Proofiness - Charles Seife [19]
If there’s any relationship at all between the mother’s hip-to-waist ratio and the proportion of sons she has—which is rather dubious—there are some other ways it could come about. Perhaps giving birth to boys (who tend to have bigger heads) stretches the pelvic ligaments more than giving birth to girls, leading to a larger post-birth increase in the mother’s hip-to-waist ratio. It’s a bit far-fetched to think that giving birth to boys can alter a woman’s hip-to-waist ratio, but it’s a heck of a lot less far-fetched than the reverse. If the effect is real, then it’s probable that the researchers accidentally mistook cause for effect.
It’s surprisingly common for researchers to do this—to declare that A causes B when in fact it’s more likely that B is causing A. The debt-causes-bad-health study, which appeared in the journal Social Science and Medicine, is a good example, because the researchers had the causality pointer in exactly the wrong direction. When the authors found that the higher a person’s credit card debt, the worse his health, they promptly concluded that bad debt was causing poor health. Perhaps, they argued, the stress associated with financial trouble was causing illness. However, it’s well-known that people in bad health are in worse financial straits than healthy people; they have to pay medical bills, and illnesses can interfere with people’s ability to earn a living. (You can see this connection most clearly with bankruptcies, which are often—more than half the time, according to one study—in families where members had serious medical problems.) The researchers who found the correlation between debt and health apparently forgot to consider that they had cause and effect bass-ackward—and instantly rushed to the wrong conclusion.
Causuistry like this affects public policy, especially in cases where cause and effect are difficult to disentangle. Drug policy in the United States, for example, is a mess, and the government is not above using a little bit of causuistry to scare parents about the dangers of illegal substances. A few years ago, a terrifying advertisement circulated in magazines. Signed by the Office of National Drug Control Policy, the American Psychiatric Association, and a number of other organizations, the ad seemed to speak from authority. “Marijuana and Your Teen’s Mental Health,” it blared. “Depression. Suicidal Thoughts. Schizophrenia.” The ad continued:
Did you know that young people who use marijuana weekly have double the risk of depression later in life? And that teens aged 12 to 17 who smoke marijuana weekly are three times more likely than non-users to have suicidal thoughts?
And if that’s not bad enough, marijuana use has been linked to increased risk for schizophrenia in later years.
The message is clear: if your kids smoke marijuana, they run the risk of going crazy—they’ll catch the reefer madness!
The problem is that the link between drugs and mental illness is a really tough one to understand. It is plausible (and there’s some evidence) that marijuana might trigger schizophrenia in some people, or perhaps exacerbate existing cases. However, there’s a large amount of evidence that people with mental illnesses—schizophrenia, bipolar disorder, depression, and other conditions—turn to drugs (and alcohol) as a means of lessening their symptoms, a phenomenon known as self-medication. However, “schizophrenia might make your kids smoke marijuana” is not nearly as scary as “marijuana might make your kids schizophrenic.” It would never work in an advertisement. For different reasons, neither would the statement “alcohol might make your kids schizophrenic,