Why Journalists need to understand statistics – Sensational Listener article about midwifery risks
Play and learning mathematics and statistics
14 October 2016Talking in class: improving discussion in maths and stats classes
3 November 2016The recent article in the Listener highlights again the need for all citizens to be statistically literate. In particular I believe statistical literacy should be a compulsory part of all journalists’ training. I have written before about this. I was happy to see letters to the Editor in the 22 October issue of the Listener condemning the sensationalist cover, which was not supported in the article, and even less supported in the original research. I like the Listener, and subscribe, but this was badly done!
The following was written by a fellow statistician, John Maindonald and published here with his permission.
Midwife led vs Medical led models of care
A just published major observational study, comparing midwife led with medical led models of care has attracted extensive media attention. The front cover of the NZ Listener (October 8) presented the “results” in particularly sensationalist terms (“ALARMING MATERNITY RESEARCH …”).
http://www.listener.co.nz/archive/october-8-2016-2/
Much more alarming is what this sensationalist cover page has made of results that are at an optimistic best suggestive.
Adjustments, inevitably simplistic, were made for 8 factors in which the groups differed. There is, with so many factors operating, no good way to be sure that the inevitably simple forms of adjustment were adequate. Additionally, there will have been differences in mothers’ circumstances that the deprivation index used was too crude to capture. Substance abuse was not taken into consideration.
Here are further links:
http://www.radionz.co.nz/news/national/314359/babies-more-at-risk-in-midwife-led-births-study
http://www.otago.ac.nz/news/news/otago622928.html
(Otago U PR)
https://www.auckland.ac.nz/en/about/news-events-and-notices/news/news-2016/10/another-perspective-on-new-zealands-midwifery-care.html
(mildly skeptical comments)
http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002134
(the paper)
I am disappointed that in its response to criticism of its presentation in Letters to the Editor, the Listener (October 22) continues to defend its reporting.
John Maindonald.
3 Comments
Dr Nic and John
I too was concerned on reading the original article for its sensationalism. But also for their interpretation of the numbers.
The less adept reader may think that the article is saying that the increase is 55% of births rather than 55% of an already low number. Actually the incidence of oxygen deprivation during the delivery is around 4 per 1000 births (depending on your source) – quite low. An increase of 55% takes it to about 6 per 1000 – still quite low. That for neonatal encephalopathy is around 3.5 to 6 per 1000 still a change of only about 2 per 1000 for a 39% increase if the Listener figures are to be believed.
However on rereading the source article I see that the wording is actually different. The source article says:
“They found that mothers with medical-led care compared with midwife-led care had
lower odds of some adverse outcomes for infants. These included 55% lower odds
of oxygen deprivation during the delivery, 39% lower odds of neonatal encephalopathy..”
That is actually quite different and it is lucky the listener mathematicians are not so astute or they might have said the midwives increased the odds by 120% and 60% respectively.
But whichever way it is interpreted, the numbers change infinitesimally as far as a new Mum-to-be is concerned (less than 1 chance in 170 for either problem) but the research should be asking the midwives to look at changes they can make to improve their odds.
Bob J.
Hi Bob
Thanks for that. I too was concerned about the baseline problem. It is such a common problem in reporting – that journalists should be aware of.
In a very real way, tutoring allowed me to do a PhD (in statistics) with a family. Consulting and stipends and grants were all there, but tutoring was pretty big. My major observation is not that most people are not in any way inherently bad at statistics, they’re poorly taught, either in high school (so they’re afraid of all math), or in university. Statistics for non-core students has got to be one of the most poorly taught subjects that exist, at least from what I’ve seen, admittedly in three universities only. Although I’ve talked to many others from all around who agree.
First, we need to make people stop being afraid of stats. Then we need to teach it well.
Otherwise, we will never see the end of this.