Coronacademics ?

14 Apr Coronacademics ?

To be clear, this post is not any sort of rigorous (or even unrigorous) analysis, but just a “noticing” of a trend that has appeared in my personal Twitter feed. As someone who follows quite a few scientists (mostly cognitive science, psychology and neuroscience), there has been, unsurprisingly, a lot of talk/retweeting about COVID-19. This goes for Twitter in general, of course, but as many research scientists have jobs that give them the flexibility to decide what they will spend their time investigating, a significant proportion of these people seem to have turned their attention to work focussing on COVID-19.

I’m curious how to think about this. People have had their own takes already, with one in the social/behavioral psychology realm going along the lines of:

1. We need people to change public behavior, therefore we (social/behvioral psych people) are the ideal experts to inform policy decisions.
2. Backlash: Stop being so opportunist. It’s irresponsible to make hasty claims with low power because they “might” save lives. They might also do the opposite.
3. Backlash-backlash: Stop being so negative. We’re all in this together and everyone can try and help in whatever way they feel best suited.

I don’t really have a strong position on this debate, but I think it is very interesting to consider the scale of academic work focussing on COVID-19 at the moment. I just spoke with a collaborator on a paper that I am working on who is at Penn and his research group has 5 COVID projects currently in the works. What percentage of academic hours in the past month have been COVID hours? And how does that vary across departments? And if you account for all the hours/resources, how much has academia effectively spent on COVID-19?

Of course it goes beyond academics too – everyone with access to an internet connection and a graphing package seems to be coming up with their own plots/models for different metrics of the pandemic. Overall I think that this is good, but it is definitely adding a lot of noise to the signal (wherever that is). Which is at least great for confirmation bias purposes.

My own opinion/bias, that epidemiologists have a strong incentive to overstate the danger of the pandemic, being helpfully supported by this plot referring to Ontario, Canada, where reality is well below the “best case” model (model data from the government of Ontario)…