Fine, I'll run a regression analysis but it won't make you happy • Hacker News Board

Re: Fine, I'll run a regression analysis but it won't make you happy

#271

Post by VancouverMan » Mon, Oct 02, 2023, 5:17 AM UTC

dghlsakjg wrote: post #215

VancouverMan wrote: post #145

Earlier quoted context omitted.

The death counts from Canada from that era are very questionable. As early as June of 2020, the "public health" authority in Canada's most-populous city (4th highest population in North America) revealed that the counting of deaths was being done using a methodology that sounds quite dubious: "Individuals who have died with COVID-19, but not as a result of COVID-19 are included in the case counts for COVID-19 deaths…

Attributing what happened in Ontario to the whole of Canada is something that happens way too often. As a Western Canadian resident, I don't think it's fair or reasonable to extrapolate about all of Canada from a single tweet from a TO health authority, and an article with a disclaimer about it's own accuracy. There well may be problems, but please keep in mind that while 50% of Canadians live between Montreal and To…

The very questionable counting of deaths seems to have been going on in the other more-populated non-Ontario provinces, too.

Alberta was using a dubious death-counting approach, as described in this April 2021 article:

"According to Dr. Deena Hinshaw, any death that has been flagged where COVID-19 is a possible cause is included in the initial count, even if the official cause of death remains unknown."

https://globalnews.ca/news/7814731/alberta-determine-covid-1...

BC was using a dubious counting approach as of April 2022:

"Fewer than half of COVID-19 deaths reported since B.C. changed counting methods were caused by the disease"

https://bc.ctvnews.ca/fewer-than-half-of-covid-19-deaths-rep...

Quebec's counting of deaths as of May 2022 (and apparently before that) was also dubious:

"Quebec’s interim public health director, Dr. Luc Boileau, has acknowledged that the province has seen a “huge” number of deaths linked to COVID-19. Quebec’s high death toll, he said last Thursday, is explained by the fact the province counts a COVID-19 death as any death involving someone who has the disease.

He said a government study from January indicated that around 30 per cent of the official COVID-19 deaths in the province’s hospitals involved people who tested positive for COVID-19 but whose principle cause of death was not the disease. He said about 40 per cent to 50 per cent of official COVID-19 deaths in the province involve people who had the disease but who died of other causes."

https://globalnews.ca/news/8801205/quebec-still-reporting-do...

Re: Fine, I'll run a regression analysis but it won't make you happy

#272

Post by defrost » Mon, Oct 02, 2023, 5:37 AM UTC

CamperBob2 wrote: post #268

panarky wrote: post #206

Earlier quoted context omitted.

>>> What about ten lives? One thousand? Ten thousand? Are 200,000 lives meaningful enough? Silver's regression model shows that for each 1% increase in the vote share for the blue candidate, the covid death rate in that state (statistically) would fall by 15.5 per million population. In areas where the blue candidate received 20% of the vote, and the red candidate received 80%, the model predicts 1,793 deaths per mil…

It's because red ideologues got the vaccine in far lower numbers. Vaccine refusal played a part, certainly, in connection with the fact that conservatives are older. In many instances, a lot older. In other instances, not really that much older. The trend is unidirectional but the degree varies by state, as you'll find when you start comparing West Virginia and Florida to Maine and Vermont. But the truth is inescapab…

From TFA:

   *The differences in state death rates are very likely because of differences in vaccine uptake*

    Just to be clear, I don’t mean to imply that COVID is intrinsically more likely to target Republicans or anything like that. Rather, my claim is that COVID is considerably more deadly in people who haven’t been vaccinated, and since Republicans are less likely to be vaccinated than Democrats, state partisanship serves as a proxy for this.

    Indeed, we can look at vaccination directly.

    ...  age and vaccination rates alone explain more than half of the variation in COVID death rates between states since Feb. 2021.

However Silver didn't mention age being strongly correlated (either way) with being vaccinated, which is interesting, given the rest of the article it'd be expected to be stated if it was.

Re: Fine, I'll run a regression analysis but it won't make you happy

#273

Post by slv77 » Mon, Oct 02, 2023, 5:38 AM UTC

candiddevmike wrote: post #252

jonahhorowitz wrote: post #238

Earlier quoted context omitted.

The measures that seem like they should have worked, but people claim they didn't are: - masks: clearly they worked for the medical professionals that were treating patients during the first waves, as they were directly exposed and didn't get sick, why didn't they work for the population at large? Did we not wear them correctly, consistently? - schools: It seems obvious that schools could be a huge vector of transmis…

RE: masks, paper masks don't do shit, gators even worse. That's what 90% of the population wore. Had we enough N-95s for everyone, I think the mask effectiveness would've increased dramatically, but I doubt folks would put up with wearing one. For me, I still wear N-95s on planes and in crowded places, I understand their effectiveness and it fits my risk/reward profile. RE: schools, they have been and continue to be…

N-95 masks work to keep you from catching the virus and cloth or paper masks (somewhat) prevented you from spreading the virus. Ideally everyone would have worn N-95 masks but the US had four problems:

1) The US didn’t have enough N-95 masks. In Asia where masks in public were more common people used masks from their home supply. I think some segments of the population were resentful that the US didn’t have access to the best masks and Asia did. 2) We had only a general idea of who was high risk. 3) We had only a general idea of the long term impacts of the virus. There was some initial concern that the virus could be like the Zitka virus where mortality rates dramatically increase with the second exposure. If that were the case the second waves could have killed off half the population with wide transmission. 4) Households aren’t homogeneous and it is near impossible to prevent transmission from a possibly asymptomatic child to a high risk adult when they share the same home.

Unfortunately given the lack of consistent messaging about masks in the US one segment of the populations believed that the masks were to protect themselves and the other segment believed it was to protect others. Those that were low risk and believed it was to protect themselves resented the masks and those that saw the masks as a way to protect others resented those that resented the masks. I think those polarized views still persist today.

Re: Fine, I'll run a regression analysis but it won't make you happy

#274

Post by fwungy » Mon, Oct 02, 2023, 5:40 AM UTC

educaysean wrote: post #258

fwungy wrote: post #224

Earlier quoted context omitted.

You don't become a career scientist or doctor by rocking the boat. A medical degree is a huge investment. Doctors have student loans and families. Anyone who sticks their neck out is at risk of losing their license or research grants.

> Anyone who sticks their neck out is at risk of losing their license or research grants. Sticks their neck out against whom? Against the Afro-Chinese/Euroamerican consortium that took kickbacks from the "medical companies" in order to keep the COVID theatre going? Those poor honest doctors around the world - if only they could speak up! If only one doctor would tell the truth!

Doctors depend on the free exchange of information to help their patients. It is one of the pathways that medicine advances. Covid was managed in a top down, hierarchical mode. Doctors who questioned it were threatened and punished by medical boards, some lost their licenses.

Had a non-vaccine treatment for covid been discovered via clinical practice it would have destroyed the EUA for the covid vaccine, i.e. that's how they were able to issue an experimental vaccine widely to the public. An effective covid treatment would have cost the corporations billions of dollars in the present, and the ongoing subscription income of annual covid vaccinations for hundreds of millions of people.

When you've got billions of income it's easy to spread it around to politicians, media, and scientists who are willing to carry the story. You just pay them well and make them sign NDAs: no one will talk. This forms a ready alliance to silence anyone who dares question the corporate narrative.

Re: Fine, I'll run a regression analysis but it won't make you happy

#275

Post by nrr » Mon, Oct 02, 2023, 5:48 AM UTC

candiddevmike wrote: post #252

jonahhorowitz wrote: post #238

Earlier quoted context omitted.

The measures that seem like they should have worked, but people claim they didn't are: - masks: clearly they worked for the medical professionals that were treating patients during the first waves, as they were directly exposed and didn't get sick, why didn't they work for the population at large? Did we not wear them correctly, consistently? - schools: It seems obvious that schools could be a huge vector of transmis…

RE: masks, paper masks don't do shit, gators even worse. That's what 90% of the population wore. Had we enough N-95s for everyone, I think the mask effectiveness would've increased dramatically, but I doubt folks would put up with wearing one. For me, I still wear N-95s on planes and in crowded places, I understand their effectiveness and it fits my risk/reward profile. RE: schools, they have been and continue to be…

Paper masks (and, by some measure, even gaiters) are for source control, and they work just fine for that purpose if the literature is anything to go by. Every software project needs it, and our squishy bodies are decidedly not an exception.

Re: Fine, I'll run a regression analysis but it won't make you happy

#276

Post by CamperBob2 » Mon, Oct 02, 2023, 6:13 AM UTC

defrost wrote: post #272

CamperBob2 wrote: post #268

Earlier quoted context omitted.

It's because red ideologues got the vaccine in far lower numbers. Vaccine refusal played a part, certainly, in connection with the fact that conservatives are older. In many instances, a lot older. In other instances, not really that much older. The trend is unidirectional but the degree varies by state, as you'll find when you start comparing West Virginia and Florida to Maine and Vermont. But the truth is inescapab…

From TFA: *The differences in state death rates are very likely because of differences in vaccine uptake* Just to be clear, I don’t mean to imply that COVID is intrinsically more likely to target Republicans or anything like that. Rather, my claim is that COVID is considerably more deadly in people who haven’t been vaccinated, and since Republicans are less likely to be vaccinated than Democrats, state partisanship s…

He actually spent a lot of time working around the obvious: that COVID absolutely is intrinsically more likely to target Republicans. Not because of any merits or demerits associated with being a Republican, but because "Republican" is strongly correlated with "old as hell" in many if not most US states, as well as "more likely to reject vaccines."

Right-wing media figures consistently fed their audience a diet of vaccine skepticism, conspiracy theories, and promotion of useless alternative treatments; e.g., https://www.nytimes.com/2021/07/17/us/politics/coronavirus-v... . Because their audience skews older, this messaging had unusually-deadly consequences.

Why there's anything controversial about any of this, I have no clue. How does it make sense for him to write "I don't mean to say that COVID targets Republicans," followed by a paragraph that states exactly why it does?

Re: Fine, I'll run a regression analysis but it won't make you happy

#277

Post by dent9876543 » Mon, Oct 02, 2023, 8:02 AM UTC

zbentley wrote: post #167

Izkata wrote: post #162

Earlier quoted context omitted.

In the US, deaths within 28 days of a positive covid test were counted as a covid death, unless they were vaccinated, in which case they also had to be sick (IIRC). So our data about how many lives vaccination saved is really corrupt, the unvaccinated count higher than it should be. On top of that there was also monetary incentive for hospitals to add covid as a cause of death for anyone who died.

Could you source those claims please? I don't mean to come off as critical, I'm just curious to read the stated rationale.

I’m not offering a source, but I’ve seen those claims made repeatedly, with no real rebuttal.

At this stage, I’m willing to take the claims at face value.

Even if the letter of the guidance was honest and impartial, I’m absolutely certain second order effects would come to bear.

If I recall right, there was even an attempt to have /all/ deaths during lockdown attributed to COVID on the misguided assumption that the only possible response was tight lockdown, and so any collateral was attributable to the cause of that lockdown, i.e., COVID. (Lots of groupthink there…) So even though we didn’t get that written into the guidance, I’m sure there would have been a strong bias to being very conservative when attributing cause to anything other than COVID.

Point is that I don’t think we even need to go as far as malicious intent to render the figures highly questionable. But then if you want to also layer nefariousness on top, the numbers become almost unbearably meaningless.

Re: Fine, I'll run a regression analysis but it won't make you happy

#278

Post by Fradow » Mon, Oct 02, 2023, 8:02 AM UTC

ryandrake wrote: post #176

logicchains wrote: post #137

Earlier quoted context omitted.

Yes, there are papers attempting that, such as this recent one, which looked at whether shelter in place policies reduced overall excess mortality: https://onlinelibrary.wiley.com/doi/full/10.1002/hec.4737 . "Using an event study approach and data from 43 countries and all U.S. states, we measure changes in excess deaths following the implementation of COVID-19 shelter-in-place (SIP) policies. We do not find that cou…

Almost nowhere on earth did "stay at home" actually happen. We all had these so-called mandates, but people (especially in more rural areas) pretty much did whatever the hell they wanted with no repercussions since there was almost zero enforcement. They were out and about, shopping for their khakis at "essential" businesses, eating at the restaurants who refused to closed (also with no consequence) and basically ign…

Nowhere on earth is a big statement. If you at least qualified it as "country-side" that would have been a bit more understandable (very low density making it pretty much impossible to control), but as others said, no, there are place where "stay at home" actually happen.

I live in Paris, and at the height of "stay at home" mandates, it was a ghost town. I had an actual reason to get out a few times, and let me tell you there were less people outside in broad daylight than there currently is in the middle of the night, which was really eerie.

There was enforcement too, since the police had pretty much nothing else to do there during the full mandates (not the latter weaker ones).

Re: Fine, I'll run a regression analysis but it won't make you happy

#279

Post by pasc1878 » Mon, Oct 02, 2023, 8:04 AM UTC

fwungy wrote: post #25

I really don't trust any of the numbers on covid. A vaccine for a dangerous illness that can affect billions of people is a very big payday. There's lots of money to spread around to politicians, corrupt scientists, and corrupt media. I'd like to see a real scientific debate with adversarial inquiry so we could know if numbers are being fudged and which ones they are. I trust science, but not greed driven capitalists…

Most of the vaccines were designed and developed in University labs - so if you trust science then you should trust them.

The COVID numbers also come from scientists in public health and statistics organisations. Why don't you trust those scientists they made no profit.

The first vaccine to be approved the AstraZenica one approved in the UK was sold at cost so that should show that it was needed.

Why do you not trust those scientists I show above.

Which scientists do you accept? and why are they different?

It did have issues in that for under 30 it had a slightly higher risk of embolisms (which is one reason that it is not use much now) and in its first tests did not have enough over 60 year olds in the trial. This latter was picked up by several things US did not approve it - how much was NIH it got nad press possibly because the not over 60s tests meant that European politicians misinterpreted so say it was unsafe and that view was not changed when it had the full tests, its PR was messed up saying things like it passed all tersts then had to backtrack making AZ look shady. Being the fitst some people pressing for patyent free vaccines criticised it from day one even thopugh much cheaper for all than other vaccines.

So everyone was critising it even though it worked.

How much of this was driven by other pharma companies lobbying because of the at cost price vs their with profit price. The AZ president has said that they would not do things at cost again as it is just not worth the issues.

Re: Fine, I'll run a regression analysis but it won't make you happy

#280

Post by uoaei » Mon, Oct 02, 2023, 8:37 AM UTC

diogenes4 wrote: post #63

staticfloat wrote: post #62

Earlier quoted context omitted.

Yes, that’s what it means for me. I’ve never heard someone use the word arbitrary to mean anything other than “a random choice”, or even “a poorly thought-out choice”. My professors in grad school explicitly discouraged use of that word anywhere in technical writing, as they felt it would immediately give the reader the impression that the actions taken in the research were not thought through. Example: “This new tec…

> I’ve never heard someone use the word arbitrary to mean anything other than “a random choice”, or even “a poorly thought-out choice”. That's shocking, I use it to mean "the result of a judgement or decision" about a dozen times a day, such as "it's not random, it's arbitrary". I had no clue people had an alternative definition for it. I'm even more surprised that otherwise ostensibly-educated people have no clue ab…

If I heard someone say that I would think they're making the point only that it was based on personal whim and not pure statistical chance. I wouldn't interpret it to mean that the person's whims are in any way reasonable, just that they're theirs.