Extreme Color Without Proper Baselining Intentionally Distorts the Message

In a Wall Street Journal article online, there was a reference to COVID infections on a state-by-state or a county-by-county basis across the nation. The calculation was based on the number of cases in a normalized population of 100,000 so that individual states and counties could be properly compared. That part of the visualization analysis was reasonable and followed analytical best practices.

See the WSJ link here.

However, the choice of color scheme and the intentional distortion of the actual data by the thresholds that were set to engage the color range for the data display were intentionally configured to portray an extremely large “percentage” of the population across the country that was experiencing the latest variant during the period of reporting. The maximum figure in the data on a per county basis was close to 300 people per 100,000 or a total of 3 tenths of one percent of the population. So less than one third of one percent of the population in the “high transmission” areas was becoming infected. But you would not know that relationship to 100% of total by looking at the color scheme since the natural understanding of a color range is that the range itself fully spans from 0 to 100% of the data figures. So the color range was skewed to show maximum red at one third of one percent of maximum.

That is called misrepresenting the data and it is done frequently and even more often now that creating “interesting” and motivating visualizations is accessible to most everyone.

Be sure to understand the data range and the percent of total when making sense of data with color ranges.