By Meghna Amin
I am not a numbers person. I can say that quite easily, as an English student, a journalist, someone who listens to all the numbers and statistics in the world around us and especially in the mass media, and has turned a little numb to it all. That’s exactly what’s happened with the storm of numbers related to the Covid-19 pandemic – the rising cases, the high death rate, how many days into lockdown we are. As the introduction of How to Read Numbers points out, however, I couldn’t really be living and engaging in a Covid-19 world, surrounded by numbers and actually ignore them. Each death is an individual person with millions of stories, but as all those deaths are adding up and all those numbers are taking over, it’s clear that the media is just a numbers game, and it turns out, as this book shows over 22 chapters, that like it or not, we are all numbers people.
Tom and David Chivers (the latter of whom teaches economics at our very own Durham University) have, from the first chapter, kindly set out some “mathsy” terms that I last saw in school in a few very clear sentences. In other words, within a few pages in, this felt like an all-round guide to Economics for Dummies. And then I got to about page 11, where, instead of wondering why I was reading about means/medians/modes through thought experiments, the real-life media examples started explaining the stats behind the headlines.
Headlines work as click-bait, whether that’s through telling the reader that cheese on toast is the nation’s favourite snack or that screen-time before sleep is killing us. Although I’d never really thought about where the truth and facts behind these headlines come from, How to Read Numbers has clarified how all these surprising headlines are maybe not that shocking once we learn the numbers that create them. Whether it’s the fact that Twitter doesn’t actually represent the whole population, or that screen time isn’t actually killing us (but just that 4 hours of e-book reading means around 10 minutes less of sleep), this book really has opened my eyes to what I’m believing and what I’m being told.
One of the most relevant chapters of the book is number 10, ‘Bayes Theorem’. The actual theorem sounds more technical than it is, which I now realise thanks to the very helpful textbook-like boxes that could be ignored dotted throughout this book. However, this chapter actually doesn’t have much about technicalities, and focuses more on realities. Considering when this book was written, in the middle of last year, of course I’d be expecting to read about the numbers that have been taking over our lives: the R number, the death rate, the rising number of cases, the rising number of hospital admissions, and so on. These have also been the numbers dominating headlines across all the national papers nearly every day. And now, although the numbers are in reality as scary as they seem, with the advice of ‘treat them with caution’ and a bit more understanding about the percentages involved, I can admit a tiny bit of relief at the fact that, if nothing else, the complex numbers dominating the media are often more about shock value than accuracy, and that’s not just when it comes to Covid-19.
It is also clearly evident in some of the other headlines, that aren’t pandemic-related. Identifying the difference between causation and correlation seems easy enough, but, with some of the case study examples throughout this book (which, to be fair to the authors, do come from a wide range of well-known publications and not just frowned-upon tabloids), it’s clear that sometimes, the stories with stats are about style over substance. The truth is, no one wants to hear about the boring. So when the media ‘cherry-pick’ (to borrow the authors’ phrase) the numbers they want to base their stories on, they’re doing it so people pick up the page or click the website and read. We’re invited in by the shock, which is enough for the headlines to disguise the truth behind the statistics.
Generally, I may not care too much about weather forecasting, or the statistical models behind YouGov polls that predicted (successfully or unsuccessfully) which party would lose or gain seats in the government elections. However, I do care about what we’re reading and hearing about, and what the news is actually telling us. The last chapter of this book, the conclusion, pretty much summed up all my thoughts and everything I’d learnt over the last 22 chapters (aside from all the maths lessons I’d also learnt by actually reading the optional learning boxes): I am not a numbers person, but that doesn’t mean anything, because I do want to be a ‘numerically responsible journalist’. The 11 bullet points of the very last few pages are the perfect reminder of how to do that, whilst the rest of the book has just left me doubting everything I’ve read in the news before (which I guess is a good thing?).
Image: Meghna Amin