Here's a little Fermi estimate I find myself forced to do. There are a number of uncertainties involved, so if anyone wants to help me refine it, then I'd be happy to hear from them. My children go to a school with about 250 pupils, and I want to guess the probability that 1/
at least one of those pupils has coronavirus.
The number of confirmed cases in Cambridgeshire was four yesterday. If the average time to get from infection to confirmation is a week, and the number of cases is doubling every two to three days, then 2/
The number of confirmed cases in Cambridgeshire was four yesterday. If the average time to get from infection to confirmation is a week, and the number of cases is doubling every two to three days, then 2/
that would suggest that the number of people now who have the virus and will in due course become confirmed cases is around 30 (with significant error bars, as 4 is such a small number). And since only a fraction of people who are infectious will become confirmed cases, 3/
the number of infectious people will be some multiple of that, which I'm going to have to guess at completely: let's say five times. So that makes 150 people in the whole county, which has a population of about 650,000. For convenience let's make that 130 people. 4/
Then the chances that any given person has it are about one in five thousand. So the chances that someone in the school is infectious are about one in 5000/250=20 -- but I stress the various uncertainties above. It's important to note that this probability is increasing ... 5/
rapidly -- in a couple of days it will probably be more like one in ten, and by the end of the week more like one in five, or it could already be that if my numbers were off. That's probably not enough for me to need to be too concerned about the safety of my own family, 6/
but looked at more collectively it is more of a problem. It means that larger schools will very likely be spreading the virus around by the end of the week if they stay open, and at least some smaller schools too, though maybe not all that much. 7/
I think it's safe to say that keeping schools open in Cambridgeshire next week would be a very bad idea. I'm uncomfortable about my children going to school this week, but next week I'll keep them at home whatever happens. 8/
Note that these calculations will vary from region to region, and may be wrong anyway, so don't try to draw any general conclusions from them: do your own calculations and draw conclusions from those. 9/
The number I'm most worried about was that wild guess that one in five people who get it will become confirmed cases. If the ratio is significantly smaller, then the probabilities become significantly bigger ... 10/10
