I think that's good news. But note, that's because many of the papers that replicate, only replicate because they're making vacuous observations. Of the papers that get attention because they make surprising claims, many of them are politically or ideologically motivated.

Partly this depends on how widely you define politically motivated. For instance it's common for supposedly scientific papers to run a model, and then advocate for certain government policies based on the output predictions. This is political activity. It's also often not scientific because the models haven't been validated, and in fact if/when the political recommendations are acted upon it renders their predictions unfalsifiable, which a disinterested scientist should not want.

This problem is much larger than grad students. Bad papers are routinely being written by career academics, professors, large groups of them.

Re:California Businesses Leave the State by the Thousands by microtherion on Sunday September 13, 2020 @09:35PM (#24465790)
Just look at the sourcing of this report: The number cited comes from the random blog of a Texas real estate company, citing a random (paywalled) newspaper. That's barely the quality of work of a really lazy high school student.
Re:California Businesses Leave the State by the Thousands by guscost on Sunday September 13, 2020 @09:37PM (#24465803)
What’s your point?
Re:California Businesses Leave the State by the Thousands by microtherion on Sunday September 13, 2020 @10:07PM (#24466039)
This is an absolutely ridiculous lack of scholarship. There's a good reason why HN prefers primary sources, and the moderators often replace links to secondary sources with the primary source. This particular submission has, at best, FOURTH HAND information (assuming that the Dallas Business Journal did not do original research).
Re:California Businesses Leave the State by the Thousands by Guthur on Sunday September 13, 2020 @09:13PM (#24465612)
Starting with a hypothesis (conclusion) is pretty standard, otherwise you are just poking around blindly.

The current "strategies" have only been sustainable because our economies had grown enough to sustain the blow, for now. Many countries don't have that luxury and eventually the West will feel it as well.

The strategies are based on a hope that a meaningful vaccine can be found in a timely fashion, and that is no where near guaranteed. So no matter which way you slice it it's all gambling with our lives.

Re:California Businesses Leave the State by the Thousands by jjeaff on Sunday September 13, 2020 @11:26PM (#24466457)
There is a big difference between a hypothesis and a conclusion.

You start with your hypothesis, then, after your research and study, you formulate your conclusion as to whether your hypothesis was correct or not.

If you start with the conclusion that your hypothesis is correct and only look for supporting evidence, that's bias.

Re:California Businesses Leave the State by the Thousands by rayiner on Sunday September 13, 2020 @09:06PM (#24465551)
Dr. Atlas had some bad judgment getting involved with the Trump administration, which has a unique ability to snap defeat from the jaws of victory. But you’re misrepresenting his policy recommendation, which is actually not that different than what has been tried in other countries:

> Dr. Scott Atlas is the Robert Wesson Senior Fellow at the Hoover Institution, an accomplished physician, and a scholar of public health. For several weeks, Dr. Atlas has been making the case in print and in other media that we as a society have overreacted in imposing draconian restrictions on movement, gatherings, schools, sports, and other activities. He is not a COVID-19 denier—he believes the virus is a real threat and should be managed as such. But, as Dr. Atlas argues, there are some age groups and activities that are subject to very low risk. The one-size-fits-all approach we are currently using is overly authoritarian, inefficient, and not based in science. Dr. Atlas’s prescription includes more protection for people in nursing homes, two weeks of strict self-isolation for those with mild symptoms, and most importantly, the opening of all K–12 schools.

Other countries had relatively brief shut downs, and had targeted reopenings based on risk profile. Schools have been reopened in Europe since May: https://www.nytimes.com/2020/05/10/world/europe/reopen-schoo...

His view is probably the correct one, at least if anybody was competent enough to implement it. We should isolate high risk patients, quarantine the sick (real quarantine, like they do in Asia, not just “go home and infect your family” we do here), and let everyone else get herd immunity.

Re:California Businesses Leave the State by the Thousands by dragonwriter on Sunday September 13, 2020 @09:11PM (#24465590)
> Other countries had relatively brief shut downs,

And if the US had actually ever done a shutdown right (or, on a smaller scale, if the few states that did do that on a state level had the legal and practical ability to impose the kind of travel restrictions against the rest of the US that almost every country in the world has against the US because of COVID), it (or those states, in the smaller scale case) might have started to recover after a relatively brief shutdown.

Re:California Businesses Leave the State by the Thousands by jjeaff on Sunday September 13, 2020 @11:29PM (#24466481)
I'm not seeing this one size fits all policy anywhere.

Many schools are in session. Many are not. Some types of businesses are shit down and other types are not.

As for heard immunity, the high r value of this virus indicates that we would need as much as 50-60% of the population to get infected before heard immunity starts working.