Applying a few data literacy principles, I've rewritten the story for a more general audience (i.e. less analytical).
[00:00:08] I was looking for something interesting to cover about COVID-19 and came across this new visualization on Tableau Public. It's actually published by the COVID-19 Data Hub.
[00:00:20] The first thing that I want to cover and mostly for this discussion is about color. There are also a few technical things I'd like to go over. The first thing that jumped out to me where these colored boxes, yellow, green (which is colored blue), and red.
[00:00:43] I'm not sure what the point behind the green is, but I expect the names are there because they understood that there might be a conflict between coloring boxes green and red, and people with color vision deficiency might not be able to distinguish the difference.
[00:01:02] Instead of using color, I think using shapes here might be a bit more practical. For example, for positive we could use a thumbs up; negative, a thumbs down; yellow could be an OK.
[00:01:27] Also, you may notice that all of the ranges are different for the colors, depending on the particular metric, so it might be a little simpler just to use some kind of imagery here.
[00:01:40] The other reason I'm thinking about using shapes is that we've got gradients in play as well, so that's a lot of color going on and it might detract from what people are looking at.
[00:01:56] I like the idea of using the gray gradient for the population, but there are some things going on with the gradient for the mask percentage. First, everything is showing up as blue because the scale is capped at one hundred percent. So instead of capping at one hundred, make that number a little bit higher.
[00:02:29] Also, it's not clear what the sorting is. It's not alphabetic; it's not by masks distributed, and we saw that it's not by percentage.
[00:02:46] The other thing that I'd like to cover from a technical standpoint is this is not a long list and I feel like we could potentially lay out this view slightly differently to eliminate the scrolling. The last point I'd like to cover is in terms of the timing. Here we see that these are set up to be 30 days and it might be helpful to show the date ranges, but I think it's OK like this as well.
[00:03:16] The confusing part is when you go over here is not knowing what the date range is. A large number of masks, seven hundred and seventeen thousand, made me think this might actually be covering the entire time since the pandemic started. But you can see it starts on September 2nd and goes to the end of October, so it's 60 days, not 30 days.
[00:03:38] Take a look below to see my final idea on what this could look like.