Did you know that under-age drinkers consume a quarter of the nation's alcohol? Hopefully not, since it's a lie. But on Tuesday, news organizations including CNN, the Associated Press and the "New York Times" were reporting that figure as fact to a shocked (okay, apathetic) nation.
The culprit in this minor hoax was a bogus study called "Teen Tipplers" put out by Columbia University's National Center on Addiction and Substance Abuse (better known as A Bunch of Blowhards, or CASA). Headed by human hairdryer Joseph A. Califano Jr., who has the distinct honor of having served as secretary of health, education and welfare under Jimmy Carter, CASA has long been noted for its opposition to anything not healthy, educational, or… welfarish.
Now, however, it can be noted not just for twisting facts and crying wolf over relatively minor problems, but for plain ol' makin' it up as it goes along. You see, though people manipulate data all the time in public policy debates -- and 99 out of 100 Washingtonians would be unemployed if they weren't pushing idiotic causes -- seldom does anyone push things as far as CASA did this week.
While it almost never happens, CASA actually managed to lie so boldly that they got caught publicly. To its credit, the "New York Times" figured out that something was up with the CASA study (it more than doubled the federal government's estimate of under-age drinking -- while using the same data!), and went on the offensive Wednesday to debunk the junk science.
As "Times" reporter Tamar Lewin summarized the problem: "The Columbia center said it had derived the data from the Household Survey on Drug Abuse, a yearly poll of 25,500 people, conducted by the Substance Abuse and Mental Health Services Administration.…That survey includes nearly 10,000 people age 12 to 20, an oversampling intended to ensure that there would be enough data from young people.…So young people made up almost 40 percent of the survey, although they make up less than 20 percent of the population. In estimating their share of alcohol consumption, the center did not adjust the data to account for the oversampling."
"It's very unfortunate," Sue Foster, the center's vice president and director of policy research, told the "Times." "We didn't reweight the data."
The eloquent Ms. Foster tries to make the whole thing sound like a wacky mix up à la Three's Company ("Oh, no! Don't let Mr. Roper find out you didn't reweight his data, Jack!"), but what it is really is fraud. No one just forgets to reweight data unless they're a high school student taking their first stats class.
It's all okay though, the center's officials tell the press, because the Household Survey probably has a significant problem with underreporting anyway. Apparently, if you don't like the data you come up with the first time around, you can just keep adding numbers until you do like the data. Someone should tell Al Gore.
Of course, this was just an ugly, silly little incident. Most Americans probably heard that minors consume a quarter of all alcohol and slowly drifted into revelries about Spring Breaks past. But the larger issue highlighted by a story like this is the press's and the public's general inability to analyze statistical and scientific data -- and mainly, here, the burden must lie on the press. While reporters certainly aren't expected to be scientists, it's their job to do at least some basic filtering and make sure that the public doesn't get its collective chain yanked.
Of course sometimes it's the reporters doing the yanking. Recently the press displayed willful statistical and scientific illiteracy with the Summer of the Shark ballyhoo. Statistically, there weren't any more shark attacks than any other year, but the networks needed to fill up the void left by a waning Chandra Levy story.
But even aside from simple sensationalism, the press routinely regurgitates factoids thrown out by public interest groups in relation to important debates without asking basic questions.
Do 12 children die every day from gunfire? Yes, but only if you include people aged 15-19 -- many of whom are in gangs and are involved in drug-related crime -- as children. Otherwise the number is closer to 1.7 a day, or 1.3 excluding suicides.
Did Joe Camel influence kids to smoke? Only if kids having been able to match a product to its spokesman (spokescamel?) at age six conclusively proves that they were destined to go on to smoke ten years down the line.
Studies such as CASA's whopper this Tuesday hit the street everyday. Busybodies who care way too much about this, or an unhealthy amount about that, peddle their wares day in and day out hoping to land language in a bill, a new government program, a huge foundation grant, or a mention in the State of the Union.
For the past few years, the press's policy has been to print first, ask questions later. This time though, they got burned -- and they deserved it. It may be overly optimistic to think that editors and reporters have learned a lesson and will ask common sense questions about the studies they get press releases about in the future, but they just might. After all, 61.3 percent of all journalists have at least a modicum of common sense. Unless, of course, they've been over-reporting.