Curiosity can save us when lies come dressed as numbers
Judging by all the nonsense that people repeat these days, most of us could do with a refresher course in basic number skills. Whether we are confusing millions with billions, the debt with the deficit, or — as in a recent BBC online article — fretting that a currency has lost 180 per cent of its value, we could do better.
One chronic sign of trouble is the lazy cliché about “damned lies and statistics”, which has been used as an excuse for complacent cynicism ever since Mark Twain misattributed it to Benjamin Disraeli. This may be because we do not trust our own statistical intuition. We are more confident in our linguistic skills: the fact that we are surrounded by lies, half-truths and political braggadocio does not make us abandon words. Yet too many of us feel we do not have the skills to tease apart truth and lies if they come to us dressed up in numbers.
That’s a shame. Accurate numbers let us answer some of the most basic questions about any phenomenon. Is it important? Is it getting better, or worse?
It is impossible to make sense of a complex world without some statistical tools in your cognitive toolbox. As Hetan Shah of the Royal Statistical Society is fond of pointing out, it may be possible to lie with statistics, but it is easier to lie without them.
Perhaps we could start by improving the way statistics and data-handling skills are taught in schools. This is easy to say from the sidelines, but not so easy to deliver.
So what might we do? Some people complain that schools focus too much on “engineering maths” — the calculus required for physics and engineering — and not enough on the statistical skills needed for epidemiology, economics, and social science, or for that matter the data-science skills that are in widespread demand.
Perhaps that is true. Yet Bobby Seagull (maths teacher, researcher and FT columnist) tells me that the situation is changing in the UK, with GCSEs expanding with statistical topics and the discipline becoming a compulsory element of A-level maths. That sounds like good news.
Still, there are only so many hours in a school week, so any honest demand to teach more statistics is also implicitly a demand to teach less of something else. For this reason some reformers focus on quality rather than simply a shift in what the curriculum requires.
Sharon Witherspoon, head of policy at the Academy for Social Sciences, argues that we need more and better-trained maths teachers, along with a greater emphasis on using number and data skills in other courses, from biology to geography. Such changes, of course, require serious political will, money, and a change in the incentives within the school system.
Will Moy, director of the fact-checking organisation Full Fact — and ever the pragmatist — told me that universities might be an easier place to start. He suggests a three-day statistical boot camp as a compulsory requirement for graduation. With almost half of young people in the UK going to university, an improvement in the skill and confidence with which undergraduates handle numbers would be no small achievement.
My own experience is that technical skills are only part of the story. Some of the basic attributes required to handle numbers are virtues such as patience and curiosity.
For example, consider the trait psychologists call “cognitive reflection”. A classic test of this ability is the question: “If it takes five machines five minutes to make five widgets, how long would it take 100 machines to make 100 widgets?”
Almost certainly, an answer popped into your head: 100 minutes. But then you probably paused for a moment and worked out the correct answer.
That moment of reflection is often missing when we deal with politically fraught claims in the media, or in our Facebook feeds. If the claim slots into our preconceptions about the world, we accept it and perhaps repeat it. If it challenges us, we reject it instinctively.
We need to train ourselves to stop and think. That isn’t easy, because neither the dark art of political rhetoric nor the context-stripping of social media is conducive to a reflective state of mind.
A second virtue is that of curiosity, which we might think of as a hunger to know more, coupled with a tolerance for being surprised. Simple questions such as: “I wonder how they know that?”; “Is that better or worse than I might have expected?”; “What exactly do they mean?” often unlock far more insight than narrow technical queries.
Of course, there is more to statistics than emotional maturity. Just as everyone benefits from learning to read, we all have something to gain from learning certain basic technical skills in handling numbers. The world would be in a better place if many senior decision makers had something rather more advanced than that.
Yet for all the concern about how we are teaching our children to deal with numbers, perhaps my own teachers weren’t so wide of the mark with their advice. They always used to tell me: stop and think, check your answer, and explain your reasoning. It was wise counsel. Some lessons stand the test of time.
Written for and first published in the Financial Times on 28 September 2018.