School Opening - Not a Binary Choice
Binary and Linear Thinking are Dangerous
This may seem a little disjointed to you. I assure you, by the end, you’ll have seen all the pieces to consider and realize the school opening arguments are disingenuous. Once you’ve viewed all the threads, you’ll see the tapestry. You’re being sold a binary choice when you need to see shades of grey.
The reality is that schools should be viewed holistically within respective communities and should have gates or trip wires for shifting from full function to reduced classes, further gates to focused support for less privileged students, with yet more marks to shifting various levels of support, and finally to full closure. Those arguing to keep schools open seem to think that because Europe has been successful, we can too. That would be true if we copied everything Europe did. Unfortunately they are only looking at copying the school part and not even all of their school measures. The United States has a much higher background prevalence of disease. The United States lacks mass quantity rapid surveillance testing. Those residing in the United States don’t adhere to procedural controls as well as those in Europe. Don’t try to counter with “we need to follow rules.” It’s a nice sentiment but fails to live in reality. We need to deal with our reality. If, however, you want to push for better enforcement of rules, please do lobby for such. Ventilation has been recognized as critical for preventing spread. Yet we’re talking about open windows in winter. Good luck implementing that.
This is creating an emotional binary choice discussion when really we should be setting determinants to drive levels for closure and levels for opening while recognizing degrees to such closed to open status. This is a compound complex issue best suited for the professional class in government and should not be oversimplified for general public consumption. It is compound in seeking best ways to reduce spread while complex in that it is intwined with other societal problems that may shift back and forth in predominance.
———————————————
Our review starts with the writing of Nina Schwalbe in The Atlantic. Why Are We Closing Schools? Keeping kids out of the classroom will make recovering from the pandemic harder in the long term, while not keeping us any safer in the near term.
“Keeping students home is unnecessary. Reopening schools doesn’t appear to meaningfully increase the level of risk faced by teachers or students, but closing them causes well-documented damage to students. Evidence from around the world—and even from New York City—shows not only that many schools should remain open, but that officials should take more steps to open up classrooms.” This clearly states the binary position to which I strongly disagree. Yet even it concedes such may not actually be universally applicable simply by using the word “many.”
“Many school districts, including New York City’s, have already implemented a number of measures to keep students, teachers, and their families safe: blended learning, active screening for illness, universal mask wearing, and increased ventilation, among others. These precautions are working, and all the early data show that classrooms do not appear to drive transmission of the virus.” How exactly are we screening? It certainly isn’t through periodic surveillance testing of the nature that would capture the asymptomatic and the more contagious pre-symptomatic. Seems there’s infighting at The Atlantic in this regard with Olga Khazan, “Some experts think that at-home testing can stop coronavirus outbreaks, and that the government should have been doing more to produce the tests.”
Back to Schwalbe, “Six feet is also more than most European countries require. In some parts of Switzerland, children are not required to distance while at school, though staff and students 12 and older must wear masks... Norway and Spain have imposed age-based distancing rules for students over 13 and 9 years old, respectively—but still mandate less than six feet. Italy and Portugal both recommend keeping students and adults three feet apart. Likewise, France, which has returned to national lockdown, recommends distancing of three feet and masks in schools for students 6 years and older.” I cannot emphasize this enough, we are not Europe. Their starting position is significantly further left on the curve.
“New York City’s own data on its partial reopening show similar results: Schools reflect the prevalence of the virus in the community, but do not drive community spread. According to New York City government data provided to me (I served as a resource for the school district in an informal, unpaid capacity), the city performed more than 74,000 tests in 1,224 schools during a three-week period in October, and just 45 students and 63 staff members tested positive.” That sounds like a lot of testing. It sounds like adequate surveillance. But it’s not! A quick google search led me to Department of Education Data at a Glance, which shows NYC has a student population of 1,126,501. That’s 6.6% of the population tested. For surveillance testing to be viable, you need to cover the population and you need to do so with an adequately tight periodicity. Three weeks ain’t adequate. Nor is a mere 6.6% of population tested. In Germany, testing occurs for all much more often on the government’s dime. Their community has less load while activities outside school are more curtailed. If you want to copy Europe, you must copy everything Covid related.
———————————————
From WAPO We’ve figured out it’s safe to have schools open. Keep them that way. Danielle Allen and Ashish Jha:
“For untold thousands of children, schools are their only source of healthy meals. And for too many children, they are a refuge from a precarious home life, a place where observant teachers can be a safety net.” This is a valid concern that shows the complexity of competing societal demands. It does not mean schools need to be open, however. Instead they can still work partially closed.
“Schools, particularly grades K-8, should be exempt from closure requirements. They should be given authority to determine, based on the broader state of the pandemic at any given time, what percentage of their students are educated remotely or in person, but we must ensure that the schools remain able to serve, at all times, those students who need in-person instruction.” Agree though this is not the same as “keeping schools open.” It is instead reducing risk by reducing the school attending population and reducing crosstalk between Covid bubbles. It helps keep the more disadvantaged kids safe by reducing contact and spread opportunity produced from the more affluent kids attendance. To reduce the number of nodes is to create an exponential reduction in contacts.
“Families should have a critical voice in making that determination.” - Disagree. Individuals should not be determining risk acceptable for others. This extends to families as each family individually should not be determining risk for other families, yet leaving decisions to families achieves exactly this.
“To meet the needs for education during a pandemic, schools and educators must be equipped with resources and capacities for infection control so they can function safely even when community spread levels increase. And educators must be supported as essential workers.” - So, how do you propose handling this when schools aren’t properly equipped? Arguing to open schools is a disingenuous argument. Instead, you should be pushing the means for them to do so safely. Yet you don’t have surveillance screening nor adequate tracing and you want ventilation to include open windows while heading into winter. Treating teachers as essential workers is a noble idea. Yet we’ve seen how essential workers have been treated in the recent past to include medical providers wearing garbage bags and food chain workers threatened and bribed - coerced - to work.
“Pandemic resilience means in-person learning is available to all students who need it throughout the whole course of a pandemic.” Disagree. Resilience is bouncing back after the fact. It is not accepting undue risk to include risk of death. Instead you need to focus on balancing risks, food security versus disease. In-person learning is not part of this equation. In person remote access, perhaps.
“Just as hospitals never close, schools should never close for some students, and for most they should be last to close and first to open.” Agree, but this isn’t the same argument being made.
———————————————
Most of the articles arguing to keep schools open trace back to Emily Oster. Let’s look at her work. A while back, I wrote a critique of her piece in The Atlantic. Here it is in its entirety, if you’ve already seen it, you may skip to the half-section break:
Today The Atlantic Let Me Down. Here’s the letter I felt I had to write them nearly two months ago:
I find it odd that The Atlantic recently published a piece by Emily Oster stating that schools aren’t in fact the disease vector concerns we feared a mere day after ProPublica publishes a piece calling out 44 outbreaks in Illinois schools. The Atlantic’s piece raises suspicion with me as it closes by calling out “Democratic governors who love to flaunt their pro-science bona fides in comparison with the anti-science Trump administration don’t seem to be aware of this growing body of evidence.” They didn’t need to put a political spin on this subject but they did. Such immediately makes me question motive behind the writing while also makes me skeptical of the piece. I gain further concern with “On social media, people shared pictures of high schools with crowded hallways and no masking as if to say I told you so... those photos of hallways don’t count” yet we’re provided with no explanation as to why such photos shouldn’t count. Concerns seem justified when considering “If school isn’t safe for everyone, why is it safe for low-income students? And if school is safe for low-income students, why isn’t it safe for everyone?” Such tries to paint protection from disease as falsely in competition with addressing racism, economic inequality, and systemic injustices. We know this to be false because as Dr. Aaron Carroll explained in the NYTimes, risks are cumulative while “we aren’t very good at discussing trade-offs. We want it all. We want to eat in restaurants, crowd into houses, go to work and celebrate occasions en masse.” This is to say that low-income students are safer at learning centers because the other students aren’t there. Here’s Dr. Carroll MD again, “Too many view protective measures as all or nothing: Either we do everything, or we might as well do none. That’s wrong. Instead, we need to see that all our behavior adds up.” And, “Instead of asking why we can’t do certain activities, we might consider what we’re willing to give up to do them more safely. Even better, we might even consider what we’re willing to give up so others can do them, too.” Consider the ProPublica piece shows an average of 180 new cases per day in children in Illinois after schools opened though they averaged less than half with 72 per day before schools opened. Packing the schools with kids is like allowing the forest to fill with dried wood. There’s no guarantee it will burn, but when it burns, it really burns. You yourselves recently published a piece by Zeynep Tufekci highlighting that K not just R varies; it seems in her advice Dr. Oster (not an MD) has ignored this.
Sincerely,
[Fffflats]
This letter writer is a former Naval Aviator and Test Pilot; in other words, he knows a thing or two about risk management.
—————
I will concede ProPublica’s reflection on Illinois schools was merely correlation. Israel, however, clearly showed causation. Per Isabel Kershner and Pam Belluck in the NYTimes, “Confident it had beaten the coronavirus and desperate to reboot a devastated economy, the Israeli government invited the entire student body back in late May. Within days, infections were reported at a Jerusalem high school, which quickly mushroomed into the largest outbreak in a single school in Israel, possibly the world. The virus rippled out to the students’ homes and then to other schools and neighborhoods, ultimately infecting hundreds of students, teachers and relatives... The lesson, experts say, is that even communities that have gotten the spread of the virus under control need to take strict precautions when reopening schools. Smaller classes, mask wearing, keeping desks six feet apart and providing adequate ventilation, they say, are likely to be crucial until a vaccine is available.“
I’d like to pause on the idea of risk being additive per Dr. Carroll for a moment. A way a brother of mine put this to me is most fitting, “Imagine a deck of cards. Take the Joker and shuffle it into the cards. Now, every action you take requires you to pull cards. Some actions require more cards to be drawn than others. If you pull the Joker, you lose.” Think a moment going to eat at a restaurant requires four cards while going to a bar for a drink or sitting front row at a choir requires six cards. Going to school may only require one or two to be drawn. Fortunately you can add cards to the deck over time. Consider it like adding two cards from the discard back to the deck and reshuffling before every draw. Unfortunately, failure to adhere to mitigation controls means you may also have to shuffle more Jokers into the deck.
—————
More recently Emily Oster wrote in the WAPO Schools are not spreading covid-19. This new data makes the case.
“This data from New York doesn’t mean that all schools are safe to open in all environments. When levels of community spread are very high, the chance of spread at school, too, is certainly higher. What it does mean, though, is that by far the most helpful thing we can do for schools is to control community spread.” This is exactly right. Schools are just as flammable as the communities in which they reside. School closure should be determined by conditions in the community. Yet this point of hers is buried under several paragraphs trying to say schools are not dangerous under a title stating “Schools are not spreading Covid-19.” Perhaps this should be changed to “Schools are not currently spreading Covid-19” to better reflect reality to include risk. This piece properly prioritizes aiming at communities holistically but it fails to consider what to do if preventing community spread has failed.
———————————————
I agree that schools should have a certain priority with a first open last close mindset. We should also adopt an understanding for levels of closure. The writings ultimately tend to make proper prioritization discussions. But their tone is that of blindly keeping schools open no matter what. They fail to appreciate we’re not Europe. We have much higher prevalence while we have very little surveillance testing. We also don’t do as well adhering to procedural controls. Many of our buildings likely have poor ventilation with winter coming. They are trying to sway an emotional drive to keep schools open. Really we should be having discussions to determine trip wires for when to close and when to re-open. And we need to be careful not to fall for the false security of thinking it hasn’t happened therefore it won’t. Especially, as in Israel, it already did. Though I’ll admit, if you give me mass rapid surveillance testing, I’ll go along with the binary open the schools option. This means instead of selling the binary choice, provide the damn tests.
* Top picture from Where’s the Greater Risk? How do we Mitigate? In the article, it emphasizes the small step with big consequences on the right side. In such, we may not perceive how small a step yields the big consequence. Yet here, I’d rather focus on the left majority of the curve. Lots of steps or a big step have yielded little consequence, hence this is the learned experience. Unfortunately the learned experience is misleading relative to what’s about to happen.
Monday, Nov 23, 2020 · 5:40:23 AM +00:00 · Fffflats
It occurs to me I gave you no book recommendations. I like to do that, so here we go:
- Michael Lewis The Fifth Risk. This highlights the danger of inept government as well as shows concern for willful malice. It showed a path of known increasing risk without benefit yet we continued to follow it. This is the definition of folly.
- Jordan Ellenberg How Not to be Wrong. Shows the folly of binary and linear thinking. Shows the risk of stories.
- Daniel Kahneman Thinking Fast and Slow shows impact of biases, story first reasoning, importance of non-linear thinking. Shows we are all susceptible to such (myself included). Shows knowing about the problem does not cure the problem.
- James S.A. Corey The Expanse series. Shows human pursuit of smaller group competitive interests in the face of obvious need to cooperate, shows deliberate inaction as evil action, shows greedy opportunism. Highlights cascading failures.
- As we’ve mentioned folly, Barbara Tuchman The March of Folly. We know we’re causing significant problems yet we seem unable to stop our march deeper into cycles of problems. (See also The Atlantic on Death Spirals.)
Tuesday, Nov 24, 2020 · 10:26:08 PM +00:00
·
Fffflats
In this piece, I concluded with “Though I’ll admit, if you give me mass rapid surveillance testing, I’ll go along with the binary open the schools option. This means instead of selling the binary choice, provide the damn tests.” I had presumed mask mandates were in effect. That was a poor assumption. Per ProPublica, “They were greeted by the principal, teachers and staff at Swainsboro Middle School who hadn’t seen them in four months. Before allowing the children to enter, a longtime receptionist beamed a temperature gun at their foreheads... Masks were optional, and about half of the children wore them... At least nine middle school teachers would be infected, including four along a single hallway; one would spend four weeks on a ventilator, fighting for her life. More than 100 students were quarantined because of positive cases or exposure. Within the first two months of school, the county would have one of the highest proportions of school-age COVID-19 cases in the state.” We should revise opening to a couple criteria, mask mandate AND mass rapid surveillance testing should do the trick. Temperature checks and questions are not adequate for surveillance. This also shows schools can in fact be sources of spread. It also suggests though doesn’t prove children as likely vectors. Such should give pause to our push for open schools. Mandates and tests work. You can try to add increased spacing likely via less in person student load and better ventilation as able after implementing the two hard means to more quickly snuff the fire and keep the burning away. The two alone should work, however. Give me mask mandates and mass rapid surveillance testing, I’ll go along with the binary choice. Mandate masks and provide the damn tests. In the absence of tests, community prevalence becomes a key measure though such would still require masks. For communities lacking mass rapid surveillance testing, you’ll likely want community prevalence marks for total closure, reduced opening limited to food and internet access, limited in person student load, and full functioning. Either hit the two big criteria, masks and mass rapid surveillance testing, or don’t be binary.