Home icon
Data Visualisation Guide

Preserving scales for comparisons

2 minutes read

Pitfalls in dataviz: scales

Small multiples is a visualisation technique in which small copies of the same chart showing different partitions of the data are arranged in a grid. They can be a great way to avoid packing too much data into a single, overcrowded visualisation.

Small multiple line charts showing the share of renewables in household energy consumption in the EU and its member states. Source: buildingsdashboard.eu

But in order to make accurate and meaningful comparisons, the y axis on the small multiples should be kept constant. Otherwise readers not paying attention to the labels on the y axis will quickly draw false conclusions from the small multiples.

Here is an example of the BBC falling into this pitfall.

4 charts showing the Covid hospital patients in England, Scotland, Wales and Northern Ireland. All charts have a different y axis

Source: @RobDunsmore

Keeping the x axis constant on all small multiples should be obvious, but this rule is not always respected, as this example shows:

6 small multiple line charts showing vaccination rates in different age classes in England. The x axis is different for every chart

Source: One in four elderly black people in England still not vaccinated, theguardian.com

Related pages

Breaking scales

Scales in line charts

Width to height ratio

Starting bars at zero

Line interpolations

Line charts: double y axes

Pitfalls in dataviz: scales