Bad practices in data visualization

By Alberto Cortinas, 2024-10-15

"Criticism of a data visualization proposal that could be improved"

The visualization I've chosen to do this study on bad practices in data visualization is this publication from Eurostat in which the creator wants to answer the question *When Europeans fly nest?*.

Bad practice in data visualization

The source data is the average age at which young people leave the parental house in several european countries, and the technical solution that the creator has implemented is a bubble chart where every country is a bubble and each bubble has:

  • The name of the country
  • The average age
  • A color defining the cluster of age that the country belongs to.
  • The bubble diameter correlated to the age (bigger bubble for countries with grater age).

The problems I see in this approach are:

It is not as self-explanatory as it should

It is difficult to understand that the bubble are located in the place the country is in the real world as the shape of the visual is far from the shape of Europe. In addition, color scales do not help the user to see which cluster has greater average age than the other: why do green countries leave parent house at older age than blue ones?. Some different colors are similar and they are difficult to differentiate (blue vs violet).

Difficulty in contextualizing data

This visual does not help to easily answer some questions like:

  • Which is the country with people leaving parental home older? And younger?
  • What's the difference between countries?
  • Which is the position vs European average?

A good visualization must help comparing data of different countries and infer relationships between them

My proposal

My proposal to improve the visual, as it's representing geographical data, is to present it using a colorpleth map in which each country will be painted with a color saturation that represents the average age at wich young people leave the parental household.

The color scale will be a gradient from blue to white, where blue represents the youngest age and white the oldest. The user can hover over the countries to see the detailed data.

I will add a bar graph that will rank all the countries in descendant order and it will be easy to see the difference versus European averages.

The legend will be placed at the bottom of the map to help the user understand the color scale.

When European fly nest?

Average age at which young people leave the parental house, 2022

Difference with respect to the European average

With this approach, I improve some aspects of the visual:

  • It is easier to identify each country
  • User can infer relationships between data (mediterranean, nordic)
  • Comparison of different countries it is more clear and easy