Huge gap Between Perception and Experienced Tensions in Dutch Society

A graphics iteration process.

In the Netherlands tensions between different groups of the population is a rising hot topic. To name a few that support this trend:

  • The Dutch elections: where popular populist and politician Geert Wilders used a repeatedly polarising rhetoric.
  • The refugee challenge Europe copes with. This results in (un)conscious racism that puts pressure on ‘the Dutch identity’ when dealing with the integration of refugees.
  • The Turkish Prime Minister Erdogan who tried to expand its power with a referendum. A lot of people in the Netherlands were surprised to see how big the group of Erdogan supporters was among the people who have a Turkish ánd a Dutch passport.

When the Central Bureau of Statistics published a population trends report: Tensions in the society and individual experiences thereof (loosely translated) I took a closer look. When I saw the charts in the report, I wanted to improve them to answer the for me important question: What is the difference between the experienced tensions in peoples’ personal surroundings and the peoples’ perception about these tensions.

End-product

Critique

At first, the chart on the left is not that bad. But we can do better when you want to guide the reader comparing the reported tensions (perception) and the experienced tensions between different groups of the population (Persons with and without a migration background, Religious groups, Poor and the Rich, High- and low-educated).

The colours in the original graph have to be improved since the data represents a sequential scale (from nothing to often), so the use of a qualitative palette is not advised. Besides, the colours seem randomly chosen and force you to look at the legend a lot. Below, you see a first version (on the left), where I:

  • Grouped the reported and the experienced tension per population category for better comparison.
  • Changed the colours into a sequential palette.
First version

This is better but comparison is still a bit hard since the categories don’t start at the same point. With the second iteration I decided to:

  • Delete the legend and place the values in the chart itself, so you don’t have to switch focus. Now you can easily read from top to bottom.
  • Change the red in the blue colour to get rid of negative connotations people have with the colour red and to create a better contrast with the background.
Second iteration

We’re getting there but after receiving feedback and a more thoroughly investigation of the data, a third iteration was needed. The feedback entailed that the difference between reported and experienced tensions were not understandable. I therefor decided to go for perception (this term was mentioned in the original report as well) and an explanation of the difference in the intro of the graph.

Then I interviewed the data more thoroughly by asking: how was the data collected? I found out that respondents were asked in the first question about the extent to which they think there are tensions in the Netherlands between certain groups of people (perception).

Detail of last iteration.

When asking who was included in the experience part of the data, a new insight occurred. Only if the respondents answered the first question with: a lot of tensions, the researcher asked about the extent to which they experienced these tensions in their surroundings. That’s why I wanted to represent this visually and added an explanation within the intro text of the complete and final visualisation (end-product).


This is a repost and part 4 of the series Data Visualisation Redesigned for the Better.

For more redesigns and other data journalism/visualisation related articles, go to my blog:

Do you stumble upon a crappy graph? Please let me know! Cheers 🙂