The time has arrived for cumulative charts
Mar 08, 2021
Long-time reader Scott S. asked me about this Washington Post chart that shows the disappearance of pediatric flu deaths in the U.S. this season:
The dataset behind this chart is highly favorable to the designer, because the signal in the data is so strong. This is a good chart. The key point is shown clearly right at the top, with an informative title. Gridlines are very restrained. I'd draw attention to the horizontal axis. The master stroke here is omitting the week labels, which are likely confusing to all but the people familiar with this dataset.
Scott suggested using a line chart. I agree. And especially if we plot cumulative counts, rather than weekly deaths. Here's a quick sketch of such a chart:
(On second thought, I'd remove the week numbers from the horizontal axis, and just go with the month labels. The Washington Post designer is right in realizing that those week numbers are meaningless to most readers.)
The vaccine trials have brought this cumulative count chart form to the mainstream. For anyone who have seen the vaccine efficacy charts, the interpretation of the panel of line charts should come naturally.
Instead of four plots, I prefer one plot with four superimposed lines. Like this:
I agree. I call this chart a "smokestack chart." I wrote about it during the early days of the pandemic: https://blogs.sas.com/content/iml/2020/03/30/smokestack-plots-cumulative-curves.html
Posted by: Rick Wicklin | Mar 08, 2021 at 02:39 PM
I ask what information the chart is intended to show. If it is that this season so far lags way behind the previous three seasons, then a chart is not needed; just state the conclusion backed up by the 1 death noted in the title. If it is to display the patterns within the seasons, then the chart fails, as only early data are available for the current season. There is still time for an unfortunate gain in deaths. If it is to focus on pediatric deaths, the chart again fails, as there are no data on total deaths. The weekly data show interesting variations week to week, but again, that is not the apparent focus of the chart. The cumulative chart beneficially removes that weekly detail.
Posted by: Richard Krablin | Mar 09, 2021 at 07:43 PM