To be great data visualizers we must first become great statisticians, examining large amounts of data and presenting it in a factual, yet rhetorical, context. When we are dealing with such a lot of data, on a daily basis, it is only natural that we become a little defensive when we feel our practices are under attack.
A somewhat unclear visualization from The National Review. Image via Twitter.com
“There are three kinds of lies,” Mark Twain quoted in his autobiography, “lies, damned lies, and statistics.” While it is true that figures can be misleading, and, consequently, so can the visualizations we draw from them, we should take this kind of quote with a pinch of salt. After all, it is up to us to ensure that our work is reliable, trustworthy and effective.
In a field in which sharp visuals and slick rhetorical devices are praised almost as highly as the data itself, this can be difficult. We must enter into a balancing act; our visuals must be striking and thought-provoking, but beneath it all they must be factual. Style over substance just ain’t going to cut it.
Power and Responsibility
Misinformation is a hot topic at the moment. Everywhere – from the political caucuses held across the US, via the UK’s referendum on the European Union, to the question of foreign involvement in the Syrian Civil War – it seems that people are either accusing or being accused. People – it would appear – really do not like being misled.
Justifiably so, of course. While most data visualizations carry nowhere near as much political weight as the scenarios outlined above, there is still an element of caution required with every visualization we produce. We receive the statistics, we build a narrative around them, we deliver this narrative to the public; this is a powerful position that should not be taken lightly!
It might be a cliché to say that ‘with great power comes great responsibility,’ but that does not mean that this maxim is not true. As data visualizers, we must act responsibly, giving the public the information they need in an easily digestible fashion, and using this information to paint an enticing, but truthful picture.
The Chinese phrase sān rén chéng hǔ – translating literally into English as “three men make a tiger” – underlines this responsibility. If three men tell you there is a tiger, you believe them, no matter what evidence you receive to the contrary. Misinformation and rumors spread quickly, we must make sure that we are not contributing to this proliferation.
The Narrative Danger
I have written at length on the power of narrative and the importance of storytelling with data. Creating a narrative with data visualization is one of the most thrilling, powerful and satisfying approaches to handling data, but – like Dr Frankenstein presiding over his monster – we would be well advised to take care.
For starters, the data always has to be our primary focus. All visualizations we make must be based upon the facts. We can make those facts look pretty, and we can draw attention to key pieces of insight which casual users may otherwise have missed, but we cannot simply invent a narrative which isn’t there. This includes drawing false parallels between irrelevant data sets, and omitting certain key figures to suit our own agenda.
Similarly, we cannot create scales and comparisons that simply aren’t there. Steering well and truly clear of the political ramifications and judging the visual on merit alone, let’s take a look at a piece of data visualization tweeted by the White House in 2015.
Image via QZ.com
We can see here, by reading the data, that high school graduation rates have increased very slightly during Barack Obama’s time in office – which is a good thing – but what about that scale? Here, the visualizers have simply used their own dimensions, chucking comparative scale out of the window and leaving us with a visual which tells a very different story to the one told by the figures themselves.
Advertising standards, libel laws and other pieces of regulation and legislature do keep us on the straight and narrow for the most part, but what about those rogue visualizers who go out and give the rest of us a bad name? Are they allowed to get away with their manipulated data and spurious conclusions?
The answer is no, at least not in the long run.
When we produce data visualization for public consumption, we are engaging in content marketing, and for this to be successful, it needs to reach the eyes of as many people as possible. Social media accounts are great for this, particularly if you can create a visual which goes viral, but it is good old SEO which is going to be doing much of the heavy lifting for you.
SEO – as you well know – has evolved enormously in recent years. While once upon a time it was all about checking the right boxes, making sure the right keywords were in there, and forging links with other websites, these days it is about much more than this. Nowadays, the big hitters of the search engine world have shown that they are serious about creating a good experience for their users, and this means quality, quality and more quality.
Bad visuals are bad content, pure and simple. Once words gets out that you are publishing skewed visualizations and massaged data, your authority is going to plummet, along with your Page Ranking and your search engine prominence. Data manipulation is not a victimless crime, and its perpetrators get what they deserve.
Authority and Accountability
So what can we do to keep our heads when presented with a tantalizing bit of data? How can we maintain our authority and safeguard ourselves against lapses of judgement? Here are a few tips.
- Don’t work on data visualization alone; have a team who you can bounce ideas off. An extra few pairs of eyes will help you to ensure that the data stays pure.
- Make the source material available. Including a link to the raw figures beneath the visualization ensures that you are accountable and proves that you have nothing to hide.
- Let the data lead you. Creating a story before you have the facts is not science, it is art. Visualizers deal in truth, so make sure that your narrative is data-led, and not the other way around.
Follow these tips, be insightful, stay truthful, and data visualization success is not far behind.
Start Visualizing Your Data! Click here for a free trial.