Well that was interesting. Genuinely, it was.
I have never really had an enthusiasm for UK politics, never mind European of Global. But now, now I do.
Tableau’s dedicated Politics month, purposefully conflicting with the UK’s EU referendum and the US presidential race, has been the underlying reason for this, exposing me to a whole world of data that I did not know existed.
Now instead of trying to form a decision by listening to the callus lies of politicians, so detached from reality that I am surprised when I don’t see that Cameron and co’s birthplaces aren’t Narnia; I can now make my own informed, objective decisions based on facts.
And it was great to see others doing the same.
Of course Tableau shouldn’t be the only one’s taking credit for this, so should all companies that are giving people the opportunity to think in a more objective way, i’m talking data viz companies, data viz preachers and the statistics departments of our governments for giving us access to more data than ever.
For us UK based data people, the Office of National Statistics (ONS) is, in my opinion, a data dreamland, providing so much data on what is often such misunderstood subjects.
So now it is up to us few, to spread the word, and make sure political decisions are, more than ever, based on objective facts rather than lies and scare mongering, and make these misunderstood subjects, understood subjects.
Click here to navigate to my politics visualisation portfolio.
My Politics #IronViz is not designed to help people make more informed decisions about an upcoming political decision (as discussed above), such as an election or referendum, it is instead designed to help people understand a recent election, help them understand voting patterns and trends, and help them understand the deeper detail.
I chose to focus on the London Mayoral election from 2016.
Why? Well having only recently moved to London I thought this was a chance for myself to better understand the politics in this richly diverse city. It is also a very recent election so I thought the outcome was still relevant and could still be used by the people of London to help understand the results.
Below I have outlined the four key questions that I wanted my audience to be able to answer, and how I represented these in my visualisation.
- Overall, how did the different boroughs of London vote.
This map is central to my visualisation. It helps answer the question noted above, ‘overall how did the different boroughs of London vote’, but also acts as a navigation tool to the rest of the visualisation, where selecting the different boroughs filters the other information accordingly.
So the first thing you are probably wondering is why the ‘lego map’ (as I have dubbed it), well when asking for feedback from my Information Lab colleagues, fellow data schooler Rob Suddaby asked that very question.
He was right to challenge my design choices, and you should always be able to justify them otherwise they should not be there.
He was right to challenge my design choices, and you should always be able to justify them , which in this case I felt I could.
- How many people took part in the election
In this summary I included detail on the number of residents eligible to vote, ‘logged electorate’, the number of good ballot papers counted ‘number of good votes’ (some votes are excluded for various reasons), and the % turnout, which is the division of the latter against the former.
These are seen as valuable metrics in terms of adding context to voting statistics, especially the turnout %.
I represented these values as text to quickly allow my readers to acknowledge these numbers before moving on to understanding the deeper detail, and if needs be they could very quickly glance over and refresh their mind on this information.
- How did each borough vote for each of the main political partiesOf course when looking to show geographical trends, representing this information on a map is one of the best way of showing this type of information as it allows for easy recognition from your audience, providing they have knowledge of the area being shown. Again I chose ‘lego maps’ to represent this information for the reasons noted in the answer to the 1st question, and also for consistency.
I decided against including the figures on the tooltips for this section as the information is available in the way in which I have answered point 4.
- How did voting patterns change from the 2012 mayoral election.People often have issues with sankeys, they see them as something that whilst great to look at, really doesn’t tell you much. I agree, in cases when they are used incorrectly. Otherwise they can be hugely informative and efficient ways of presenting flow data (and of course beautiful).
My viz is not a traditional sankey, one that is used to show relationships, interactions and flows between different elements, but instead has been adapted to show the change between one point in time to another, of the same object.
I’d like to give credits to certain members of the data viz community who helped shape this piece of work.
Chris Love, for his blog on sankeys, which helped me to understand the underlying maths behind creating a sankey diagram which I transferred into an alteryx flow.
Lisa Ding, for her great blog on dynamically formatting numbers, which means my values can be summarised to different levels. I have now used this trick on numerous projects that I have worked on.
A random blog, for the link of which I have lost, that discussed the different types of maps including a ‘lego map’.
And many others of course!