Now that I’ve introduced Tableau from the perspective of my working life, I wanted to go off-topic a bit by looking at it from the perspective of a learner. That means for this post I won’t be talking about the main theme of this blog, namely data analysis in the context of UK higher education fundraising – so I’d advise you to bail now if that’s what you’re here for.
In this post I’m going to talk about my experience of submitting an entry to #MakeoverMonday. This is a weekly initiative led by Andy Kriebel (Head Coach at The Information Lab) and Andy Cotgrave (Technical Evangelist at Tableau), the latter of whom says the aim of it is to improve people’s data literacy (he elaborates on that here). It’s been running for several months now but I only got involved for the first time a couple of weeks ago, and the reasons for this have been captured well by both Chris Love and Adam Medcalf. I won’t repeat what they’ve said, but I’m pleased that these two data visualisation experts have said it publicly, and whereas before I felt like my efforts weren’t up to scratch I now believe that it doesn’t matter as long as I’m learning as I go along.
But I digress. Let’s get back to the evening of 11 August, in which I set myself a 1-hour time limit to create a visualisation in response to that week’s #MakeoverMonday data on Olympic gold medals. You can see my final submission above, and the original visualisation is here.
It’s all well and good to have faith in producing simple visualisations, but before I’d even opened the date file I had one thought: what has everyone else done? I started with Andy Kriebel’s own blog, where he’d summarised the changes he made to the original visualisation. Andy also collates each week’s entries and adds them to a Pinterest board, and even though that week’s entries weren’t present yet I tried to draw inspiration from the previous submissions. This approach had the opposite effect to what I’d intended: I felt demoralised and overwhelmed by the quality of what I saw. Once again I felt like I wasn’t good enough, but by returning to that feeling I recognised that admiring the work of others wasn’t going to encourage me to do something of my own. There was only one thing for it: time to look at the data.
So it was that the familiar feeling returned: I felt overwhelmed. How was I supposed to make sense of this? How could I possibly represent all of this data in a single visualisation, and do it in under an hour? The clock was ticking…
That’s when the obvious occurred to me: just because I have 981 rows of data doesn’t mean I need to use it all. In fact, Andy Kriebel took a similar approach by focusing on the top 25 countries – so what if I looked at the top 10 by total medal count in descending order?
What, precisely, should I do with this?
It doesn’t tell much of a story at this point, but it should be obvious from glancing at the chart that the USA has dominated the medals count. That’s interesting, but what’s more interesting is that the Soviet Union is in second place, so let’s limit the data to just these two competitors and bring in the “Edition” field to see how things change over time (I’ve dropped Country into the Colour mark here, using red for the Soviets of course):
Ok, so the Soviets only appear between 1952 and 1988. And look at that: they’re absent from 1984, whilst the USA are absent from 1980. What’s going on here?
Time to consult Wikipedia. The article on the 1980 Summer Olympics explains that many countries boycotted the Games that year as they were held in Moscow. A smaller number of Eastern Bloc countries returned the gesture 4 years later when the Games were held in Los Angeles. Clearly there’s a story to tell here: it’s the Cold War in microcosm, and things came to a head in the early 80s.
This is all well and good but that bar chart looks pretty uninspiring to me. I need something exciting like Andy’s dot plot, but how I do produce something like that? The Tableau community is very supportive and I’m sure there’d be a tutorial out there on this, so I start Googling.
Within moments I glance at the clock: I’ve already spent more than half an hour getting to this point. To hell with it, I’m just going to have to use a bar chart. There’s nothing wrong with that – it doesn’t obscure or distort the story, so there’s no reason why I should feel compelled to invest my time and energy learning how to build something less obvious.
Let’s simplify things even further and show the highest-ranking of the two countries by total medal count at each Olympics between 1952 and 1988.
You can see how we’re much closer to the finished product at this point. In fact, from here we’re mostly adding some finishing touches, starting with the narrative: I used annotations at key points to tell the story of how the Soviets entered in 1952 but didn’t immediately beat the USA, though once they did they took a while to beat. I also wanted to point to the boycotts of the 80s, before the Soviets had one last hurrah at the 1988 Games and then the state itself collapsed.
Finally, I fiddled with the formatting for the last ten minutes. I came across a particular difficulty here: I wanted to use the same colour when referencing the Soviet Union in the title as I had for them in the chart, but that was difficult to do as I’m colour blind and I couldn’t figure out a way in time to determine the hex value of that red. In the end I resorted to using this website to get it, although I’m still not sure whether or not there’s an easier way.
And that’s it – my 1-hour #MakeoverMonday. Am I proud of it? Actually, I am – I found a unique story to tell within the data, and I managed to convey that story clearly without devoting a significant amount of time to the effort.
How would you have told this story? How would your choices have differed to mine? And do you believe that the thought I had halfway through – a simple chart is fine, but a less obvious choice would have been better – has substance? Let me know.