Some people claim that the best parents can do is hold their breath when their children become teenagers and exhale when they enter their twenties. The teenage years are notoriously difficult, for both teens and their parents. Many sweet, innocent ten-year-olds transform into surly, moody thirteen-year-olds who would rather do anything than be seen with their parents. Not all adolescents rebel so dramatically, of course, but the teen years are a time of transition as kids go through puberty, become more independent, seek the approval of their peers, and experience increased academic pressures at school. Not many of us would choose to relive those years.
If parents were to list their biggest worries about their teenagers—the reasons they hold their breath—three dangers would make everyone’s list: pregnancy, violence, and substance abuse (including alcohol). Countless promising lives are thrown off track by one or more of these problems, and a great deal of parental and societal efforts go into preventing them. In this chapter we will consider teenage pregnancy, which occurs at a very high rate in the United States. Millions of dollars have been spent on prevention programs that don’t work. Meanwhile, programs that incorporate the story-editing approach have been shown to be effective.
Let’s begin with Amanda Ireland, a 2008 graduate of Gloucester High School in Massachusetts, who has some advice for her younger friends: don’t become a teenage mother. Amanda speaks from experience, because she gave birth to a daughter when she was in the ninth grade. Sometimes classmates came up to her in the hall to tell her how lucky she was, but Ireland set them straight: “It’s hard to feel loved when an infant is screaming to be fed at 3:00 a.m.” Christen Callahan, another teen mother from Gloucester High, concurs. “I don’t think it’s the best idea for anyone, including myself,” she says. “You lose everything.”
But Ireland’s and Callahan’s advice didn’t seem to sink in, if what happened with a group of younger girls at Gloucester High is any indication. In June of 2008 these girls made international headlines for having allegedly formed a pregnancy pact, each vowing to get pregnant as soon as possible. Although school officials deny that there was any such pact, one thing is clear: there was a fourfold increase in pregnancies at the high school over the previous year. Seventeen girls became pregnant, none older than sixteen years of age.
Pundits were quick to point their fingers and give reasons for the spike in pregnancies. Liberals blamed the school for not providing sex education and for banning contraceptives from its health clinic. Conservatives pointed out that Massachusetts had rejected federal funds to provide abstinence-centered education. Others suggested that the girls were imitating teenage celebrities such as Jamie Lynn Spears, who became pregnant at age sixteen (and, coincidentally, gave birth to a baby girl the same week that the Gloucester High story hit the news).1
Whatever the cause, the problem is not limited to Gloucester High. At T. C. Williams High School in Alexandria, Virginia, seventy teenage girls were pregnant or became mothers in the fall of 2008. Nancy Runton, the school nurse, claims that not all these pregnancies were accidental. “I’ve known girls who’ve made ‘I’ll get pregnant if you get pregnant’ pacts,” she says. “It’s a status thing. These girls go around school telling each other how beautiful they look pregnant, how cute their tummies look.” Then reality sets in. Cynthia Quinteros, a T. C. Williams student, gave birth to a boy when she was fifteen. A year later, this is her typical day: she gets up at 6:00 a.m., feeds and dresses her son, Angel, and drops him off at a day care center located in the high school so that she can start classes at 7:50 a.m. After school she plays with Angel and starts her homework, then leaves Angel with her mother while she works as a cashier at a supermarket from 5:00 p.m. to 10:00 p.m. She returns home, finishes her homework, and goes to bed, only to repeat the routine at 6:00 a.m. the next day. “I love Angel,” she says. “But if I didn’t have him I wouldn’t have to work after school, I could study more, I could be a normal teenager.”2
When reading about young mothers such as Amanda Ireland and Cynthia Quinteros, it is hard not to engage in some tongue-clucking. What were these girls thinking? And let’s not forget the fathers of their children. Didn’t these boys and girls know the risks they were taking and the consequences of becoming teenage parents? But before pointing fingers, it might be useful to remember that nearly anyone could have been in their shoes. How many of us have had experiences like this when we were teenagers? We are alone with our first love, experiencing a rush of emotion and passion more intense than we ever thought possible. Maybe we’re in our parents’ empty house, in a deserted park, or in the backseat of a car. We fumble with buttons and clasps and zippers and—well, here is where the story diverges. Some of us put on the brakes, deciding to wait until we were older before having sex. But that was probably a minority of us. According to recent statistics, one-third of teens report that they have had sex by the ninth grade and 65 percent have had sex by the twelfth grade. (If these percentages seem high, it should be noted that they have actually dropped a bit over the past fifteen years.)
Some of us who chose to have sex when we were teenagers did so with careful planning—for example, by purchasing and using condoms. Many of us, however, did not. Today, 38 percent of teens report that they did not use a condom the last time they had sex. We don’t need statistics to tell us how this plays out: a pair of fifteen-year-olds don’t make careful plans before having sex, and when they are alone in a house or in the backseat of a car, they think, “Surely nothing bad will happen—it’s just one time, right?” Lots of teens are lucky and don’t get pregnant or contract a sexually transmitted disease (how many of us, reading these words, just said a silent word of thanks to the gods of reproduction?). But many teens are not so lucky. Today, nearly one in three girls becomes pregnant before the age of twenty in the United States, resulting in more than 400,000 births a year. The rate at which teenagers become pregnant, contract sexually transmitted diseases, and have abortions is much higher in the United States than in most other industrialized nations, including Canada, England, France, Japan, and the Netherlands. The costs are high: teenage mothers are more likely to drop out of high school, to remain single parents, and to live in poverty, and their children are more likely to be victims of abuse, to be placed in foster care, to have below-average general knowledge, math, and reading skills when they enter kindergarten, to drop out of high school, and to go to prison as teenagers or young adults—even after we take into account the socioeconomic status of the mothers and other relevant factors.3
Teenage pregnancy is not a new problem, of course, and for decades many schools have offered sex education to try to reduce it. Depending on your age and where you went to school, you might have attended a single furtive session conducted by an embarrassed gym teacher or several classes on reproduction and birth control as part of a comprehensive family life curriculum. Many localities also have programs to support teen mothers and reduce the likelihood that they will have another child in their teens, because repeat births among teen mothers are disturbingly common. The question is, do these programs work? And what does the story-editing approach have to offer? As we will see, there are some surprisingly effective strategies that can be easily adopted by schools and parents.
The history of sex education in the United States can be traced back to the beginning of the twentieth century, when Victorian taboos against public discussions of sexuality began to change. By the 1920s, more than twenty-five hundred high schools in the United States offered some form of sex education, though the approaches varied widely. In some schools, students were subjected to graphic pictures of the ravages of sexually transmitted diseases, whereas other schools adopted a more quaint approach. (In high schools in Kansas, students took field trips to local jewelers to learn how to shop for wedding rings.) By the 1930s, some colleges offered rudimentary sex education. At Russell Sage College, for example, a women’s college in New York State, the college president and his wife met with students to have frank discussions about family life.4
Sex education was always controversial, but it became even more so in the turbulent 1960s. Mary Steichen Calderone, a physician who was the medical director of Planned Parenthood, pushed for a more open discussion of sex and birth control, rather than an exclusive emphasis on abstinence. In 1964, she succeeded in reversing a policy by the American Medical Association that prohibited doctors from discussing birth control with their patients. Meanwhile, against the backdrop of the 1960s counterculture movement and its advocacy of “free love,” conservative groups became alarmed at changing societal mores about sex and the open dissemination of information about birth control and abortion.
The battle over sex education continues to this day. In 1996, the United States Congress passed Section 510 of Title V of the Social Security Act, which provided money to states for sex education, but only if the approach was abstinence-based. The act defined abstinence education as a program that “has as its exclusive purpose, teaching the social, psychological, and health gains to be realized by abstaining from sexual activity,” and stated that, among other things, “sexual activity outside of the context of marriage is likely to have harmful psychological and physical effects.” (These were the funds that Massachusetts refused to accept, which, according to some, led to the increase in pregnancies at Gloucester High School.) Today, sex education continues to be part of the culture wars in the United States. To liberals, the idea that hormone-intoxicated teenagers will “just say no” when half clothed in the backseat of a car is naive. To conservatives, it is parents—not the schools—who should decide what kind of information teens get about sex. And some are concerned that providing information about contraception will make premarital sex more tempting to teens.5
The politicization of sex education reached absurd heights in the 2008 presidential campaign, when a television ad for John McCain claimed that Barack Obama was in favor of teaching “comprehensive sex education” to kindergarteners. “Learning about sex before learning to read?” the ad intoned. “Barack Obama. Wrong on education. Wrong for your family.” The truth is that when President Obama was an Illinois state senator he supported a bill mandating “age-appropriate” sex education, which would have given kindergarteners information about inappropriate touching and sexual predation, not descriptions of intercourse and offers of free condoms. And there was an opt-out clause allowing parents to exclude their children from the sessions if they wished.6
Instead of trying to score political points, we should use the scientific method to find out what works and what doesn’t when it comes to preventing teenage pregnancies. Yes, people disagree on moral grounds about the appropriateness of premarital sex and birth control. But regardless of where one stands on these issues, it is to everyone’s benefit to find out whether educational programs increase or decrease the likelihood that teens will have sex and whether they prevent unwanted pregnancies. The Obama administration endorsed this approach, making federal funds available for sex education programs that have been “proven effective through rigorous evaluation.”7
Unlike some of the other problems we will encounter in this book, teenage pregnancy has been the subject of rigorous research in recent years. Several well-designed, experimental tests of sex-education programs have been conducted. The reason for this, I suspect, is that because many people view teenage pregnancy and sexually transmitted diseases as medical problems, they are amenable to scientific tests of programs designed to prevent these outcomes. Nonetheless there are many examples of “blistering” in this area (the term I use for programs that common sense tells us should work but that in fact do not; see chapter 1).
Let’s begin with the question of whether abstinence education makes teens less likely to engage in sexual intercourse. A paper published by the Heritage Foundation, a conservative think tank, concludes that it does: “Authentic abstinence programs are… crucial to efforts aimed at reducing unwed childbearing and improving youth well-being.” A careful look at the evidence, however, reveals that this conclusion is at best premature.8
In 1997, researchers examined all published studies that tested the effectiveness of different kinds of pregnancy-prevention programs. They did so using a technique called meta-analysis, which statistically combines the results of individual studies to produce an overall estimate of effectiveness. The results were encouraging: girls who participated in the prevention programs were significantly less likely to get pregnant than girls who did not. When the type of program was examined, however, it was discovered that the programs that discussed and distributed contraception did better than abstinence-based programs.9
The same year this review was published, the United States Congress authorized an evaluation of abstinence-based programs, namely, those that were funded by the law that mandated abstinence-only sex education. One might question whether the cart was put before the horse here—providing fifty million dollars a year for a program and then evaluating it—but in any event, an evaluation was conducted by Mathematica Policy Research, a nonpartisan private research firm. The study used a true experimental design: at four different schools, two thousand adolescents were randomly assigned either to participate in an abstinence program or to a control group that did not participate. The bottom line? None of the programs had a detectable effect on the participants’ likelihood of abstaining from sex or on the age at which they first had sex. For example, on a follow-up survey conducted between forty-two and seventy-eight months after the abstinence programs began, the percentage of teens who said they were virgins was identical among those who took part in the program and those who did not (49 percent in both cases).10
We should acknowledge a weakness of all studies of teenage sexual activity: they rely on teens’ reports about their sexual behavior. Obviously, the researchers cannot follow adolescents into their bedrooms or the backseats of cars and observe what they actually do. Participants may not be completely forthcoming about their sexual escapades, making the effects of the programs hard to evaluate. Researchers attempt to deal with this problem by ensuring that participants’ responses are completely confidential; and, indeed, many teens do report sexual activity on surveys.
Another limitation of the Mathematica study is that it looked at only four abstinence programs. Although the researchers attempted to choose programs that were exemplars of the abstinence approach, perhaps there are other programs that are doing a better job. To provide a broader picture, a different set of researchers conducted interviews with a nationally representative sample of more than seventeen hundred teenagers who ranged in age from fifteen to nineteen. They divided the teens into three groups: those who had never received any sex education (10 percent), those who had received abstinence-only sex education (23 percent), and those who had received comprehensive sex education that included both abstinence messages and information about birth control (68 percent). The results? Teens in the abstinence-only group were no less likely to abstain from sex, and were no less likely to get pregnant, than teens who didn’t receive any sex education. However, teens who had received the comprehensive sex education, which included information about birth control, were somewhat less likely to engage in sex and much less likely to get pregnant.11
An advantage of a national survey is that it includes a representative sample of all Americans, allowing us to generalize beyond a few specific programs. A disadvantage, of course, is that the participants are not randomly assigned to receive one or the other type of sex education. Although the researchers statistically controlled for variables that might influence the results, such as gender, age, race, family makeup, socioeconomic status, and area of residence, we can’t rule out the possibility that unmeasured variables biased the results (as discussed in chapter 2). But in conjunction with the Mathematica study that randomly assigned teens to receive abstinence education or to a control condition that did not, this survey certainly calls into question whether abstinence training has any effect.
To be fair, the absence of a statistically significant effect is a null finding and does not prove that the approach is ineffective (for example, maybe the sample sizes in the studies were too small to detect any benefits). Perhaps the fairest summary of abstinence-based programs comes from an in-depth report issued by the National Campaign to Prevent Teen and Unplanned Pregnancy, a nonprofit group devoted to lowering the rate of teen pregnancy in the United States. This report concluded that “studies find no strong evidence that any particular abstinence program delays the initiation of sex, although they hint that future studies may find a positive effect for one or more programs.”
In fact, one recent study found that a version of an abstinence program was effective. In this study, African American sixth and seventh graders attended two four-hour sessions that stressed the importance of abstinence. The kids also received education about sexually transmitted diseases and training in how to resist peer pressure to have sex. The authors of this study were quick to point out that their intervention would not meet federal guidelines for abstinence education. It was not moralistic or religious, and it stressed that kids should avoid sex until they were ready to handle the consequences, rather than mandating that they wait until marriage. Further, the program did not disparage the use of condoms; instead, it corrected any misconceptions the kids had about condoms’ effectiveness. In a way, then, this intervention was a hybrid of abstinence education and sex education. The mixture was effective: about one-third of the teens who received the intervention reported that they had sex over the next two years, compared to 47 percent of the teens who had been randomly assigned to a control group.12
Many other studies have found that educating teens about birth control prevents pregnancies. Further, fears that giving teens information about birth control will make them hop into bed with their friends are unfounded. There is no evidence that sex-education programs make teens more likely to have sex.
Lameesha Lee, a resident of High Point, North Carolina, gave birth to a healthy baby girl when she was fourteen years old. At an age when many teenage girls are discovering boys, social networking sites, and high school algebra, Lameesha dropped out of school to raise her child on her own (her boyfriend was in prison at that point). As we have seen, having a child in one’s teens puts both the child and the mother at risk for all sorts of bad outcomes. These problems are compounded when teen mothers have a second child, which is not a rare occurrence. In the United States, about 25 percent of teen mothers give birth to another child before reaching the age of twenty.13
Some programs have thus targeted teen mothers in an attempt to prevent second pregnancies. Two such programs, one in North Carolina and another in Colorado, came up with an approach that would be applauded by economists: pay teen mothers not to get pregnant. The North Carolina program, called Dollar-A-Day, was the brainchild of members of the School of Nursing at the University of North Carolina, Greensboro. The Colorado program, also called Dollar-A-Day, was founded by the local chapter of Planned Parenthood. As their names imply, the programs pay teenage mothers a dollar for each day they are not pregnant. There are also weekly meetings at which the young moms can socialize, talk about their problems, and get advice from professionals and other older adults.
Do the programs work? Lameesha Lee, who took part in the North Carolina program, thinks so. “I know I don’t want any more kids for a good long while.” The authors of an article about the North Carolina program also think so: they describe their effort as “a model program for others to emulate.” Jean Workman, a public health educator, agrees: “We have found that those who remain in the program and faithfully attend don’t have second pregnancies. The girls who miss two to four weeks are the ones who get pregnant again.” A more formal evaluation of the North Carolina program reached the same conclusion, finding that ten of the sixty-five girls who took part (15 percent) became pregnant during the first five years of the program, a rate that “was substantially lower than the… 30–35 percent rates reported for other programs.”14
Now even I, a professional skeptic about the ways in which social interventions are evaluated, find this pretty impressive—at first glance. But if we learn nothing else from this book, we should learn to take a closer look at such data. Doing so, unfortunately, leaves considerable doubt about whether these programs have any effect. First, Jean Workman’s observation is a clear case of correlation without causation. The fact that girls who stick with the program don’t get pregnant doesn’t mean that the program is responsible for that desirable result. It could be a classic case of a “third variable”: things that increase the girls’ risk of a second pregnancy (e.g., psychological problems, drug use, their family situation) also make them more likely to drop out of the program; or, put differently, girls who are unlikely to get pregnant in the first place are more likely to stick with the program. The fact that only 15 percent of the girls who took part became pregnant is suggestive, but without a control group we can’t be sure that the program was responsible for this low rate. In fact, it is not clear how low the rate actually is: as mentioned, the rate of second pregnancies appears to be closer to 25 percent than the 30–35 percent the authors of the evaluation used as their comparison point.
Unfortunately, there is further reason to be skeptical of these programs. Researchers conducted an experimental test of the Colorado Dollar-A-Day program and the results were disappointing. They randomly assigned teenage girls who had recently given birth to their first child to one of four conditions: the standard program in which mothers attended weekly support meetings with other teenage mothers and received a dollar for each day they were not pregnant; a group-only condition in which the girls attended group meetings but were not paid (the paid and unpaid mothers attended different meetings); an incentive-only condition in which they received a dollar for each day they were not pregnant but did not attend any meetings; and a control condition in which they did not attend meetings or receive any money.
The incentives worked to increase attendance at the group meetings. In the group-only condition, in which no incentives were offered, only 9 percent of the young mothers invited to participate attended one or more of the meetings. When offered a dollar a day, 58 percent attended one or more meetings. So far so good—the money got the girls (many of them, at least) to show up. But did the meetings have the desired effect of preventing repeat pregnancies? Unfortunately, there is no evidence that they did. Nor is there any evidence that the incentives alone reduced pregnancies. By the end of a two-year follow-up, 39 percent of the girls had become pregnant again, with no significant differences in the percentage of pregnancies between any of the four conditions. It didn’t matter whether the researchers looked at all the mothers randomly assigned to the four conditions or only at those who actually attended the group meetings. Consider, for example, the girls who were randomly assigned to the dollar-a-day-plus-group-meetings condition and who actually attended the meetings. About one-third of these girls became pregnant again, which was not statistically different from the percentage in the control group. The researchers even hint at a negative effect of the group meetings: “Some of the young mothers… tended to reinforce one another’s inconsistent contraceptive behavior and ambivalence about postponing future childbearing.” Bringing together at-risk teens in a way that backfires is a phenomenon we will encounter again (see the next chapter, on adolescent behavioral problems). The verdict: Dollar-A-Day pregnancy prevention programs are an example of blistering, and possibly of bloodletting as well (a commonsense approach that does more harm than good).15
Suppose that school officials in Gloucester, Massachusetts, hired us to help reduce the pregnancy rate, or that officials in North Carolina and Colorado asked for our advice on how to help teen mothers postpone future pregnancies. By now we know that abstinence-based programs are unlikely to work by themselves; nor will offering teen moms a dollar a day to attend weekly meetings have the desired effect. We also know that providing information about contraception will prevent pregnancies, but this is a political hot potato, particularly in a town like Gloucester, which is predominantly Roman Catholic. Even if we could convince the school district and parents to start with comprehensive sex-education classes, what else might we do?
According to the story-editing approach, we should try to change teens’ narratives in a way that will make them less likely to engage in unprotected sex. Well, what sort of narrative places teens at risk for becoming pregnant, or for getting someone pregnant? Research shows that teens who feel disengaged from their school and community and feel alienated and socially excluded are particularly at risk. Now, I don’t mean to imply that unprotected sex is a deliberate, conscious way in which teenagers rebel. Rather, feeling alienated is likely to put kids at risk in many ways, influencing who they hang out with and how much they plan for their futures. Put differently, kids who feel like they have a stake in their communities and have clear goals for the future are less likely to put themselves at risk by having unprotected sex.16
How, then, can we change teens’ narratives from a sense of alienation to one of engagement—from the belief that “I don’t fit in here” to “I’m a valued member of my school and community”? It turns out that an effective approach is to have teenagers engage in regular volunteer work in their community. Now, at first blush this might seem a little hard to believe. What does volunteering at a soup kitchen or a nursing home have to do with teenage pregnancy? What it does is take advantage of the do good, be good principle discussed in chapter 1, namely, the idea that one of the best ways of changing people’s self-views is to change their behavior first. Involving at-risk teens in volunteer work can lead to a beneficial change in how they view themselves, fostering the sense that they are valuable members of the community who have a stake in the future, thereby reducing the likelihood that they engage in risky behaviors, including unprotected sex.
There are at least three well-tested programs showing that this approach works. The first to make this discovery, the Teen Outreach Program, was founded in 1978 by Brenda Hostetler, a St. Louis school administrator. Teen Outreach is typically implemented in high schools in grades nine through twelve and has two components: supervised community service and weekly classroom sessions in which the students discuss their volunteer experiences and issues related to teen development. The students choose their volunteer activity from a wide range of alternatives, including working as aides in nursing homes and hospitals, taking part in walkathons to raise money for charities, and peer tutoring. They are asked to perform a minimum of twenty hours of volunteer service over the course of the year, though most put in many more hours than that.
The classroom sessions involve group discussions, role-playing, guest speakers, and class exercises. In addition to providing an opportunity for students to discuss their volunteer experiences, the meetings focus on such issues as teen relationships, values, communication, peer pressure, goal-setting, and sexuality. As part of the latter discussion, teens receive sex education, including information on contraception and sexually transmitted diseases. However, the researchers involved in testing the program stress that sex education is not the central focus of the program and that it in fact takes up less than 15 percent of the classroom curriculum.17
After its debut in St. Louis, Teen Outreach spread to other cities, with the sponsorship of the Association of Junior Leagues International and the American Association of School Administrators. To the great credit of its founders and sponsors, the program was tested more rigorously than most social interventions, including an experiment in which students at twenty-five sites were randomly assigned to take part in Teen Outreach or to a control group that did not. The results of this experiment were promising: the girls who took part in the program were half as likely to become pregnant as girls in the control group. The program works equally well in preventing first pregnancies as well as second pregnancies in teenage mothers. The program also had positive effects on the kids’ academic performance.
Teen Outreach increases teenagers’ sense of autonomy and connection with adults and peers, especially among young teens. That is, the volunteer work changes the participants’ narratives about themselves, fostering the view that they are valued members of the community. Remember, the teens who are most likely to get pregnant are those who feel disengaged from their school and community and feel alienated and socially excluded. Volunteer work can be an effective remedy for such alienation, making teens feel more engaged and connected.
The astute reader, however, will have noted that Teen Outreach involved community service and a classroom component, making it difficult to tell whether it was the volunteer work per se that was responsible for its success. The classroom component involved sex education about contraception, which we already know can help reduce pregnancy rates. Maybe it was the classroom instruction alone, and not the volunteer work, that was the key to the program’s success.
Fortunately, another program, Reach for Health Community Youth Service (RFH), provides an answer to this question. RFH was implemented in the early 1990s in a New York City middle school that primarily served African American and Latino students. Like Teen Outreach, the main component of the program was community service, in which seventh- and eighth-grade students spent an average of three hours a week doing volunteer work in a variety of community settings, including nursing homes, senior citizen centers, medical clinics, and day care centers. As in Teen Outreach, the students participated in weekly classroom sessions that covered issues about teen development, including sex education. But the program differed from Teen Outreach in that the students in the control condition received the classroom curriculum as well. That is, classes (eighteen in all) were randomly assigned to receive just the classroom component of the program or the classroom component plus the volunteer service, which allows us to assess the added benefits of the volunteer work.
The results were encouraging. First, the researchers looked at the proportion of students who reported that they were virgins before the program and remained abstinent after the program. By the tenth grade—two years after they took part in the RFH program in middle school—50 percent of teens in the community service condition reported that they were still virgins, compared to only 37 percent of teens in the control condition. Second, the researchers asked the teens, when they were tenth graders, whether they had had sex in the previous three months. Forty-one percent of teens in the community service condition reported that they had, compared to 58 percent of teens in the control condition. Both of these results are statistically significant. Thus, engaging in volunteer work delayed the age at which teens had their first sexual experience and reduced the frequency of sex two years after the program had ended.18
Another intervention that incorporated community service was the Quantum Opportunities Program (QOP). In 1989, a pilot program funded by the Ford Foundation targeted at-risk youth in five American cities. Beginning in the ninth grade, and continuing through high school, the kids received 250 hours per year of educational assistance (for example, homework help, peer tutoring) and 250 hours per year of developmental assistance (for example, life/family skills training, help with job applications, planning for college). In addition, the students engaged in 250 hours of community service per year, receiving a small stipend for their participation. The intervention was tested with an experiment, in which teens from families receiving public assistance were randomly assigned to take part in the program or to a control group that did not. The results? Compared to the kids in the control group, those who participated in the program were less likely to have children, more likely to finish high school, and more likely to go to college.19
Given these encouraging results, the Ford Foundation and the United States Department of Labor funded an even larger-scale intervention in seven different American cities. Whereas the pilot program included only 125 students, the full program included nearly six hundred students (between fifty and one hundred teens per site). These teens were randomly selected from a larger pool of eligible participants and took part in the program while attending grades nine through twelve.
Unfortunately, the full program did not replicate the beneficial effects of the pilot study. There were no significant differences between the teens who participated and those randomly assigned to a control condition, either in educational outcomes (e.g., the likelihood of finishing school) or in behavioral outcomes (e.g., the likelihood of becoming a teen parent). In fact, there was some evidence that the program backfired: the teens who took part in QOP were more likely to be arrested or charged with a crime than were teens in the control group, qualifying it for my bloodletting award. Thus, rather than having the intended beneficial effects, the program turned at least some kids into criminals.
What happened? It turns out that each site was given a fair amount of latitude in how they implemented the QOP program, and none of the sites adopted the entire curriculum. In particular, most of the site managers decided to focus on the mentoring aspects of the program and none fully implemented the community service component—the very component that we know, from the Teen Outreach and Reach for Health programs, has beneficial effects! Sadly, more than $15 million was spent on a five-year intervention in which a key ingredient (community service) was eliminated. Even more sadly, the program retained a component—bringing at-risk teens together on a regular basis—that had already been shown to have negative effects, as we will see in chapter 6 (possibly explaining the increase in crime rates in the intervention group).20
The fact that policy makers learned so little from past research—at huge human and financial cost—is made even more mind-boggling by being such a familiar story. Too often, policy makers follow common sense instead of scientific data when deciding how to solve social and behavioral problems. When the well-meaning managers of the QOP sites looked at the curriculum, the community service component probably seemed like a frill compared to bringing kids together for sessions on life development. Makes sense, doesn’t it? But common sense was wrong, as it has been so often before. In the end, it is teens like Amanda Ireland, Christen Callahan, Cynthia Quinteros, and Lameesha Lee who pay the price—teen mothers who might have been helped by relatively inexpensive interventions proven to work.
There are clear lessons here for school administrators who want to reduce teenage pregnancy: abandon abstinence-only sex education (at least the moralistic version that refuses to discuss birth control), implement sex education that includes a discussion of contraception, and encourage teens to engage in volunteer work. If you have children in school, you should find out what kind of sex education and volunteer programs the school offers, and ask what the evidence is for the effectiveness of the programs in use.
But what about you and me in our everyday lives? There are some pretty interesting implications for those of us who are parents of teenagers. The clear message is that we must do what we can to prevent our kids from feeling alienated and disengaged. This isn’t exactly new advice, of course—it’s making it happen that can be hard. Alienation is as much a part of adolescence as acne, as epitomized in the movie The Wild One, in which Marlon Brando’s character, Johnny, the leader of a motorcycle gang, is asked what he is rebelling against. “Whatta ya got?” he responds.
But alienation isn’t inevitable. The key is to make your teen feel like part of a larger whole that is making a difference. One way to do this is to get your teen involved in volunteer work. But ladling soup at a homeless shelter is by no means the only way for a teen to feel connected. Working on a national political campaign can work well, too (witness the passion of young Obama volunteers in 2008 and the fervor of young conservatives in 2010). Teens can also become involved in any number of local civic and political issues. A side benefit is that they will make friends with peers who are also involved in the community, rather than with people like Johnny, the leader of the motorcycle gang (or his modern equivalent).
That’s all well and good, you might say, but what about teenagers who spend all their time in their rooms texting their friends or playing video games and who show no interest in soup kitchens or political campaigns—especially (gasp!) if their parents suggest it? Good question, because the more people feel like they are forced into something, the less likely they are to enjoy doing it and want to keep at it (see chapter 4). Although there are no magical solutions here, there are things parents can do. One is to steer children toward their interests. If they love sports, help them find volunteer opportunities in that area, such as becoming a referee or assistant coach in a league for disadvantaged children. If they love music, they can organize a group to play at nursing homes or day care centers. Another approach is for parents to do volunteer work with their children. We sometimes forget how much we are role models for our kids; they are keen observers, and often learn more from what we do than from what we say.