Minimizers throughout the pandemic have made many false claims to suggest that the COVID incidence is less than what measures like positivity and wastewater levels show. The evidence paints a different picture.
A statistical test known as the Pearson Correlation Coefficient (r) provides a value when looking at two data sets. The formula is complicated looking and would be very difficult to calculate by hand.
In short, it’s using the x- and y-values of two different data sets to determine how closely they are related. The value for r is always between -1 and 1. If the r=1, the data sets is perfectly correlated. That’s extremely rare in the real world. At the opposite end, an r of -1 means that the data has a perfect negative correlation. An example of a perfect negative correlation would be a graph of two lines with the formulas y=x+1 and y=x-1, again, very rare with real data. A value of zero means that there is absolutely no correlation between the two variables.
If you are interested in greater detail on correlation coefficients, this paper is a good start and where I found this table. From it, you can see how the strength of a correlation is viewed in different disciplines.
Here’s what I discovered over the weekend in the COVID data. I have a table that has wastewater, the percentage of confirmed COVID emergency department visits, and positivity rates from tests. I was very surprised to see just how strong the correlation was between wastewater and ED visits (0.89). I decided it might be worthwhile to look at the coefficient between all three variables. This is a static image with the values of the data through 7/31/24.
I have tried to set up a view on this page that will update the correlation coefficients when I do my data update each week.
