From Harriet Coleman
In his review of a book about misunderstanding statistics, Mike Holderness seems to have fallen into a trap himself (15 September, p 57). “Roughly 9 per cent of women suffer breast cancer,” he writes. Then: “So, out of 1000 teetotal women, about 90 will suffer breast cancer.”
In fact what we have is a 9 per cent risk for a total adult female population. That includes unknown percentages of women who drink one glass of wine a day, or a bottle of gin a day, or are teetotal, or get pissed out of their minds only on Saturday nights, or used to but haven’t touched a drop since 1985…
The risk of breast cancer rises “6 per cent for every drink consumed on a daily basis”, we are told. Holderness appears to assume the baseline used is the (unspecified) rate for teetotal women, not the total female population rate, but I wouldn’t count on that without seeing the report.
Wording that personalises the statistic to a rise in “a woman’s” risk of breast cancer is misleading if not meaningless, because every woman, until she starts drinking daily, is somewhere else on the scale from zero drinking to daily drinking, so we don’t know what her risk level was before.
Advertisement
Mike Holderness writes:
• Hoist by my own petard, and by the urge to simplify.
Nice, France
