Using data to tell the story we want to tell

I love a good headline, but even more, I love probing data “facts” presented by the media who too often tend to search for and favour information in a way that confirms prior beliefs.

Scrolling through the news feed on my phone recently, my attention was grabbed by:

“More women than ever having children over the age of 45”

The Yahoo! article references an Office for National Statistics output which shows that, true, the number of women having children over the age of 45 is going up and is at the highest level since records began in 1938… However, I pondered what the actual story is in the numbers. 

Looking at every year since 1938, I plotted both the ‘total number of births for women over 45’ (blue bars) and the ‘percentage of all births this represents’ (red line) to produce the following chart:

Apart from the baby boomers era post-war (late 1940s), we can observe a pretty good correlation between the number of over 45 births and the percentage of all births. 

I then carried out the same analysis this time for women aged 40 or more.

Again, since the lowest occurrences of the late ‘70s to early ‘80s, we can see an increase in total births and percentages, however for the age range of 40+ the absolute numbers appear to have plateaued.

Comparing the right-hand Y axes of each chart, it’s already much clearer that the proportion of births to women over 45 are of a significantly smaller magnitude.

The graph below provides a much more balanced context for the analysis of the age of mothers in the UK over time and puts the over 45s into a more realistic perspective.

My challenge to the headline is that while technically true , the absolute numbers of women over 45 is so small (less than half a percent of all births are to women over 45), and the increase so tiny (9 more in 2018 than 2017) that it creates a focus on a subset of the population, without providing any context.

In other words, the headline crafts and promotes a narrative that is fairly inconsequential to the entire data set. A sort of confirmation bias (selectively using data to backup an already held belief) that can be fairly damaging to forming our opinions and making good decisions.

At Trym Consulting, we focus on relaying information that passes the ‘So What?’ test, providing information that encompasses the big picture, is the best focus of your energy and attention, and that will inform decision-making.

About the author

Jon’s client work extends across Manufacturers and Retailers with a scope that covers high-level supply chain strategy through to detailed operational reviews and performance improvement. 

We are independent supply chain and warehouse consultants who specialise in data analysis, leading strategy, and bringing a fresh perspective to your supply chain challenges.

© Trym Consulting