Warning lights should always flash before our eyes whenever journalists mix raw numbers and percolating percentages in the same paragraph.
Numbers clearly state actuality. Percentages are the ups and downs of context. Regardless of the axiom attributed to Stalin that one death is a tragedy and a million is statistic, if two people die, two people die. If we talk about a 200 per cent increase in deaths, however, it can sound like a gargantuan increase. But if the original number is two deaths, it means four more people have died, for a total of six. Still tragic, but not quite the fire bell statistical shift the percentage suggests.
As I pointed out to a friend, a recent National Post article by the consistently reliable health reporter Sharon Kirkey is an uncharacteristically characteristic example of the error. It focused on how the so-called Delta variant is throwing a curve ball at pandemic re-opening plans in Canada and elsewhere. Alas, the article itself should probably have come with a “caution sharp curve ahead” sign.
“Britain reported 48,553 new cases on Thursday, while the number of people in hospital rose 42.8 per cent in a week, Bloomberg reports,” Kirkey wrote as part of broader reportage on the pressing global case for vaccination urgency.
There doubtless is such a pressing case. But surely there is a better way to make such a case than arithmetical switcheroo tactics that risk a skewed understanding of the situation. If you track the current raw numbers, for example, you find that the GOV.UK Coronavirus website reports 3,823 in hospital with COVID on July 14 and 4,658 in hospital as of Tuesday – in a U.K. population of 66.65 million.
Of course, every single hospitalization is a concern. No one should minimize any one of them. But the raw numbers present a reality that’s far less fear-inducing than the one-week 42.8 per cent figure, which only obscures in the absence of proper context.
So why not just say so?
Yet resistance to “just saying so” seems to border on becoming embedded in current journalistic practice at least as far as certain issues are concerned. Even the venerable Wall Street Journal let its copy editing guard down Tuesday in a COVID story on the uptick in U.S. caseloads. Reporter Jennifer Calfas writes that new coronavirus cases have averaged “32,287…cases each day over the past week” and, conferring causality, that “Coronavirus-related hospitalizations have also jumped, rising 35.8 per cent between July 7 and July13 compared with the previous seven days….”
Jumped? From which plateau across whose crevasse to what mountain top? Calfas doesn’t say, and the normally crackerjack WSJ editors apparently didn’t think to ask.
The point here is not to engage the COVID vaccination debate. As I’ve written, available data leaves me in no statistical doubt about the public benefits of as many of us as possible “getting the jab” in order to put paid to this infernal pandemic. Nor is the point to point fingers at two hard-working reporters doing their jobs under intense pressures. I cite their stories only as immediate illustrations of a broader shift I perceive at the institutional level of the news generation industry.
Part of that shift is deployment of data to provoke emotion rather than convey information. Alarm-inducing number juggling serves a particular political (or regulatory) end. We can see some of the results in real-time, real-world events. After days of reporting on the perceived dangers of the Delta COVID variant, stock markets were rattled Monday by fears of a renewed pandemic nightmare. The markets appear to have regained their balance fairly rapidly, but still. Fun with figures sent a demonstrable shock through the system.
An excellent editorial in the WSJ makes the case as well for a much deeper threat posed by social media companies bending the knee by using figurative sleight of hand (block that metaphor!) to conform to a State-mandated narrative – whether it be about COVID or anything else.
“For millions of Americans, the increasing over coordination between social-media firms and government is undermining rather than increasing confidence in authorities,” the editorial notes.
Distrust that applies to the tech platforms can reasonably be expected to apply to the remnants of mainstream media as well. The outcome is compounding damage from the failure to provide a full and fair picture of any given circumstance. In the ensuing chain reaction distrust can, tragically, seem like the prudential choice.
It’s not a heart-healthy choice for any populace. At a certain point there’s an inevitability to asking: “If they’re not lying, why can I see their noses growing?” Apart from the congenitally gullible, most people have a tingling sense of when they’re being misled even if they don’t immediately spot the how and why of the misleading. Forceful evidence of this is the instinctive discount we now give to electoral, and even legislative, political language.
Without ever having read a word of George Orwell, we experientially take to heart his words of 75 years ago: “Political language – and with variations this is true of all political parties, from Conservatives to Anarchists – is designed to make lies sound truthful and murder respectable, and to give an appearance of solidity to pure wind.”
We have enjoyed the latitude to operate from Orwell’s dictum precisely because other institutions around us, journalism primary among them, were there to counter the ill effects of political wind. Of course, such institutions had their own interests, their own biases, their own cases to make. But we could at least rely on them to counterbalance each other and, for journalism especially, adhere to the basic principle of not simply drafting in the State’s slipstream.
I myself fear, when I perceive the growing comfort of institutional journalism to follow rather than question, to obscure rather than clarify, that we might have missed the large sign and flashing lights cautioning us: “Road Ahead Washed Out.”
Convivium publishes texts that do not necessarily reflect the views held by Cardus, the Convivium team, or its editors. In the spirit of discussion, dialogue, and debate, we ask readers to bear in mind that publication does not equal endorsement. Thanks for reading. Join the conversation!