At what point do randomly sampled tests make sense? I guess now you'd test too many negative and would need too many tests to get reliable results. (you need t

At what point do randomly sampled tests make sense? I guess now you'd test too many negative and would need too many tests to get reliable results. (you need to have a decent sample of sick people for statistical significance). Than again the infections also cluster locally a lot, so you can't extrapolate for a country well..