Analyzing 100 years of election results in one chart

Hi! This is David, taking over for this year’s fourth installment of the Weekly Chart series. We’ll take a look at a historical trend in UK elections: through an area chart created by Sam Joiner of The Times that illustrates how sometimes, the simplest visualizations can be the most effective.

If you’re creating charts in a newsroom, a general election will be one of the busiest times of the year. From live results to visual deep dives, there’s a lot of raw data to collect, analyze, visualize and contextualize, and a lot of room for creativity in data visualization. However, sometimes simple visualizations can have the most striking effect. In his article about the history of female MPs in British Parliament, Sam explores the history of gender imbalance in British Parliament:

In this chart, you can see the gender (im-)balance in the parliament from 1918 right up until after the 2019 general election. When asked about the thought process behind the visualization, Sam explains:

One of the key decisions we had to make was whether to show the absolute numbers of MPs or the stacked percentage. Since the total number of MPs has changed over time, a chart showing the actual numbers of male and female MPs would have technically conveyed more information, but would also have been harder to read in practice.

Left: A rejected first draft of the chart showing the absolute numbers, rather than percentages.

The absolute numbers have historical significance, but they are not the story we wanted to tell as it is the difference between male and female that is more striking (and the narrative of the article). That’s why we’ve decided to include them as annotations in the final chart, but take the normalized percentage as the basis for the display.

As with many charts, the majority of the work was in the research phase rather than in the actual visualization. The data was initially sourced from Wikipedia, cross-checked with the Parliamentary Archives, and then merged into a single dataset. That doesn’t have to be a bad sign though, as Sam explains:

Sometimes when you spend days or even weeks finding data, it can feel like you are underpaying a chart’s significance if you come up with a way of visualizing it in a matter of hours. But the key is in the presentation, rather than the time it takes to create something.

That’s it for this week. To play around with the chart yourself, click on “Edit this chart” in the top right corner of the visualization to open an editable copy in Datawrapper. Stay tuned for next week’s Weekly Chart, which will be brought to you by our map specialist Anna.