How to turn data into a story

By Paul Marsden

The BBC released data on the geographical patterns of voting in the EU Referendum last June earlier today.

The figures show how each council ward voted on June 23rd, allowing the corporation to form to a number of conclusions about the background of people who voted to leave and those that wanted to remain in the EU.

The ward-by-ward voting also demonstrated how people living so close to each other thought so differently about the referendum.

But how do you take data like this and make it into a story? Here are four key steps.

  •  Find the data

In many cases this can be quite tricky. There is some data, particularly that held by private companies, that will never be accessible. But increasingly a lot of public data is being made available.

Due to a drive for transparency government data from councils is often made available, national government produces a variety of datasets and independent bodies such as the Office for National Statistics published figures. There is even continental data.

In this particular case the BBC made Freedom of Information requests to local authorities around the country to get the ward-by-ward vote tallies. Some authorities had already published the information.

The BBC story on local voting figure 

  •  Organise the figures

This is a vitally important. You need to ensure everything is in rows that make sense. In this case this means making sure you are assigning the right vote to the correct ward, not mixing up the leave and remain votes and making sure all the results are logical – i.e. given it was a close vote in most areas, if one side has more digits, like an additional extra zero at the end, check it. It might be an error.

Once you are finished the completed data should be organised in a clean manner, like below.

The BBC EU ref spreadsheet

  •  Create a visualisation with the numbers

This stage should really help you with identifying what main stories are. You are going to take data on paper and make it into information that makes more sense to readers.

Oldham voting data

If I take the voting data for Oldham and put it into Datawrapper, an open source tool that takes much of the heavy lifting out of creating visualisations I can create the following.
The figures are based on the percentages voting for each side.

This stacked bar graph works well because this is a referendum and therefore there are only two choices for voters. Using two contrasting colours also makes it easy to see where the divide in the vote lies.

The dot plot graph works well because it allows the reader to spot where the change over occurs from Leave to Remain. It also makes it clear just how big the margin of victory was for the leave campaign in many wards in Oldham.

This ward-by-ward pie chart is interactive and allows readers to see clearly what the final tally was in their ward. This type of visualisation would be especially handy for smartphone audience (an increasing demographic) as it is very clear.

You can also see how you could present the data across the country in an interactive map here. 

  •  Find the story

Visualising the results makes it easy to see that Oldham representated a clear victory for the Leave campaign, which has made it a popular area to canvas views on Brexit among the national media.

However like many areas of the country there was a big split in the vote across its wards. If you were asking about Brexit and the European Union it is likely you would receive different responses in Hollinwood to Werneth.

Therefore those outliers, which represent either extreme of the debate in the town, could continue to prove fertile ground for local reporters on the issue of Europe and Brexit as the process of Britain leaving the European Union gathers pace over the next few months.

Main Image: DARPA/Public Domain

Related posts

Leave a Comment