How NOT To Use Statistics

We all know that desperate, ignorant, or misguided people can use statistics, as errant theologians sometimes use Scripture, in the way a drunk man uses a streetlamp — for support, not light.

This is especially common, it seems, when addressing social ills. Unfortunately, when it comes to drug policy, true data leads to conclusions that are anything other than true. Religious conservatives, inoculated against critical analysis of information by a culture that disdains the academic while relying on the emotion, often fall prey to this in arguing social policy — with disturbing results. Stats are that part of the arsenal that make the user look solid, unless the one wielding them can’t figure out which end is up.

I experienced this a few months ago while debating with a friend, a 20-year veteran of prison administration, about the legalization of marijuana. Don is foursquare, vehemently, solidly against it. I am, as I’ve made clear, in favor of legalization. I readily admit that he knows more about prisons and criminals than I do, and I know more about . . . well, some things, one of which, in this conversation at least, was the correct use of statistics.

Don pointed out that more than 95 percent of the men serving drug-related sentences in his particular prison, a high-security unit in Western Washington, had used marijuana before moving to cocaine, meth, heroin, and other narcotics. And he’s right — as a matter of fact, I bet the number is higher than that. Point conceded.
Only a very few men serving drug-related sentences haven’t used marijuana; for those who did, marijuana was clearly the “gateway” drug Joe Friday and every cop since then has warned about. His belief, then, is that if 95 percent of the drug criminals he knows of began with pot, then pot is the problem and should be outlawed.

But that point — that an overwhelming number of drug users and drug criminals started with marijuana — says only that. It doesn’t say, not even close, that 95 percent of people who use marijuana go on to use hard drugs. That would be significant, but Don’s stats don’t say that. It sounds ironclad, but it’s an argument from the wrong end and a terrible way to formulate social policy.

Most hard-drug users began their mornings with Rice Krispies, enjoyed baby carrots at lunch, and snuck extra cookies after dinner. Tell me that 95 percent of Snap, Krackle, and Pop fans go on to methamphetamines, and there might be a point to cereal prohibition. But to conclude that the slurping of kiddie cereal is the primary contributing factor to future drug use falls wide of the mark, we can all agree, particularly because it turns out that most cereal-eating kids don’t go on to anything harder than Multi-Grain Cheerios.

True: An enormous percentage of drug criminals started their mornings as children with cereal. False: Cereal consumption inexorably leads to drug usage. True: A tremendously high percentage of drug criminals smoked marijuana before they found heroin. False: A tremendously high percentage of marijuana users become drug criminals. Switch around the nouns, and the argument stands or falls.

Unfortunately, that happens too often, and those rigidly devoted to law and order seize upon stats like those offered by my friend to make false — and societally damaging — conclusions. I was dismayed, frankly, that a man as experienced and intelligent as Don would handle data so clumsily. But arguments like his have prevailed in marijuana legislation for decades. I understand the difference between baby carrots and marijuana, and I would caution anyone who uses weed that it could — could — get out of hand, kind of like beer, brownies, and butter. But I wouldn’t assume that virtually every pot smoker I know will go on to manufacture meth in a trailer or sell coke to schoolkids. I didn’t, and I bet you didn’t, either.

There are myriad social factors that contribute to hard-drug use and addiction. We’d all be better served if legislators focused on those things rather than continue a failed, illogical policy of criminalizing pot. The things that contribute to addiction and criminal behavior are complicated, but aren’t common to everyone who smokes. Responsible pot smokers aren’t contributing to major social dysfunction. They’re breaking the law only because there’s a law in place to break, and it’s time that we acknowledge that all that’s rolled up in a joint is a little bit of dried weed — not the decline of civilization as we know it.

Leave a Reply