Statistics never lie, which is why one plus one equals window
In our post-truth world, marketers should stop relying on statistics to prove their worth. Instead, statistics should inform our stories rather than define them, writes David Dunn.
For many people, mathematics is the essence of fact. No matter the complexity, mathematicians pride themselves on being rigorous when solving problems, happy in the knowledge that numbers deliver irrefutable facts and indisputable proofs.
Estimations, approximations and partial truths have no place in mathematics. Unless, of course, we’re talking about statistics.
Statistics deals with “the collection, analysis, interpretation, presentation and organisation of data”. All five verbs used to define statistics are subject to human error, either intentional or unintentional, making a mockery of the idea that numbers never lie.
While statistics are there to help us understand the world, they can often be manipulated, misinterpreted and misrepresented.
Nicely balanced piece. I’m a creative and have just completed three units of stats as part of my studies in psychology. I have an increasing appreciation of how stats can inform the creative process, but I shuddered when I read this;
https://www.theguardian.com/science/2017/feb/01/high-tech-war-on-science
Thank you, and thanks for sharing – scary indeed! Thought I’d return the favour, and share another Guardian article that helped inform this piece. Think it’s particularly relevant given your studies:
https://www.theguardian.com/politics/2017/jan/19/crisis-of-statistics-big-data-democracy
Thanks in return David, missed that one.
Whenever I broach the subject of the replicability crisis with my lecturers, I tend to get a mumbled response. I think this is hitting academia particularly hard, especially when there are those such as Senator Malcolm Roberts who would seize upon such observations for their own questionable ends.
While we’re playing swaps, here’s another;
https://www.theguardian.com/science/2017/jun/27/profitable-business-scientific-publishing-bad-for-science
Excellent article. Well said!
Yes, you can lie with statistics. But it’s a lot easier to lie without them. I love that people use Trump winning the election to prove that statistics are a lie- but look at the lies he tells without them. Personally I think he’d find it harder to lie with them.