Timely topics provide ideal fodder for data visualisation. Last year, the Scottish Independence Referendum spurred on an outpouring of infovis content.
The outcome of this referendum was largely influenced by the inclusion of 16 and 17-year-old-voters, who were eligible to vote in this referendum, the first time the voting age has been lowered in the UK. Of those eligible, 80% voted. A poll on election day found that 16- and 17-year-old voters were the age group who voted Yes most often (71 per cent to 29 per cent) — a significant factor in how close the results were.
The buzz of activity on the subject by young people likely influenced the volume of social media commentary throughout the lead-up and during the election, which provided a huge quantity of data for analysis.
We’ll look back to see how visualisations predicted and explained the close result of 55.3% voting No and 44.7% Yes to the question“Should Scotland be an independent country?”
How the Yes Vote Was Won or Lost
Creator: indy-scot.com / Graeme Lyon
Source: Twitter
Method: Processing
This is a long scroller showing a variety of charts that used social media data to predict the outcome. Data was collected over a six month period, tracking the inclusion of yes/no and hope/fear in tweets with the hashtag #IndyRef. Editorially, it was a nice touch to use Scottish slang words “…take a wee trip”, “I’m feart”. I also liked that an analysis of the big issues was created by monitoring mentions of topics such as oil, Europe, currency and poverty.
#IndyRef Election Day
Creator: The Drum
Source: Twitter
Method: CartoDB
While this map is quite visually appealling, I’m not sure it explains much. It plays through on a timeline, showing the geotagged location of tweets which are colour coded (green for Yes and red for No). I’m also seeing yellow bubbles appear, but it’s not clear what these mean. I’m assuming that larger bubbles indicate a higher volume of tweets, but it moves through the timeline too quickly to gather much detailed information. I think this could be improved with some interaction. If, for example, the user could slow down the animation, and click on bubbles to read the quotes, that would give us more context. It is interesting to see that tweets on the event orginated from every corner of the globe, though.
How Close Was the Scottish Referendum Vote?
Creator: BBC
Source:Scottish Census, Population Surveys, Election Results
Method: Unknown
This article has a series of choropleth maps examining the referendum vote in constituencies across Scotland according to a variety of metrics such as vote share, turnout, political party support, unemployment and electorate age. Key insights result from comparison the voting results with unemployment (a higher unemployment rate often resulted in a Yes vote), and with electorate age (older populations voted No while young populations voted Yes).
Scottish independence referendum: the result in three maps
Creator: The Guardian
Source: Election Results
Method: Google Maps
More choropleth maps, but these ones show a bit more clearly what just one map on the BBC site attempts to show – the split between Yes/No votes in the constituencies. In the BBC map, vote share is shown on one map, with percentage of No vote in pink and Yes in green. It’s a little easier to see what’s going on when these are separated into two maps as in the Guardian asset, but because the colour shading is shown in a semi-transparent overlay and the differences between colours are minimal, it’s still a little tricky to read. The voter turnout is shown on a third map.
These maps would benefit from tooltips showing the percentages when the user hovers over the various constituencies. Contrast with The Most Detailed Maps You’ll See From the Midterm Elections from The New York Times in 2014, and you’ll see a more elegant solution to displaying this type of information.
No comments yet.