Data visualization is quite often a struggle to represent multiple relevant dimensions preserving the readability of the plot. The paper presents an underutilized approach of colorcoding, in which the position of an element in a three-dimensional array of data is represented with a single color. To illustrate the technique I address the question of population ageing. European population is ageing rapidly, but the process is not happening uniformly in all parts of Europe. Regions differ quite a lot: Eastern Europe still undergoes demographic dividend; Southern European regions form a cluster of lowest-low fertility; Western Europe experiences the greying of the baby boomers; urban regions attract young professionals and force out young parents; peripheral rural regions lose their youths forever… How can we grasp all the differences at a glance?

For each NUTS-3 region the unique color is produced by mixing three distinct color spectrums in the proportions that reflect deviations from European average in the share of elderly populating (aged 65+) -- magenta, population at working ages (15-64) -- yellow, and kids (0-14) -- cyan. It is important to note that this map is not meant to be able to inform the reader of the exact population structure in a specific region. Rather, it provides a snapshot of all the regional population structures, facilitating comparisons between them.

https://i.imgur.com/iznvA6o.png

Figure 1. Colorcoded map of population structures in European NUTS-3 regions in 2015.

Colorcoding is a useful and intuitive way of displaying three variable datasets at once. The obvious drawback of the map is that it is not colorblind friendly, and there is no way to make it so because color is the main player in this dataviz.