The math of abstinence education

I'm pretty happy to see that more and more states are refusing federal money for "abstinence only" education. All the evidence says that abstinence education doesn't work, and in fact increases the amount of dangerous teen sexual behavior.

Why does that happen? Well, we can describe the answer with math.

Suppose that the damage done to teens by sex is a function of their level of sexual activity - the more sex partners a kid has and the more frequently a kid has sex, the more harm the kid is likely to experience (from disease, pregnancy, emotional damage, bad reputation, etc.). And suppose the function is convex, like the function pictured here - i.e., each small increase in sexual activity hurts a kid more than the last. Now suppose kids are naturally pretty evenly distributed on this curve - X% of kids don't have sex until marriage, X% have one partner, X% have 2, etc.

So in this setup, abstinence education will force kids to move to one extreme of the distribution or the other - either you listen to your teacher, and you have no sex at all, or you decide abstinence isn't worth it, and you go buck-wild. Because the function is convex, that means that half the kids will experience a minor benefit, and half the kids will be totally and utterly screwed (both literally and figuratively). By leaving no room for a middle ground, abstinence-only education, eve if it works on half of kids, basically condemns the other half to a life of disease, degradation, and teen pregnancy.

Now maybe conservatives disagree with my assumption that the "damage from sex" curve is convex. Maybe they think the opposite - that having sex with one person before marriage is just about as bad as having sex with one hundred. And maybe lots of conservatives do think that. Or maybe they think that abstinence-only education can actually make kids want to have sex less. I have to say, I think the evidence is strongly against both of those viewpoints.

So next time a conservative tells you why abstinence education is a good idea, just ask him if he thinks the harm-sex curve is concave or convex. That'll teach him!


Political science addendum: If you read the article about abstinence-only education funding, you'll find that it's about $204 million, and that "most of it [goes] directly to community organizations." A little research reveals what yo could probably guess - these "community organizations" are mostly Christian religious organizations. Christian religious activists form the main voter and activist base for the Republican party, who in turn pays these activists lots of taxpayer money. This kind of direct money-for-votes exchange is known in political science as clientelism. When practiced on a large scale, it is commonly believed to increase corruption and slow economic growth.