ooops -- I figure that both factors are important. But I am beginning to doubt the meaningfulness of the 'infection rate' as a statistic. If you weren't and aren't doing random sampling of people who don't claim to be sick, then they are unmeasured. If there are hardly any of these asymptomatic spreaders, then this is fine. But if they are the majority than the picture that emerges from the data we get isn't the correct one.