How not to do Coronavirus Analysis

Bad simple regression in my analysis? It’s more likely than you think!

Do not use lm(Total.Covid.Cases ~ variable, data=data) and call it a day; amplify the experts instead. Don’t use Google Trends either. Please.

There are so many things that need to be thought of here. Mask usage, mask policy, municipality vs state… just because you find something significant, doesn’t make it very relevant. Feature Selection in a year’s time is going to be very curious to find how it spread… but for the time being, wear a mask.