Scott Maier’s “Numbers in the News”
April 1, 2009
Perhaps journalists aren’t quite as maths-phobic as they’re portrayed. Among the many interesting snippets from research carried out by Scott Maier from the University of Oregon is the fact that “perhaps mathematical incompetence in the newsroom is not as rife as commentators … contend” (Numbers in the News, Journalism Studies 3 (4), 2002).
He bases his conclusion on a three-month audit of maths-based stories in a US daily regional newspaper. Although he identified 43 types of errors during this period, I was surprised to find that none involved the use of percentage change (in my experience, most reporters struggle with this), until he revealed the reason for this – the paper in question (the News and Observer, North Carolina) had identified a spate of such errors and so had initiated an evidently successful education campaign to eradicate them.
The success prompted Maier to observe: “improving math peformance in the newsroom is not a quixotic quest”. I am quietly relieved!
But back to my point about errors not being as prevalent as commonly feared. Maier observes that many of the errors identified were relatively minor, and that not only were the vast majority of articles error-free, but that numbers often were used in ways that added depth and aided comprehension.
A second (again, surprising) piece of information is the extent to which maths skills (albeit of a basic nature) crop up in general news reporting. Just over a half (53% but with a 4% margin of error at the 95% confidence level) of the stories in the main news section (the A section) of the paper were maths-based. This came as something of an eye-opener to my students, who no doubt had hoped a career in journalism would enable them to evade any further encounter with mathematics.
I remain unconvinced the proportion of stories in UK newspapers involving maths skills is this high – my own entirely unscientific, impromptu “audit” (ie, two minute trawl) of the Liverpool Daily Post only revealed a count of about 15%. Whether this reflects a fundamental difference in reporting styles between the US and the UK, or simply the sketchy nature of my instant audit, is an interesting question.