Monday, December 29, 2008

Abstinence pledges worse than ineffective

The Washington Post reports this morning

Teenagers who pledge to remain virgins until marriage are just as likely to have premarital sex as those who do not promise abstinence and are significantly less likely to use condoms and other forms of birth control when they do, according to a study released today.

The findings indicate that pledges are not only ineffective at stopping premarital sex but counter-productive in discouraging teens from taking reasonable precautions.

One explanation for the findings may be that pledges convince teens that they don't need to take precautions. After all, you don't need condoms or other forms of birth control if you're not going to have sex in the first place. However, in the instances when sex does happen, the teens are unprepared. As the study indicates, the "instances" seem to be as frequent for the pledged teens as for the unpledged teens.

Another explanation is that pledges put up barriers between teens and responsible adults who might help arrange for contraception. It is difficult enough for a teen to start the "I need protection" conversation with a parent; imagine how much harder it is when the teen has pledged chastity.

Yet another explanation may be that parents become less vigilant regarding their children's behavior once a pledge has been made.

Pledges and abstinence programs generally are based on the unrealistic notion that we shouldn't send "mixed messages" to teens. However, the mixed message seems unavoidable. We should discourage teenagers from engaging in sex, but if they do, we want them to take precautions against pregnancy and disease.

The good news seems to be that teens as a group are getting the mixed message despite the actions of some adults. Teen birth rates and pregnancy rates are at relatively low levels. More comprehensive programs that recognize the inherent mixed message may help to reduce these rates more.


Anonymous said...

Dave, I'm surprised you missed the obvious correlation!=causation argument. Teens who feel themselves at risk but believe this behavior is unacceptable take the pledge as a way of reducing their risk. It may even work! There is a serious sample selection problem here. I'd love to know if there is any exogenous variation one could use to identify the true effect....

Dave Ribar said...

The author of the study, "Patient Teenagers? A Comparison of the Sexual Behavior of Virginity Pledgers and Matched Nonpledgers," which just appeared in Pediatrics used propensity-score matching (PSM) to address the selection issue.

A criticism of PSM is that it does not address selection from unobservable characteristics, so it can't establish causation.

The inputs into the PSM don't seem to back up your assertion. The pledged teens had more negative attitudes regarding sex; in general, they appeared to be at lower risk for sex.

It doesn't look like there are good exogenous sources of variation in this particular study (just about everything that would affect pledging would also affect sexual activity). However, random-assignment experimental studies of abstinence-only programs also failed to detect positive effects (source: ).

The main difference between the the experimental results and the current study is that the experimental results also failed to find any association with contraception.

Anonymous said...

My mistake for not actually reading the study. I like the experimental design better, and the results make somewhat more sense to me. Just a couple of other random thoughts: (1) What about other instances where people (teenagers or otherwise) "pledge" not to do something -- smoke, for example? Does that work? (2) How would these teen-agers respond to a commitment mechanism like StickK that would put some money on the line?

Dave Ribar said...

We should distinguish between minors, especially very young minors, and adults. An issue here is whether young minors can make these types of commitments. Another issue is whether the "commitments" are being coerced.

For adults, the StickK program seems to have some good results for smoking, drinking, and weight loss. It might be harder to apply to sexual activity (for instance, who acts as the referee?).

Even if the incentives work for some teens, there still is the potential issue of taking precautions for the others.

rknil said...

These studies are great at ignoring reality.

Reality: Teens will be pressured by peers into having sex when they don't want to or are unsure, but they will almost never be pressured by peers into NOT having sex if they have decided to go that route.

This is why these pledges ultimately fail.

However, that does not mean alleged adults with alleged real world experience should stop dissuading teens from having sex. The idea that the approach of responsibility will never succeed/has always failed is one of stunning ignorance.

Remember: The one, true solution is to promote the approach of responsibility. Don't back the approaches of ignorance or the people who support them.

Responsibility -- you can't go wrong with it.

Dave Ribar said...


How do you come to the conclusion that the studies have ignored reality? For instance, some of the studies randomly assigned teenagers to either receive or not receive a particular type of "education" program; the studies then compared the actual subsequent behavior of the teenagers. This is about as real as it gets--try an actual intervention and see if it changes behavior. For better or worse, that behavior would be influenced by all of the things that you mentioned.

The successful programs (and there do seem to be some out there) emphasize responsibility, include ways to deal with peers, but also educate about contraception.

rknil said...

How do you come to the conclusion that the studies have ignored reality?

From reality. (In my reality, I look skeptically upon people who try to pressure people into doing things, especially when those things have no effect on the life of the person applying pressure. I tend to think there are some hidden [or not-so-hidden] consequences.)

Sadly, some (many) people do not share this insight. I call them the anti-responsibility crowd. I have a fairly clear view of their message: They are wrong.

I fail to see how someone taking a pledge that has little chance of succeeding is any form of education. Perhaps this is the point, but it's obscured by the raving lunacy of the anti-responsibility crowd.

This is the ultimate straw-man argument, created by a group of people who are 100 percent wrong about this entire issue. They try to deny being wrong because they don't want to admit their mistakes in life. Instead, they try to pressure people into making similar mistakes. In short, these are not the people we want to have any say in education on this issue.

Until there is a massive overhaul of the attitudes of the anti-responsibility crowd -- and that probably won't happen without a lot of head-pounding and mud-slinging -- it'll be hard to resolve this issue. But the work must begin because right now we have people trying to claim the responsible choice is not the smart one. Their motives are highly questionable, and they must be excluded from this discussion as much as possible.

Remember: The one, true solution is to promote the approach of responsibility. Don't back the approaches of ignorance or the people who support them.

Responsibility -- you can't go wrong with it.

Dave Ribar said...


Declaring everyone else to be "anti-responsibility" doesn't seem to be all that helpful, especially when you haven't clearly indicated how people are supposed to be responsible.

A pledge/purity/abstinence advocate would surely claim that she is emphasizing responsible behavior (don't do something that is going to lead to pregnancy or STDs) and acting responsibly by sending this message.

An advocate of contraceptive education would likewise claim that she is emphasizing responsible behavior among the many teenagers that have, for whatever reason, decided to become sexually active.

I still don't see how studies that have looked at each type of program in real world settings have ignored reality. This seems to be a criticism of the programs more than the studies, unless you're arguing that the studies didn't compare the programs against "real world" alternatives.

rknil said...

The studies are flawed because they focus on disingenuous pledges. I could pledge I'm going to the moon every day, and every day I would fail. Maybe I could get a bunch of people to do the same thing. Using these studies, the results would say we should stop teaching math and science because none of us went to the moon.

I think it's quite clear where I'm going, but the anti-responsibility folks want the rest of society to make the same bad decisions they did. That's why this debate continues to be couched in nonsensical parameters.

I've explained why these pledges fail. Instead of focusing on those causes, though, people keep screaming their heads off about the failures themselves.

We're never going to make any progress on this issue until the focus is on peer pressure. And we have to get to the true motives of the people who oppose these approaches so strongly. I don't see that second part happening any time soon, as people these days lie often for the sake of convenience, so we'll have to stick with that first part.

Remember: The one, true solution is to promote the approach of responsibility. Don't back the approaches of ignorance or the people who support them.

Responsibility -- you can't go wrong with it.