It's time to end the consensual hallucination of fall in-person classes

[spork]

A bit of a lite wildcat strike at Spelman. Many faculty are refusing to return to face-to-face https://www.ajc.com/education/some-spelman-faculty-refusing-to-teach-in-person-due-to-covid-19-safety-concerns/65UMGAILTBETDGZOE4EUUPM4MU/

Another story on the same event: https://www.theatlantavoice.com/articles/spelman-faculty-hault-face-to-face-classes/.

[Caracal]

Is it time to revive? Rice University is going online for the first two weeks given a high positivity test rate. In terms of vaccination, 94% of students are vaccinated and 82% of the faculty. They have a mask mandate indoors and are testing. Obviously, the fact they are testing is why they detected the virus in the Rice community. Faculty can get an exemption to keep teaching in person.

https://coronavirus.rice.edu/news/provost-fall-course-instruction-update-aug-19
https://coronavirus.rice.edu/news/dou-please-read-important-updates-our-fall-start
https://www.khou.com/article/news/health/coronavirus/rice-university-covid-virtual-learning/285-9413fc02-9035-4ef0-9a77-5989b2a97e1f

I mean, what's the end game? Vaccines aren't perfect, but they should blunt transmission and will prevent the vast majority of severe disease. That doesn't mean we should just do nothing, but online teaching isn't sustainable in the long run. This won't go on forever, but I don't think there's any reason to think this is all going to be gone by next semester.

As long as we're still having waves of infection and hospitalization, we haven't reached a "new normal". Specifically, if the area surrounding an institution is having a rising rate of hospitalizations for covid, remote teaching may be the most prudent choice. On the other hand, if/when we reach a point where there is a steady state of 2-3 hospitalizations (or whatever)  for covid over a period of at least a couple of months, then it's probably realistic to see it as the forseeable future reality.

I don't know when that's coming, but I think we have to think about the harms on both sides. Quite honestly, I have to include the personal harms in that. I find remote teaching depressing and it takes a lot of the joy out of my job. Besides, if reasonable precautions like masks are required, the evidence suggests that teaching should be relatively safe. There weren't any outbreaks I'm aware of in the last year connected to classes.

[Stockmann]

Is it time to revive? Rice University is going online for the first two weeks given a high positivity test rate. In terms of vaccination, 94% of students are vaccinated and 82% of the faculty. They have a mask mandate indoors and are testing. Obviously, the fact they are testing is why they detected the virus in the Rice community. Faculty can get an exemption to keep teaching in person.

https://coronavirus.rice.edu/news/provost-fall-course-instruction-update-aug-19
https://coronavirus.rice.edu/news/dou-please-read-important-updates-our-fall-start
https://www.khou.com/article/news/health/coronavirus/rice-university-covid-virtual-learning/285-9413fc02-9035-4ef0-9a77-5989b2a97e1f

I mean, what's the end game? Vaccines aren't perfect, but they should blunt transmission and will prevent the vast majority of severe disease. That doesn't mean we should just do nothing, but online teaching isn't sustainable in the long run. This won't go on forever, but I don't think there's any reason to think this is all going to be gone by next semester.

The endgame should be to just get it right, like Singapore. It's not that we, humanity, don't know how to handle Covid, it's that most places have botched at least one key aspect and/or have faced low compliance - anti-maskers and anti-vaxxers in the US and indeed much of Europe, abject failure in vaccination schemes in Australia and New Zealand (and in indeed in several of the wealthier parts of Asia vaccination was also botched), etc. Generally, it's been failures of leadership/governance (Sweden, France, Tanzania most spectacularly of all), societal failures (Russia) or both (Mexico, Brazil, India).
At this point the answer to f2f classes is almost blindingly obvious: a multi-layered approach involving vaccine mandates, precautions such as open windows, facemasks, etc, and testing and contact tracing. Where things are particularly bad, a gradual re-opening could be done, keeping bigger classes online and initially having only practicals, labs, etc f2f.
What bugs me most about not getting it right is that a lot of these measures would not have been unfamiliar to a medieval plague doctor, and even vaccines aren't that new. In some respects, modern disease management has been a victim of its own success - not least in the widespread belief, in much of the world that has been developed for a long time, that infectious disease is a Third World problem.

[Caracal]

Is it time to revive? Rice University is going online for the first two weeks given a high positivity test rate. In terms of vaccination, 94% of students are vaccinated and 82% of the faculty. They have a mask mandate indoors and are testing. Obviously, the fact they are testing is why they detected the virus in the Rice community. Faculty can get an exemption to keep teaching in person.

https://coronavirus.rice.edu/news/provost-fall-course-instruction-update-aug-19
https://coronavirus.rice.edu/news/dou-please-read-important-updates-our-fall-start
https://www.khou.com/article/news/health/coronavirus/rice-university-covid-virtual-learning/285-9413fc02-9035-4ef0-9a77-5989b2a97e1f

I mean, what's the end game? Vaccines aren't perfect, but they should blunt transmission and will prevent the vast majority of severe disease. That doesn't mean we should just do nothing, but online teaching isn't sustainable in the long run. This won't go on forever, but I don't think there's any reason to think this is all going to be gone by next semester.

The endgame should be to just get it right, like Singapore. It's not that we, humanity, don't know how to handle Covid, it's that most places have botched at least one key aspect and/or have faced low compliance - anti-maskers and anti-vaxxers in the US and indeed much of Europe, abject failure in vaccination schemes in Australia and New Zealand (and in indeed in several of the wealthier parts of Asia vaccination was also botched), etc. Generally, it's been failures of leadership/governance (Sweden, France, Tanzania most spectacularly of all), societal failures (Russia) or both (Mexico, Brazil, India).
At this point the answer to f2f classes is almost blindingly obvious: a multi-layered approach involving vaccine mandates, precautions such as open windows, facemasks, etc, and testing and contact tracing. Where things are particularly bad, a gradual re-opening could be done, keeping bigger classes online and initially having only practicals, labs, etc f2f.
What bugs me most about not getting it right is that a lot of these measures would not have been unfamiliar to a medieval plague doctor, and even vaccines aren't that new. In some respects, modern disease management has been a victim of its own success - not least in the widespread belief, in much of the world that has been developed for a long time, that infectious disease is a Third World problem.

Totally agree.

[histchick]

Is it time to revive? Rice University is going online for the first two weeks given a high positivity test rate. In terms of vaccination, 94% of students are vaccinated and 82% of the faculty. They have a mask mandate indoors and are testing. Obviously, the fact they are testing is why they detected the virus in the Rice community. Faculty can get an exemption to keep teaching in person.

https://coronavirus.rice.edu/news/provost-fall-course-instruction-update-aug-19
https://coronavirus.rice.edu/news/dou-please-read-important-updates-our-fall-start
https://www.khou.com/article/news/health/coronavirus/rice-university-covid-virtual-learning/285-9413fc02-9035-4ef0-9a77-5989b2a97e1f

I mean, what's the end game? Vaccines aren't perfect, but they should blunt transmission and will prevent the vast majority of severe disease. That doesn't mean we should just do nothing, but online teaching isn't sustainable in the long run. This won't go on forever, but I don't think there's any reason to think this is all going to be gone by next semester.

As long as we're still having waves of infection and hospitalization, we haven't reached a "new normal". Specifically, if the area surrounding an institution is having a rising rate of hospitalizations for covid, remote teaching may be the most prudent choice. On the other hand, if/when we reach a point where there is a steady state of 2-3 hospitalizations (or whatever)  for covid over a period of at least a couple of months, then it's probably realistic to see it as the forseeable future reality.

I don't know when that's coming, but I think we have to think about the harms on both sides. Quite honestly, I have to include the personal harms in that. I find remote teaching depressing and it takes a lot of the joy out of my job. Besides, if reasonable precautions like masks are required, the evidence suggests that teaching should be relatively safe. There weren't any outbreaks I'm aware of in the last year connected to classes.

Last year, our university system was required to mask on campus, and to distance where possible.  Granted, the Delta variant wasn't in our state, at least not the levels that are assumed now, but most of us were not able to get to get a first dose of vaccine until the middle of the term.  We had some cases, but relatively few.  This surprised us all, given that the nursing cohort was able to distance due to number of students vs. the small size of the classroom available.  Those at high risk, or with members of the household at risk, could request to work remotely in some form or fashion. It wasn't ideal, but it worked.  I was scheduled to teach most of my load online anyway, so even that hasn't been a big deal for me.  I like it.

That said, the system is not requiring masking this year, and we can't require it in the classrooms ourselves.  No accommodations for COVID-19, no spacing in classrooms.  Most students do not wear masks.  Given that our county is well under 40% vaccinated and we are having severe spread right now, faculty and staff are quite anxious for ourselves and those around us. 

I'll also add that our local ICUs are filled to capacity.  Most are COVID patients, according to the hospitals. 

[Vkw10]

Our chief medical officer just did an email blast pointing out that vaccines, masking, testing, and reporting positive tests via our contact tracing app are IMPERATIVE to having ONSITE classes and events ALL SEMESTER. We can't require (idiot governor), so administrative strategy seems to be half pleading, half threatening to go online mid-semester.

Provost said during university town hall that faculty can not require masks, should not pass masks out in class, but can announce that masks are available are available in every department office. Also said that if course schedule says on-site, the university standard of 85% of scheduled class time must be F2F applies and we can do up to 15% in another modality, just as we do when attending conferences during semester.

Provost said it's up to faculty to set attendance and make up policies. Suggested faculty might want to routinely stream class on Zoom to help students who are self-isolating after exposure keep up. Told us we didn't have to interact with Zoom attendees or if count them as attending, just be sure students knew that they wouldn't be able to earn in class points via Zoom if that was our policy.

Classes start Monday. Looked like about half of employees and students were masking indoors last week. Despite the suggestion that we could go online mid semester, I expect us to be F2F all year.

[Aster]

All of our classrooms are stocked with a box of face masks for anyone who wants one. A face mask, not the box I mean.

[whynotbc]

Update on Rice: mostly false positives, which is good news. They are staying remote for the first two weeks given the planning that already went into that.
https://coronavirus.rice.edu/news/covid-19-response-weekly-update-august-22

[Caracal]

Update on Rice: mostly false positives, which is good news. They are staying remote for the first two weeks given the planning that already went into that.
https://coronavirus.rice.edu/news/covid-19-response-weekly-update-august-22

I really don't understand what anybody was thinking with that. It would have been a positive rate of over 75 percent for the entire campus. By comparison, the counties having the worst infection numbers are at 25 percent positivity rates, and that's a sample of people who are coming in to be tested, often because they have symptoms, not surveillance testing.

If you're in charge at Rice and you get those figures, I can't really understand how your first thought wouldn't be "wait, that can't be right." It seems totally bizarre that you'd change course and move online for two weeks based on those numbers before you got somebody to make sure there wasn't a mistake.

It just doesn't speak well for the competence of the people making these decisions.

[mamselle]

The fellow I know there, who teaches in the sciences and has a wife doing chemo at the moment, thinks it was a good idea.

M.

[Puget]

Update on Rice: mostly false positives, which is good news. They are staying remote for the first two weeks given the planning that already went into that.
https://coronavirus.rice.edu/news/covid-19-response-weekly-update-august-22

I really don't understand what anybody was thinking with that. It would have been a positive rate of over 75 percent for the entire campus. By comparison, the counties having the worst infection numbers are at 25 percent positivity rates, and that's a sample of people who are coming in to be tested, often because they have symptoms, not surveillance testing.

If you're in charge at Rice and you get those figures, I can't really understand how your first thought wouldn't be "wait, that can't be right." It seems totally bizarre that you'd change course and move online for two weeks based on those numbers before you got somebody to make sure there wasn't a mistake.

It just doesn't speak well for the competence of the people making these decisions.

I don't think you read that quite right-- per the linked article the initial test positivity rate was 2%, not 75% (75% was the percent of positive tests from asymptomatic individuals). 2% might not seem high, but is a lot higher than the usual rate for routine testing on campuses (.24% for them historically as the article notes). And, they did detect the anomaly based on the oddities in the data, as the article discusses. Should they have caught it sooner? Probably, but 2% fell in a zone where it was unusual, but not so improbable as to spark an immediate deep dive into the data to see what was up. These are the hardest types of data issues to catch. Standard checks by a person or automated data pipeline looking for unusual patterns (like big discrepancies between test providers)would certainly be a good idea, and one hopes they (and others) implement that going forward.

That said, we've been getting a master class in how even apparently smart and well-educated people often don't understand probability and statistics at all well. I'm not talking here about the obviously uninterested in the truth anti-vaxxers, but about seemingly well-intentioned actors who persistently misunderstand things like:
--base rates (e.g., in highly vaccinated populations a high percent of infections will be in vaccinated people but that does *not* mean the vaccines are ineffective!)
--confounding (e.g., if most older people are vaccinated and fewer younger people, than vaccinated people with infections may have as or more severe cases than unvaccinated people with infections and that again doesn't mean vaccines are not effective! See Simpson's Paradox)
--selection bias (e.g., if you are looking only at people who *chose* to get tested, most likely because they are symptomatic, viral loads may be as high in vaccinated as unvaccinated cases, whereas it turns out that is *not* true when looking at all cases identified through non-self selected testing).

I could go on. . .

[Caracal]

Update on Rice: mostly false positives, which is good news. They are staying remote for the first two weeks given the planning that already went into that.
https://coronavirus.rice.edu/news/covid-19-response-weekly-update-august-22

I really don't understand what anybody was thinking with that. It would have been a positive rate of over 75 percent for the entire campus. By comparison, the counties having the worst infection numbers are at 25 percent positivity rates, and that's a sample of people who are coming in to be tested, often because they have symptoms, not surveillance testing.

If you're in charge at Rice and you get those figures, I can't really understand how your first thought wouldn't be "wait, that can't be right." It seems totally bizarre that you'd change course and move online for two weeks based on those numbers before you got somebody to make sure there wasn't a mistake.

It just doesn't speak well for the competence of the people making these decisions.

I don't think you read that quite right-- per the linked article the initial test positivity rate was 2%, not 75% (75% was the percent of positive tests from asymptomatic individuals). 2% might not seem high, but is a lot higher than the usual rate for routine testing on campuses (.24% for them historically as the article notes). And, they did detect the anomaly based on the oddities in the data, as the article discusses. Should they have caught it sooner? Probably, but 2% fell in a zone where it was unusual, but not so improbable as to spark an immediate deep dive into the data to see what was up. These are the hardest types of data issues to catch. Standard checks by a person or automated data pipeline looking for unusual patterns (like big discrepancies between test providers)would certainly be a good idea, and one hopes they (and others) implement that going forward.

That said, we've been getting a master class in how even apparently smart and well-educated people often don't understand probability and statistics at all well. I'm not talking here about the obviously uninterested in the truth anti-vaxxers, but about seemingly well-intentioned actors who persistently misunderstand things like:
--base rates (e.g., in highly vaccinated populations a high percent of infections will be in vaccinated people but that does *not* mean the vaccines are ineffective!)
--confounding (e.g., if most older people are vaccinated and fewer younger people, than vaccinated people with infections may have as or more severe cases than unvaccinated people with infections and that again doesn't mean vaccines are not effective! See Simpson's Paradox)
--selection bias (e.g., if you are looking only at people who *chose* to get tested, most likely because they are symptomatic, viral loads may be as high in vaccinated as unvaccinated cases, whereas it turns out that is *not* true when looking at all cases identified through non-self selected testing).

I could go on. . .

Oh I see, that makes far more sense.

Agree with you about all of that. The base rate errors drive me nuts...

[Puget]

Agree with you about all of that. The base rate errors drive me nuts...

Did you know that 90% of base rate errors are made by right-handed people?
(I say smugly as a lefty)

[arcturus]

Agree with you about all of that. The base rate errors drive me nuts...

Did you know that 90% of base rate errors are made by right-handed people?
(I say smugly as a lefty)

I assume that this is also the reason that 90% of all automobile accidents happen within 10 miles of home?

[Puget]

Agree with you about all of that. The base rate errors drive me nuts...

Did you know that 90% of base rate errors are made by right-handed people?
(I say smugly as a lefty)

I assume that this is also the reason that 90% of all automobile accidents happen within 10 miles of home?

Well, we'd need to know the base rate of driving within 10 miles vs. more than 10 miles from home, but seems likely. You can add many more examples (majority of injuries happen in the home, etc.)