The Washington Post reports the following, under the headline "Abstinence-only programs might work, study says":

Sex education classes that focus on encouraging children to remain abstinent can persuade a significant proportion to delay sexual activity, researchers reported Monday in a landmark study that could have major implications for U.S. efforts to protect young people against unwanted pregnancies and sexually transmitted diseases.

Only about a third of sixth- and seventh-graders who completed an abstinence-focused program started having sex within the next two years, researchers found. Nearly half of the students who attended other classes, including ones that combined information about abstinence and contraception, became sexually active.

The findings are the first clear evidence that an abstinence program could work.

Here's how the study was conducted:

The study released Monday involved 662 African American students from four public middle schools in a city in the Northeastern United States. It was conducted between 2001 and 2004.

Students were randomly assigned to go through one of the following: an eight-hour curriculum that encouraged them to delay having sex; an eight-hour program focused on teaching safe sex; an eight- or 12-hour program that did both; or an eight-hour program focused on teaching them other ways to be healthy, such as eating well and exercising. The abstinence-only portion involved a series of sessions in which instructors talked to students in small groups about their views about abstinence and their knowledge of HIV and other sexually transmitted diseases. They also conducted role-playing exercises and brainstorming sessions designed to correct misconceptions about sex and sexually transmitted diseases, encourage abstinence and offer ways to resist pressure to have sex.

Over the next two years, about 33 percent of the students who went through the abstinence program started having sex, compared with about 52 percent who were taught only safe sex. About 42 percent of the students who went through the comprehensive program started having sex, and about 47 percent of those who learned about other ways to be healthy did.

The abstinence program had no negative effects on condom use, which has been a major criticism of the abstinence approach.

On first blush, this might appear to be a victory for the Focus on the Family types.  But there are a couple of caveats.

First and foremost, the abstinence-only programs were not moralistic.  The kids were not encouraged to "wait until you are married", but merely to "wait until you are ready".   And in fact, as written elsewhere:

Jemmott and colleagues indicated that the abstinence-only program used in the study was unusual. In fact, it would not have qualified for abstinence-only federal funding because it did not rely on moral principles, nor did it criticize condom usage.

So right away, we see that this is hardly the type of abstience-only program touted by the conservative christian block. 

Secondly, the study relied on the self-reporting of ths students themselves.  And I would suggest that a student who went through an abstinence-only program is more likely to lie about his/her failure to abstain, when compared to a student who went through a comprehensive program.

Finally, even the researcher doesn't buy into the notion that abstinence-only is the way to go:

"There are populations that really want an abstinence intervention. They are against telling children about condoms," he said. "This study suggests abstinence programs can be part of the mix of programs that we offer."

That seems to be a direct contradiction of the Washington Post's headline.

Furthermore, the Washington Post glosses over something that medical journals have noted about the study:

Although the abstinence-only program appeared more effective in delaying sexual initiation, it had little or no effect on other sexual behaviors including multiple sex partners, engaging in unprotected sex, and consistency in condom use.

So in other words, with an abstience-only program, about 10% more kids delayed having sex than kids in a more comprehensive program, and when they did so, the program had done nothing to discourage multiple sex partners and unprotected sex.

That's a program that "works"?