If you are tired of bad COVID data interpretations by reporters and your annoying uncle, then this article is for you. While your humble author chases hackers for a living, the discipline of cyber is sufficiently adjacent to data science as to warrant some insight
Now – if the number of COVID cases is growing, then either the infection rate is increasing, the number of tests is increasing, or both. When one goes up and the other down, then the number of cases will depend on relative intensity. Let’s look at some examples:
If the number of tests stays the same and the infection rate increases, then you will see more cases. This is bad news, because regions can always do more testing – and this would produce more cases. The graph below is an example.
If the number of tests increases and the infection rate stays the same, then you will see more cases. This is not bad news, although any region would prefer to see the infection rate going down. The graph below is an example.
Many more cases exist, obviously, including the number of tests going up, and the infection rate going down. This is great news, and if the relative intensities are favorable, then the resulting number of cases will go down. The graph below is an example.
Data scientists generally recommend that citizens pay closest attention to infection rates, because this closely tracks the progress of the virus. Good charts for infection rates can be found here. This is where our examples were obtained.
Oh – and in case you were wondering, the first graph is for Florida, the second graph is for California, and the third graph is for New York. This explains why Floridians should be concerned, Californians should be guarded, and New Yorkers should be satisfied.
You know a bit of data science now, so please turn off the TV and go review the infection rates in your state. If you don’t like the results (#Arizona, #South Carolina, #Oklahoma, #Texas, #Utah), then call your governor and demand action.
