To my mind, nothing embarrasses a newspaper more than a lack of vigilance over statistics—what I've taken in recent blog posts to calling "numerical illiteracy." One egregious offender is the New York Times.
Consider a recent piece by contributing op-ed writer Ezekiel Emanuel (a brother of our mayor) on sexually transmitted diseases among seniors. He implied that they're rampant. "Combine retirement communities, longer life, unfamiliarity with condoms and Viagra—and what do you get? You get an S.T.D. epidemic among the Social Security generation that rivals what we imagine is happening in those 'Animal House' fraternities."
To prop up this lurid picture, Emanuel offered a "startling statistic": "In 2011 and 2012, 2.2 million [Medicare] beneficiaries received free sexually transmitted disease screenings and counseling sessions."
What makes this number startling? Emanuel said it was "about the same as the number who received free colonoscopies to screen for colon cancer, amounting to about 5 percent of all those on Medicare." But 5 percent means one senior in 20 is taking advantage of screening and counseling offered gratis. That number strikes me as low—especially among a population supposedly being swept by an STD epidemic. The comparison Emanuel made to colonoscopies further undercuts his argument; a colonoscopy is something nobody in his right mind wants anything to do with unless a doctor insists.
To bolster his case, Emanuel passed along statistics from the Centers for Disease Control and Prevention that he said "show rapid increases in S.T.D.'s among older people." Between 2007 and 2011, he reported, "chlamydia infections among Americans 65 and over increased by 31 percent, and syphilis by 52 percent." Emanuel said these numbers paralleled "S.T.D. trends in the 20- to 24-year-old age group, where chlamydia increased by 35 percent and syphilis by 64 percent."
I fault Emanuel less for sensationalizing his story than his editors for letting him get away with it. Tracking his numbers back to the CDC, I found a chart that showed that from 2007 to 2011 the incidence of chlamydia among men aged 20 to 24 rose from 932.1 per 100,000 men to 1,343.3. Among women in that age group it rose from 2,940.4 to 3,722.
Among men over 65, the jump was from 2.6 to 3.3 per 100,000; among women the jump was from 1.8 to 2.1.
As for syphilis, its frequency among 100,000 men 20 to 24 rose from 13.5 in 2007 to 23.4 in 2011, and among women from 2.0 to 3.8. In the 65 and older population, the increase among men was from 0.5 to 0.8. There were too few cases among women even to register as a percentage; the absolute number of cases dropped from 9 to 3.
To stick with absolute numbers, among Americans 20 to 24 there were a total of 545,934 cases of chlamydia and syphilis in 2011; among Americans 65 and older there were 1,202. To paint a picture of these STDs rampaging through both populations not only took bravado on Emanuel's part, it required credulous enablers sitting at desks where skeptical editors belonged.
Earlier in January I marveled on the Reader's blog, the Bleader, at a Times story that claimed black women in Tennessee die of breast cancer at a rate 14 times that of white women (the correct ratio was 1.64 black women per white woman). The mistake came back to bite the Times, which ran a correction a few days later. Other statistics in the original story made it clear where the number 14 came from and what it actually represented. But how could such a preposterous number have reached print in the first place? (I asked the Times's public editor in an e-mail, but she didn't get back to me.)
Where was the editor who should have yanked Seth Stephens-Davidowitz back to reality?
I also wrote earlier this month about a story by contributing op-ed writer Seth Stephens-Davidowitz, who'd relied on some numbers of his own concoction to reach a conclusion that fell apart as soon as it was dabbed with common sense. "My study of anonymous, aggregate data from Google searches suggests that contemporary American parents are far more likely to want their boys smart and their girls skinny," he wrote. "Parents are two and a half times more likely to ask 'Is my son gifted?' than 'Is my daughter gifted?' . . . Parents were more likely to ask about sons rather than daughters on every matter that I tested related to intelligence, including its absence."
Stephens-Davidowitz wrote that he'd put his finger on a clear-cut case of sexual bias. "For every ten U.S. Google queries about girls being gifted," he told us, "there are 25 about boys." Yet the reality is that "at young ages, when parents most often search about possible giftedness, girls have consistently been shown to have larger vocabularies and use more complex sentences. In American schools, girls are 11 percent more likely than boys to be in gifted programs. Despite all this, parents looking around the dinner table appear to see more gifted boys than girls."
Where was the editor who should have yanked Stephens-Davidowitz back to reality? "May I suggest," an appropriately incisive query might have put it, "that what many of your parents see when they look around the dinner table are daughters who have demonstrated all their lives that they're gifted, and sons who haven't shown the first sign of it? No wonder they worry about the boys! If you don't think about this, your 'think piece' could be laughed off the page."
That's what editors do—save writers from themselves. Which leads to this metaphysical question: Does a red flag wave if no one's looking for red flags?