> Just take their mobile phone number, and text them en masse with a quick Y/N question as to whether or not they got sick?
I've been wondering why this isn't just being done for everyone in the country, every few days?
"The main symptoms of COVID-19 are X, Y, Z. If you think you may have COVID-19 please respond "Yes" to this text message. This is free. Your response is private and used to understand the spread of COVID-19 in our country. Please visit http://some.link for more information"
If someone responds "yes" you could follow up with some more questions, if that was useful. And then text them a few days later to see if they're feeling better or worse, or maybe ask them to describe how they feel.
There are lots of problems with this idea: people lying or not taking it seriously; people not responding; not everyone has a cellphone; some people will be worried about privacy, etc etc...
But surely there'd be something useful to learn from it? And it seems like it should be easy to do, really, if you got the mobile providers on board (or just order them to help).
In Italy they're personally tracking every confirmed case. Meanwhile, here in NYC, I called 311 to report my suspected case and that wasn't even a thing they were tracking. We just don't have our shit together over here, not at a state or national level.
Vague and/or non-comparable data provides decision deniability and narrative flexibility in many possible futures. Open-source analysis and watchdogs provide one level of defense.
I recently completed a survey conducted by NORC which asked a series of questions that seemed to test whether I had contracted Covid-19. I received the survey invitation in the mail so it must have been randomized. We'll have to wait to see the results.
In the uk at least we have an app that you can update with your current health status/feelings and they can track the Wuhan flu with that. They are encouraging users to report in once a day even if you have no symptoms.
I saw it suggested on Twitter yesterday that the political polling firms should be doing some coronavirus polling, tracking trends in self-reported cases the same way they track public opinion.
I'm not sure if that would be scientifically useful in any way, but it sure would be interesting.
I've been wondering why this isn't just being done for everyone in the country, every few days?
"The main symptoms of COVID-19 are X, Y, Z. If you think you may have COVID-19 please respond "Yes" to this text message. This is free. Your response is private and used to understand the spread of COVID-19 in our country. Please visit http://some.link for more information"
If someone responds "yes" you could follow up with some more questions, if that was useful. And then text them a few days later to see if they're feeling better or worse, or maybe ask them to describe how they feel.
There are lots of problems with this idea: people lying or not taking it seriously; people not responding; not everyone has a cellphone; some people will be worried about privacy, etc etc...
But surely there'd be something useful to learn from it? And it seems like it should be easy to do, really, if you got the mobile providers on board (or just order them to help).