Colours are often used in data visualisations to emphasise variation, ease interpretation and make them aesthetically pleasing. However, we do not all see colours in the same way and you should take this fact into account when you design figures. Specifically, as a start, you should make sure that everybody can actually see the variation made by the colours in your figure. If not, it is easy for the audience to misunderstand or misinterpret data visualisations (Franconeri et al. 2021).
Luckily, there are many easy ways to improve your figures in order to make them more colourblind-friendly. Here is a list of seven recommendations:
- Read the three-part series on colorblindness by Lisa Charlotte Muth: How your colorblind and colorweak readers see your colors, What to consider when visualizing data for colorblind readers, and What’s it like to be colorblind.
- Make sure your visualisation is not dependent upon various colours. There are several ways in which you can reduce the number of colours in your figures. See, for example, this post with 10 ways to use fewer colours in your visualisations.
- Use a colourblind simulator to explore how your figures will look for colourblind people. You can use Sim Daltonism for Mac and Colour Simulations for Windows. In R, you can also use the package colorblindcheck (see the vignette for examples).
- As a rule of thumb, try not to use green and red in the same figure. There are many ways in which you can make mistakes, and one of the easiest ways is to use green and red together. This is not always a problem, and there are ways in which you can use specific versions of red and green (see, for example, this post), but if you use green and red, make sure to double check that they are indeed colourblind-friendly.
- Use colourblind-friendly palettes. There are several colourblind-friendly palettes you can rely on, if you want to make sure that your figure is colourblind-friendly. In R, for example, you can use
ggthemes::scale_colour_colorblind(). - When you create maps, use the colourblind-friendly package viridis. Noteworthy, these colours are not only colourblind-friendly but also look better than most of the default map colours currently available.
- Follow ColourBlindAwareness (@colourblindorg) on Twitter. They share a lot of relevant material as well as good and bad examples of figures using colours.
I continue to make mistakes with colours when I make data visualisations. Luckily, people are good at telling me that they have problems reading my figures when I do not consider colourblind-friendly templates. It is easy to forget the fact that a lot of people are not able to see colours in the same way as yourself, but great figures should be easy to read for everybody.