Lies, Damned Lies And Statistics

I missed John Stossel’s ABC special on the War on (Some) Drugs last night, but I heard that he made Asa Hutchinson look like an ass, and it apparently hit a nerve. Bob Weiner has gone after him with a lot of unadulterated Bravo Sierra.

…Stossel pushes his inaccurate points that the drug war ‘creates crime’ when it is precisely the opposite: drug use generates murders, domestic violence, and date rapes. He soft pedals marijuana use, with assertions by an archetypical long haired user that ‘marijuana hasn’t killed anyone,’ but has no one pointing out that marijuana is the second leading cause of car crashes as well as the primary drug in teen drug treatment…

Of course, he provides no evidence that drug use, per se, generates murders or domestic violence. I certainly can’t imagine marijuana doing that.

And I’d wager that most date rapes are caused by alcohol.

As for marijuana being the second leading cause of car crashes, I’m aware of no data on that subject. I think he’s probably just making it up. And as for the teen drug treatment story, there’s no context to describe what such “treatment” might be or why it might be necessary (though it’s often basically a Kafkaesque imprisonment, in which denials of drug use are taken to be proof of drug use).

The drug warriors are getting desperate.

[Update at 9:40 AM PDT]

On the marijuana causing traffic accidents thing, he was apparently repeating an old lie of Barry McCaffrey’s. From Jacob Sullum’s Reason article on the subject:

“Marijuana is now the second-leading cause of car crashes among young people,” McCaffrey wrote in USA Today a couple of years ago. This claim surprised Dale Gieringer, California coordinator of the National Organization for the Reform of Marijuana Laws, who called McCaffrey?s office for the source.

Gieringer was referred to the National Highway Traffic Safety Administration. A NHTSA spokesman confirmed that marijuana is the second most common drug detected after fatal crashes but emphasized that it is not necessarily a cause of those accidents.

As Gieringer noted in his newsletter, a 1990-91 study by NHTSA found that 52 percent of drivers in fatal crashes had alcohol in their blood, compared to 7 percent with traces of marijuana. In analyzing the role that drugs played in the crashes, NHTSA found “no indication that marijuana by itself was a cause of fatal accidents.”

I want Mr. Weiner to write on the chalkboard a thousand times, “Correlation is not causation.”