Lies, Damned Lies, and Statistics.
It was Mark Twain who famously popularised the saying, “There are 3 kinds of lies: lies, damned lies, and statistics.”
He was not wrong.
Got a weak argument? Find a stat and present it craftily to strengthen your point. Struggling to persuade someone? There’s always a way to use a stat inappropriately to create influence.
Here’s an example:
Your sales manager tells you, “We’ve seen a 100% increase in sales this week.”
You rub your hands together in glee, mentally planning that holiday by the coast you’ve been wanting to take for ages.
But then you realise…
Last week you made one sale, this week you made two.
Statistically, that’s a 100% increase. In real terms, it means you made R20 instead of R10.
You sadly shelve your beach holiday plans. Again.
The truth is, statistics are not as reliable as they used to be. As far back as 2017, an article in The Guardian stated, “The declining authority of statistics – and the experts who analyse them – is at the heart of the crisis that has become known as ‘post-truth’ politics.”
Of course, it’s not just the politicians who manipulate the way they used statistics.
Imagine if I, for example, questioned five CEOs on cyber security. Three of them tell me they need to spend more on keeping their data safe. Here’s how I can dress that up:
“Almost two thirds of South Africa’s CEOs admit to not spending enough on data protection.”
My survey sample was, of course, infinitely too small to generate any kind of reliable data, but that didn’t stop me making a bold declaration that sounds like a cold hard fact.
What is they say about never letting the facts get in the way of a good story…?
So, are statistics just a dressed-up form of lying? And if so, what are the implications for transparency, honesty and integrity?
It’s an important question, and one I’m not sure has a satisfactory answer.
I read a story recently about a woman who’d had a car accident. She wasn’t wearing a seatbelt so was thrown of the vehicle – which was then hit by another car that was unable to stop in time. Had the woman been wearing her seatbelt, she would still have been inside the car and thus crushed in the second accident.
Her conclusion?
It’s safer to not wear a seatbelt.
You may also recall the incident in which Def Leppard drummer Rick Allen lost his left arm in a car crash in Sheffield, England, on New Year’s Eve in 1984.
He tried to overtake another driver but lost control of his vehicle, hitting a wall. His arm became caught in the seat belt and was severed. As a drummer who needed both his arms, he felt he would have been better off not wearing his seatbelt.
And yet statistically, seat belts reduce the risk of death for drivers by around 50%. So the question is: Would Rick Allen have kept his arm but lost his life if he hadn’t been wearing a seatbelt?
We’ll never know, and that’s the issue.
Statistics, when obtained scientifically and objectively by credible methods, should be reliable. And yet their interpretation and application can be anything but objective – as the above two examples illustrate.
I believe the secret lies in the way in which statistical data is obtained and reported.
In any kind of research, transparency is key. It builds trust, fosters collaboration, and ensures methodologies, data and analysis are always accessible for others to verify.
Research scientists, for example, pride themselves on making their findings reliable and reproducible. Openly sharing their data and methods means other scientists can validate their findings. This leads to greater confidence in the scientific community and beyond.
But scientists are very different to politicians, corporate leaders, and even marketing executives, all of whom love to spout statistics whenever possible to shore up their points of view.
So how do we know when to accept the data, and when to view it with more than a healthy dose of scepticism?
The most important thing, I believe, is to examine the source of the information. Where (or who) does it come from, and where does the money trail lead? In other words, who paid for or commissioned the data, and why?
Which leads me back to politicians and other civil servants, many of whom are renowned for flaunting figures that support their policies and agendas.
Our own police force is a case in point.
A statement released to the media in September last year stated that there was a 1.92% decrease in the number of women murdered in South Africa in 2023/24 than in 2022/23.
At face value, that seems like an improvement. An admittedly small one, yes, but an improvement none the less.
Until you realise this still means 966 women were murdered in the first quarter of 2024/25.
I’m pretty sure the families, friends and colleagues of those women take absolutely no comfort in the fact that the year-on-year stats show a decrease in the number of murders. All they understand is that the people entrusted with protecting the citizens of South Africa are failing miserably at their jobs.
Bottom line: always check the source of the statistics and ask whether it’s in their interest to massage the data to show them in a more favourable light.
This is especially important when it comes to social media. What sort of account or page is displaying the statistics? Just because something is shared widely, or with a tone of authority, it doesn’t mean it’s true.
We all know how easy it is for people to use social media platforms to spread ‘fake news’ and false information with just the click of a mouse.
Another very important question to ask is are you getting the full story, or are you just seeing the cherry-picked stats that benefit the person presenting them?
Former UK Prime Minister Boris Johnson was a genius at this. He once, for example, told the press that more people were employed at the start of 2022 than before the pandemic. His stats, however, were only based on the number of payrolled employees. Once you took the total number of people employed into account (such as those who are self-employed), the stats showed a decrease.
It’s easy, in the light of everything I’ve been discussing, to become despondent about the overall reliability of statistics. But I’m not at all advocating we lose our faith in them.
At JGL, we are always on a quest to promote honesty, integrity and transparency in ourselves, our country, and our government. As such, its beholden on us, and every one of you, to work hard to separate the fake from the facts, the truth from the fiction, and to avoid becoming swept away by the raging torrent of data we are presented with every day.
Because if we don’t continually question, check and analyse, we’ll become complicit in the subtle privatisation of the truth.