Europe’s arms exports to Libya: what’s the best visualisation to use?


The Guardian Datablog recently published data showing Europe’s arms sales to Libya, after the embargo was lifted, between 2005 and 2009.

I’ve decided to put three visualisation websites and software to the test: which one will produce the most coherent map of European countries’ arms deals with Libya?

In a talk I found the other day, Amanda Cox, a member of the New York Times Graphics department, said that in the best visualisastions, ‘something surprising pops out.’ So that’s what I’ll be looking for in my own maps.

The three websites and software I’ve chosen to test are Many Eyes, Tableau Public and OpenHeatMap.

Creating my dataset

At this point, I should draw attention to how Simon Rogers describes this data in the Datablog article: ‘There are some caveats… these are licenses, so actual sales could be less. They also don’t show who the end-user is. So, for example, some of the French licenses are undoubtedly granted for UK companies exporting via Paris. The data is perhaps deliberately obscure.’

Although this consideration won’t change how my maps look, it’s always important to know as much about the data you’re using as possible and to acknowledge that the maps won’t provide a completely accurate visualisation of arms sales.

Once I’d transferred the Guardian dataset to my own Google Spreadsheet, I deleted the columns showing the different arm types because I wanted my map to show how much each European country’s arms licences to Libya amount to.

There are also figures showing the total revenue of arms licences to Libya for Europe as a whole. I put those to one side, to use later on. So, my columns read ‘country; year; total.’

I’m ready to visualise.

Many Eyes

I was hoping for a map that would show a year by year progression of EU arms licences to Libya. To achieve this, I had to change the way my data was presented so it looked like this.

After much chopping and changing, I was satisfied with the result.

You can view the five maps on one screen, which is ideal for comparing the countries’ arms licences to Libya from year to year – the progression really is striking.

You can also view one map at a time and change the year in the scale:

The main problem with this map on Many Eyes is that there’s little room for manoeuvre. Once you’ve created the visualisation, you can’t make changes to the scale.

You really have to adapt your dataset to Many Eyes because it won’t go out of its way to understand your data. If it doesn’t recognise one of your countries, it will offer you a list of alternatives that are completely off the mark:

Malta had to be left out on this occasion.

So although the result was great, Many Eyes wasn’t the easiest to operate.

Tableau Public

With this software, I was able to create a map where the size of circles was proportionate to the amount of money the licence exports brought to the selling country.

What’s great about Tabeau Public is its flexibity: it lets you drag and drop the columns and rows you want to use in your visualisation. You can be more creative with colours and shapes than on Many Eyes, too.

You can also create different visualisations from the same dataset and publish them together on a dashboard.

As well as the map, I created a chart of year by year arms licenses to Libya from the EU countries as a whole, much like the one from the initial Guardian article. It’s a useful chart because it immediately shows that after 2006, EU countries as a whole made more and more money from granting arms licences to Libya.

Thanks to its dashboard tool, I think Tableau Public offers a more complete understanding of this particular dataset.

Open Heat Map

Out of the three, this was the only website I had never tried before so I was expecting to struggle slightly. In fact, it was so easy that on my first attempt, I managed to create a great interactive map.

You simply press play and it runs through the years and the countries appear in a more or less intense blue according to how much money they made selling arms licences to Libya that year. Click on the map to have a go.

Although the three websites and software present coherent visualisations of the data, I think Tableau Public stands out with its dashboard option. Coming back to Amanda Cox’s remark on good graphics, ‘something surprising pops out.’

Claire Gilmore (@ClaireEGilmore)

Advertisements
This entry was posted in Data Journalism Experiment, How is data journalism used?, Visualization Experiment and tagged , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s