Very interesting post on Nightingale, Unfair Comparisons: How Visualizing Social Inequality Can Make It Worse, Eli Holder December 1, 2022. Briefly, it is about how charts showing averages (means) may make it easier for the viewers biases to draw poor conclusions; this is particularly evident when the data is segmented by race.
A proposed solution is to better show the individual data points (people), so that it is easier to see the spread of individuals (especially those categorized by race). Thus, one is less likely to assume that all members of a particular race score the mean. One can see the different people, some low and some high. One way to display this is to use a Jitter chart, where individual data points are shown along with and indication of the mean.
I like this approach for reasons outlined in the article – shows that not all individuals are the mean, some people overlap with people in other segments – but I also like that it may remind viewers that the data represents individual people.
Philip Treacy has a nice write up on Jitter in Excel Scatter Charts on myonlinetraininghub.com. I will need to give it a try when the opportunity strikes.
