Numbers are tricky.
Some are too big to think about—like the ‘googol’ (10 to the power of 100), which has so many digits (101) that storing it in our heads would hypothetically turn our brains into a black hole. Even for fathomable numbers, our brains are hardwired to process them like sensory information.
Instinctively, but not absolutely—we need some sort of reference point.
(Try explaining how an orange tastes to someone who’s never had one…)
It’s all relative
Talking in terms of percentages, averages and statistical significance is our most basic way of keeping track of how a number relates to other numbers. But is this enough to make things clearer for us?
Not really. Here’s an example.
You have $200 in savings. What would you like your bank to do?
- Increase it by 50% and then decrease it by 50%
- Decrease it by 50% and then increase it by 50%
Had to think about that, didn’t you?
(And—in case you didn’t have time—the answer is ‘Nothing’. Note, the answer changes if you have a larger number. Try the same question with $400 as the starting point—this time, you’re better off taking the first option.)
If our brains can get a bit baffled by dealing with relatively small numbers, how do we make bigger ones meaningful?
You need to create a perspective—build it into a story.
Framing a fact (or fiction…)
Like every other, this story is all about the audience, not the author. The beginning is framed by how much they know, the middle defined by how much they want to know and the end depends on what they are supposed to do with the information.
But even then, communicating the numbers is harder than it sounds. Let’s say our story is a story about risks—that is, the severity of an event’s consequences and how likely they are to happen.
We would like to think that we could frame a rational story around our data and, as responsible decisionmakers, our audience would objectively weigh these numbers up.
We are all susceptible to the framing effect—a cognitive bias that makes us see losses as more important than equivalent gains. This means that if the consequences of a decision seem severe enough, we won’t really care about the likelihood of them happening, even if we see the numbers.
Don’t believe me? Well, there have been:
- 12 shark attacks in Australia this year
- 112 road deaths in Australia in 1 month.
Admit it, you’re still more afraid of swimming at the beach than you are of driving to it.
And even when both the consequences and the likelihood are severe, personal stories (the unresearched kind) trump everything.
(In my experience.)
Big data is watching you
All this ambiguity and, so far, we’re still only talking about manageable data.
Imagine a set of numbers so big, complex and unstructured that normal statistical analysis cannot do it justice. That’s big data, and it increasingly shapes our business strategies—soon we will need to frame stories in an ocean of numbers that is big enough to mean… anything.
But translating those numbers can tell us amazing things about ourselves. In his book, Everybody Lies: What the internet can tell us about who we really are, data statistician Seth Stephens-Davidowitz collected data gleaned from anonymous Google searches to develop profiles of how we think. (Because we always ask Dr Google when we want an answer to life’s problems.) He found these answers in real data—data that varied wildly from the way people responded to surveys in controlled conditions.
A huge undertaking—and with something of that scale, how can we be sure that we’re on the right track? Because, sometimes, it all goes wrong. A program called Google FluTrends was designed to gather search information about flu symptoms and predict illness trends faster than the official Centres for Disease Control. Things started well, but by 2013, FluTrends had missed the mark by as much as 140%.
Obviously, communicating these numbers properly is as much about asking the right questions as it is about giving the right answers.
There’s a company that’s been at the heart of both questions and answers ever since it started in 1996, when 2 computer science students decided that they wanted to organise this new thing called the internet. They made a server network out of second-hand computers in their dorm rooms and created some algorithms for sorting through a mountain of data.
They named it ‘Google’, after the googol.
And they both lived happily ever after.