Wisconsin Youth Futures Technical Report #11
Youth At Risk For Early Sexual Activity and Teenage Parenthood

University of Wisconsin-Madison/Extension,Cooperative Extension - 1994

Stephen Small
Dept. of Child and Family Studies
University of Wisconsin-Madison
1330 Linden Drive
Madison, WI 53706

Karen Bogenschneider
Dept. of Child and Family Studies
University of Wisconsin-Madison
1300 Linden Drive
Madison, WI 53706

Keynote presentation made at the conference, "The Teen Sexuality Challenge...Bridging the Gap Between Research and Action," Green Bay, WI, October 29, 1991.


Current research indicates that fifty-three percent of females between the ages of 15 and 19 are sexually experienced. Teenagers at the end of the 1980s were more likely to be sexually active than teenagers at the beginning of the decade despite the advent of programs to encourage sexual abstinence during this same time period. Moreover, each year approximately one out of ten teenage girls between the ages of 15 and 19 become pregnant. Given the high rates of teenage pregnancies, an important question that needs to be addressed is: What factors put teens at risk for teenage pregnancy? The purpose of this presentation is to review what is known about factors related to sexual activity among teenagers. We propose that a risk-focused ecological perspective is useful for understanding teenage sexuality. Characteristics of the teens and of their families are predictive of which teens are sexually active. In addition, factors outside the family such as schools, work, and neighborhood characteristics can influence rates of sexual activity among teems. The ecological perspective is not only useful for thinking about the problem of teenage pregnancy, it is also useful for thinking about how the problem can be addressed.


KAREN: The question we will begin with today is "Do we know enough to prevent early adolescent sexual activity? Today, we'd like to talk with you about a way of thinking about prevention. David Hawkins and his colleagues at the University of Washington began reviewing other fields for effective prevention techniques. The heart and lung disease prevention models that were being pioneered at Stanford and other places across the country seemed to hold promise. Their approach differed from other approaches in that if focused on factors that increase the risk of heart disease - a family history of heart disease, too much smoking, too little exercise, a diet high in fat.

What was interesting about this approach is that studies indicated that people were able to prevent heart disease by reducing the risk factors - by focusing on preventing smoking, by helping people understand the importance of exercise and diet, and by promoting lifestyle changes. The core of the risk-focused prevention approach is quite simple. To prevent a problem from happening in the first place, identify the factors that increase the risk of that problem and then address those factors - either eliminate them or reduce their effects. Or identify factors that protect against that problem and support or enhance those factors.

STEVE: This leads to the question, "Will this approach that began in the health arena also work in human development?" The clear message from the literature on child development is that human development is not influenced by one factor buy by a whole mosaic of factors (Bronfenbrenner, 1986; Segal, 1983). Children grow up, not in isolation, but in ever-widening environments. Children are influenced first and foremost by their family (see Figure 1) but also by their peers, their school and work settings, and the community in which they live. Development has no single cause; rather multiple factors working together influence development.

Figure 1: The Ecological Model

If we are serious about preventing teen pregnancy and delaying early adolescent sexual activity, this model suggests the need to address risk factors at multiple levels of the child's ecology.

Figure 2: Definition of Risk Factors

Now, what do we mean by risk factors? Risk factors are individual or environmental hazards that increase a youngster's vulnerability to negative developmental outcomes or problems. The presence of risk factors does not guarantee a negative developmental outcome, but rather increases the odds or probabilities that a problem will occur (Werner, 1990).

KAREN: Even in the face of overwhelming odds, some children exhibit a remarkable degree of resilience which leads to the question, "What is right about these children? What protects them?" (Werner, 1990).

Figure 3: Definition of Protective Factors

Protective factors are individual or environmental safeguards that enhance a youngster's ability to resist stressful life events and hazards and promote adaptation and competence (Garmezy, 1983, Werner, 1990). Protective factors are sometimes the opposite of risk factors; the main difference, however, is that risk factors lead directly to disorder while protective factors operate only when a risk is present (Rutter, 1987).

STEVE: An ecological, risk-focused approach also assumes that risk factors are cumulative and sometimes multiplicative. The more risks individuals face, the greater the likelihood that they will experience a problem. Similarly, the more protective factors individuals possess, the less likely they are to experience a problem when faced with stressful events or hazardous conditions.

Finally, an ecological, risk-focused approach to teen sexuality and pregnancy prevention assumes that because the causes of these behaviors are complex, so are the solutions. It assumes that there are no magic bullets, in spite of the fact that politicians, program planners and citizens tend to look for single factors, magic bullets and quick solutions to complex problems. The piecemeal, band-aid approaches that result usually focus on a single risk factor to the exclusion of other know risk factors. Such approaches are usually not very effective.


We will now review the literature on risk and protective factors related to teenage sexual activity and teenage pregnancy. Most of our review ill focus on factors related to early onset of sexual intercourse among teens, since being sexually active is a necessary condition for becoming pregnant or contributing to one's partner becoming pregnant. Given our limited amount of time and our interest in the development of prevention programs, we will give most of our attention to those risk factors and protective factors that we are in a position to do something about.

STEVE: We will consider risk and protective factors at several levels of the human ecology. We will start by examining characteristics of teens themselves that are associated with sexual activity. Next we will consider characteristics of the family that are related to whether or not teens are sexually active. Finally, we will consider factors outside the family, such as the school, peer group, and neighborhood, that are predictive of which teens are sexually active. For each level of influence we will give some illustrations using data from the Teen Assessment Project, or TAP, a program I know many of you have worked on along with me.

It is important to note that we cannot yet be certain that each of the identified risk factors is a cause of early sexual activity; for some of them all we know at this point is that they are correlated with the incidence of the behaviors.



Let's begin with risk factors at the individual level. A number of characteristics of the teens themselves have been linked to whether or not they are sexually active. Obviously age is important; as you can see in this slide, the older the teen, the more likely it is that they are sexually experienced (Hayes, 1987). Boys are more likely than girls to be sexually experienced at each age; however, by the late teens there are few differences between boys and girls in rates of sexual activity. Thus, as teens get older, they are increasingly likely to emulate adult sexual behavior.

Figure 4: Age and Sexual Activity

Now let's look at TAP data regarding contraceptive use as a function of age and gender; contraceptive use increases with age in both males and females. It is worth noting that younger teens are less likely to use contraceptives (Brooks-Gunn & Furstenberg, 1989). If a girl is under 15 when she first has intercourse she is nearly twice as likely to become pregnant in the first 6 months of intercourse than if she is over age 17 (CSSP, 1986).

These age differences do vary by race. African-American adolescents initiate intercourse at younger ages than whites for both males and females, although the difference between AfricanAmerican and whites has narrowed significantly over the past ten years. One explanation for the race difference is that AfricanAmericans report a greater tolerance for sexual activity outside marriage than whites; it is not clear whether these race differences are basic value differences or attitudinal adjustments to harsh living conditions (Hayes, 1987).

Figure 5: Contraceptive Use by Age


Another factor that has consistently been linked to an adolescent's decision to become sexually active is school performance. Teens who are getting low grades or who have been retained in a grade are more likely to be sexually active. In contrast, teens who are having success in school are likely to have higher educational aspirations; high educational aspirations, in turn, are associated with delaying the time of first intercourse (Flick, 1986; Dryfoos, 1990; Rosenbaum & Kandel, 1990). I looked at the relationship between grade point average and rates of sexual activity among high school students. As you can see, teens with higher GPA's are less likely to be sexually active.

Figure 6: Sexual Activity by GPA


A third individual level factor that may be important is a history of sexual abuse, at least in the case of females. Those who have worked with teenage mothers have reported for some time now that a disproportionate number of teenage mothers have been sexually abused. A study that came out last year corroborates what those in the field have been reporting. Over half of a sample of teen mothers (N=41) had been sexually abused at some point prior to age 18 (Butler & Burton, 1990). This leads to the question of how many female adolescents are sexually abused. Nationally, it is estimated that about one in four females (25%) have been sexually abused by age 18 (Finklehor, 1986).

Steve used the Wisconsin TAP data to examine the relation between sexual activity and a history of sexual abuse.

Figure 7: Sexual Activity by Sexual Abuse

This figure shows that teens who have been sexually abused are 3 times more likely to be sexually active than teens who have not been abused.

This next figure shows that nearly the same relationship holds for teens who have been physically abused.

Figure 8: Physical Abuse and Sexual Activity

Clearly, a history of physical and sexual abuse significantly increases the likelihood that an adolescent will be sexually active.


There is some evidence to suggest that teenagers who are sexually active prior to age 16 tend to engage in other unconventional behavior such as using alcohol, marijuana, or other drugs more than their peers. This may reflect the fact that these teens tend to be risk-takers generally (Flick, 1986; Hayes, 1987). It also may be that the use of alcohol or other substances may reduce inhibitions, thus making sexual activity more likely. Clearly, substance use and sexual activity are related; recent data suggest that cigarette smoking and alcohol use tend to begin before sexual activity, especially among girls (Rosenbaum & Kandel, 1990). (Or, as many old-time movies would have us believe, sex leads to smoking.)

For purposes of illustration, I computed the frequency of alcohol consumption and related that to the probability that a teen was sexually experienced.

Figure 9: Alcohol and Sexual Activity

This next graph shows the percentage of 10th graders who are sexually active as a function of alcohol use. (We are controlling for age in this graph because we know that both alcohol use and sexual activity are correlated with age.) As the frequency of alcohol use increases, so does the likelihood that a teen is also sexually active. Only 10% (or 1 in 10) of the 10th graders who do not drink are sexually active while nearly three-quarters (or 7 out of 10) 10th graders who drink several times a week are sexually active.


Let's turn now to characteristics of the individual that might serve as protective factors.


One factor that has been studied is self-esteem. Some have argued that teens who are high in self-esteem may be better able to resist pressures to be sexually active. The evidence however, has been mixed. For some samples of female teens it has been a protective factor, for other samples it has been unrelated.(1)

(1) A significant relationship has been found in two small studies based on selected samples (Chilman, 1980), yet a recent analysis by Mott (1983) using the 1982 wave of the NLSY shows no significant relationship. Rosenbaum & Kandel (1991) found self-esteem and lack of control to be inversely related to early sexual experimentation, but only for girls.

For boys self-esteem has either been found to be unrelated to sexual activity or positively related to sexual activity. In other words, some studies have actually found that self-esteem of boys is higher if they are sexually active.

Figure 10: Self-esteem and Sexual Activity in Females

In the TAP data we did not find a relationship between self-esteem and sexual activity for boys, but we did for girls. The next graph shows the percentage of girls who are sexually active as a function of their level of self-esteem. Of the high self-esteem girls, only 15 percent were sexually active while for low self-esteem girls, 29 percent were. These groups represent the bottom and top quartile of the sample (i.e., the top and bottom 25%). Stated a little differently, girls with low self-esteem were twice as likely to be sexually active than girls with high self-esteem.

Closely related to self-esteem is feelings of self-efficacy. By self-efficacy, we mean adolescents who perceive that they have a large measure of control over their lives and that they are competent and capable of choosing and shaping their destinies. These teens are less apt to be sexually active, more likely to use birth control, if they are, and less apt to have an out-of-wedlock birth than adolescents who feel that their lives are largely beyond their control (Hanson, Myers & Ginsburg, 1987; Hayes, 1987).


Several studies have suggested that self-perceptions, not selfesteem, are at the heart of sexual decision-making. Selfperception is a sense of who one is, who one can be, and who one wants to be (Hayes, 1987).


Closely related are future aspirations. Adolescents who express low career aspirations have consistently been found to have an earlier mean age at first intercourse than peers with higher aspirations. Generally, these results have been similar for both males and females. According to Marian Wright Edleman of the Children's Defense Fund, the best contraceptive is a real future. An important aspect of self-perception among teenagers is their educational, occupational, and family formation expectations. If the economic future appears bleak, there may be little incentive to delay sexual activity or childbearing. On the other hand, if teens perceive good reasons for delaying parenthood, they are less likely to be sexually active and more likely to use contraception if they become sexually active (Adams, Adams-Taylor & Pittman, 1989; Dash, 1989). Thus, we would consider a positive outlook on the future to be a protective factor.

Although we have labeled this an individual factor, it could also be thought of as a community factor. Teens living in impoverished inner city areas with high rates of unemployment are at risk for becoming teenage parents; this may be due to the teen's accurate perception that they have fewer economic opportunities than adolescents in more affluent communities (Wilson, 1987).

Figure 11: Sexual Activity Rates by Future Plans


KAREN: Having looked at characteristics of the teens that are predictive of sexual experience, let's turn to characteristics of their families. What family factors are predictive of sexual experience in teens?

PARENTAL MONITORING - The first family factor we will consider is parental monitoring. How closely do the parents monitor the activities of the teens? By monitoring, we mean knowing where your kids are, who they are with, and what they are doing.

To illustrate this point, consider the TAP findings from one Wisconsin county. As one can see in the next figure, teens who are not closely monitored are at greater risk for early sexual experience than their peers (Flick, 1986; Brooks-Gunn & Furstenberg, 1989; Hayes, 1987; Hanson, Myers & Ginsburg, 1987; Small & Silverberg, 1991). Half the teens who are in the low parental monitoring group (the bottom 25% of the sample) are sexually active, while only 11%, or about 1 in 10, of the teens in the high parental monitoring group are sexually active. Parental monitoring is one of the most powerful family variables we have looked at for early sexual activity and other problem behaviors as well. The effect of parental monitoring appears to be stronger for younger teens than for older teens.

Figure 12: Parental Monitoring and Sexual Activity

A John Hopkins survey found that the living room couch has long since displaced the back seat of the car or the beach as the place of intercourse (Dryfoos, 1990); nonmarital intercourse is most likely to occur at the home of the young woman or her partner (Chilman, 1982; Hanson, Myers & Ginsburg, 1987).


A second family risk factor is low socioeconomic status. Families with few financial resources and/or low levels of parental education have teens who are at greater risk for early sexual activity and earlier timing of childbearing. The process by which socioeconomic status affects sexual activity is not clear. The effect of SES may be largely indirect, through such factors as success in school or outlook on the future.

For purposes of illustration, I examined the relation between rates of sexual activity and level of father's education, which is a proxy for socioeconomic status or SES. Notice that the teens from families where fathers have less education are more likely to be sexually active than teens who come from families where fathers have more education. The relationship is pretty linear. The higher the father's education, the less likely the teen is to be sexually active.

Figure 13: Father's Education and Rates of Sexual Activity


A third family factor is family structure. Teens are more likely to be sexually active at an earlier age if they are from non-intact or female-headed households (Dryfoos, 1990; Hayes, 1987; Higgins, 1988; CSSP, 1986; Rosenbaum & Kandel, 1990); also females are more likely to be sexually active if their mothers or siblings were teen mothers (Dryfoos, 1990).

Figure 14: Family Structure and Sexual Activity

This graph shows the relationship between sexual activity and family structure. Teens who are in two-parent, biological families are the least likely to be sexually active; followed by teens whose parents are divorced but have a shared custody arrangement; followed by single parent, mother-based households; followed by step families. As you can see, teens who live in single parent, father-headed households are the most likely to be sexually active.(2)

(2) There are a number of possible explanations for this: 1) Single fathers may do poorer job monitoring their teens; 2) Teens who are engaging in various risky behaviors are more likely to be living with their fathers. Often the mother, who originally had custody of the child, had difficulty controlling him or her. As a result, custody has shifted to the father.

It should be noted that while females whose mothers are single and do not date are no more likely than girls in two-parent families to engage in early sex, girls whose mothers have either remarried or who date frequently are more likely to engage in early sex (CSSP, 1986).


STEVE: Families may also serve a protective function, decreasing the likelihood that teens are sexually active.

FAMILY CONNECTEDNESS AND SUPPORT - The literature suggests that one protective factor may be parental support, although the findings are mixed. Some authors conclude that parent/child communication does not appear to affect teens' sexual behavior. In other studies, however, if a teen has a positive relationship and is able to communicate with his or her parents, he or she is more likely to delay the onset of sexual experience (Brooks-Gunn & Furstenberg, 1989; Dryfoos, 1990; Hayes, 1987). Mother's communication with sons and daughters was associated with later intercourse in one study, but son's discussion with father was linked to early sexual activity, suggesting that fathers may be encouraging their sons to be sexually active (Brooks-Gunn & Furstenberg, 1989; Dryfoos, 1990). A study by Moore, Peterson & Furstenberg (1986) provides some insight into what might be going on. In families where parent-daughter communication was high and parents had traditional values, teenage daughters were less likely to be sexually active compared to families where parent-teen communication was good but parents had more liberal values. This suggests that whether or not parent-child communication inhibits sexual activity by teens is dependent on what gets communicated.

Figure 15: Parental Support and Sexual Activity

In the TAP data, parental support was a protective factor for girls but no boys. Females who reported that they had a warm and supportive relationship with their parents were less likely to be sexually active than their peers who did not report such a close relationship. Females who had a supportive relationship with their parents were about 13% less likely to be sexually active.


KAREN: Families are influential, but they are far from being the only influences on the behavior of their adolescents. Factors outside the family are also important and we would like to note some of these factors.

PEER INFLUENCE - Peers take on increasing importance as children enter the teen years. In terms of sexual activity, some studies have shown that adolescents, especially white girls, are more likely to be sexually active if they believe rates of sexual activity are high among their peers. Sexual activity is more closely related to what teenagers believe friends do than what is actually going on (Dryfoos, 1990). Unfortunately, teens often overestimate the number of their peers who are sexually active (Hayes, 1987). Male partners also exert considerable pressure on teenage females. As you can see in this slide, the pressure to be sexually active peaks in about ninth grade and is higher for females than males.

Figure 16: Pressure to be Sexually Active

The next graph shows a similar peak in the ninth grade for having sex as a way to get or keep a boyfriend or girlfriend. It is worth noting that females are significantly more likely than males to believe that having sex is a way to maintain or initiate a relationship.

Figure 17: Have Sex to Get a Boy/Girlfriend

STEVE: Teenage females are more likely to be sexually active if they are in a committed relationship (e.g., going steady) and if they have had a relatively long-term relationship with a partner. Not surprisingly, girls with boyfriends are 9 times more likely to be sexually active. Notice that very few girls who do not have a boyfriend are sexually active. For boys, being in a steady relationship with a girl is much less important.

Figure 18: Sexual Activity Status and Whether or Not Teen Has a Boyfriend or Girlfriend

The typical pattern has been that partners gradually advance their degree of physical intimacy over time. However, there is some evidence that the process is less gradual now than in years past (Brooks-Gunn & Furstenberg, 1989; Flick, 1986; Hayes, 1987).

This next graph shows the relationship between sexual activity rate and the amount of time a teen spends with their boyfriend or girlfriend. As you can see, the more time that is spent together, the greater the likelihood the couple have sexual relations.

Figure 19: Sexual Activity Status and Time Spent with Boyfriend or Girlfriend

EARLY DATING - Finally, early dating is also linked to early intercourse according to Brooks-Gunn & Furstenberg (1989). Early dating may be related to early physical maturation, which is also linked to early experimentation, especially among males (BrooksGunn & Furstenberg, 1989).


KAREN: ACADEMIC FAILURE - Academic failure increases the risk of early sexual experience among both blacks and whites (Brooks-Gunn & Furstenberg, 1989; Dryfoos, 1990; Higgins, 1988). Conversely, those who score high on intelligence tests, are academically motivated, and are doing well in school are less likely to initiate sexual activity at a young age.

POSITIVE SCHOOL EXPERIENCES - Positive experiences in school, broadly defined, may reduce the risk of unplanned pregnancies (Brooks-Gunn & Furstenberg, 1989; Quinton & Rutter, 1988; Rosenbaum & Kandel, 1990). In other words, schools may serve as protective factors if they provide teens with positive experiences. These positive experiences can stem from academic pursuits, social success, a special relationship with a teacher, the opportunity to take positions of leadership, or success in nonacademic pursuits such as sports, music, or art.

Steve examined the relation between the teenagers' degree of school attachment and rates of sexual activity.

Figure 20: School Attachment and Sexual Activity

Teens who have a low attachment to school are about 3 times more likely to be sexually active than those who are highly attached to school. Students who are highly attached to school enjoy going to school, find it meaningful, and are pretty certain that they will graduate.

Programs that are aimed at helping children achieve school success may be particularly important in reducing rates of teenage pregnancies (Dash, 1989). For example, the Perry Preschool Study in Ypsilanti showed that females who received a high quality preschool program had fewer children at age 19 than females in a control group (Berrueta-Clement et al., 1984).

STEVE: THE CHURCH - Another potentially important institution is the church. Some studies have shown that teens who attend church regularly are less likely to be sexually active than those who do not attend or seldom attend religious services. These studies generally show that commitment to religion is more important than religious affiliation (Hayes, 1987; Rosenbaum & Kandel, 1990).

KAREN: THE WORK PLACE - One potential influence we decided to look at was the teen's involvement in paid employment. Teens who worked more than 20 hours per week were more likely to be sexually active than teens who worked less than 20 hours per week or not at all.

Figure 21: Hours of Work and Rates of Sexual Activity

This next figure shows the sexual activity rate of 11th graders as a function of the number of hours they are employed. It is interesting to note that there is not much difference between 11th graders who are not employed and those who work less than 10 hours per week. However, 11th graders who work 20 hours or more per week are about 30% more likely to be sexually active.(3)

(3) This may be due to a number of things including the fact that teens who work a lot are more likely to be around older teens and adults for much of their day; they are less likely to be supervised by parents and are more likely to have parents who are permissive; they are also more likely to have a low grade point and less likely to be invested in school. Teens who work a great deal outside the home may be more independent than others and more influenced by their peers.


Another factor that we thought might be important but that has seldom been studied is neighborhood characteristics. For example, Garbarino and Sherman (1980) found that rates of child abuse tended to be lower in neighborhoods where the neighbors looked out for each other and for each other's children. We thought neighborhood monitoring might also be related to sexual activity. Teenagers may be hesitant to engage in sexual activity if they felt that their behavior was being monitored by neighbors. I tested this possibility with the TAP data.

Figure 22: Neighborhood Monitoring

We measure neighborhood monitoring by asking teens questions like, "If you were to do something wrong and a neighbor or other adult in your community were to see you, would they tell your parents?" This graph shows that teens who report low neighborhood monitoring are about twice as likely to be sexually active than those who report high neighborhood monitoring. This is a pretty impressive effect when you consider that neighborhood monitoring is a fairly distal factor when compared to factors within the individual or family.

KAREN: COMMUNITY DISORGANIZATION - In the future, we plan to look at other community factors that may jeopardize youth. It is our guess that sexual activity may be partially related to the mobility rates and the degree of community disorganization. By community disorganization, we mean when neighbors don't know each other, parents don't talk to each other, and few community norms or standards exist regarding curfews, drinking, and age of first dating. This community disorganization, which can occur in middle and upper-middle class neighborhoods as well, may be a risk factor for sexual activity.


Adolescents respond to the messages adults send. Adult norms regarding nonmarital sexual activity may influence adolescent sexual decision-making as well. Does the community have widely known and generally agreed upon norms for nonmarital sexual activity? Do adults in the community provide positive role models regarding sexual behavior and models of mutual respect between men and women?

STEVE: MEDIA INFLUENCE - Television programming and advertising give young people lots of clues about how to be sexy but little information about how to be sexually responsible (Hayes, 1987). In a recent survey, teenagers cited television and movies as the fourth most important source of information on both sex and birth control. We really don't know about how the media affects adolescent sexual behavior, but given the millions of dollars a year advertisers spend, I think we can assume it has an impact. Yet it remains for researchers to discover the long-term effects of television's portrayal of sexuality on human development for a generation who has grown up with TV (Hayes, 1987).


Alienation or a low commitment to conventional values and institutions or alienation is related to early sexual activity (Rosenbaum & Kandel, 1990). Youngsters who feel emotional ties to their family, school, or community are more apt to accept societally accepted values and expectations for their behavior.

To build closeness between children and their families, school, and community, we need to provide opportunities for involvement, the skills necessary to be successful in this involvement, and rewards for appropriate behavior (Hawkins, n.d.). These attempts to foster bonding to social institutions and consequently, these institutions' goals and values, may attenuate sexual activity (Steinberg, 1991).


Thus far, we have examined a range of risk and protective factors at various levels of the adolescent's social ecology. As we discussed earlier, the more risk factors that are present, the greater the likelihood that a problem will occur. We examined this possibility with these data by looking at the cumulative effects of several of the risk factors we have talked about.

Figure 23: Cumulative Risk Factors

For each teen we noted how many risk factors they were exposed to, and we determined the probability that they were sexually active given different exposure to risk. Note that as the number of risk factors increases, so does the likelihood that a teen will be sexually active. Only 9% of the teens who have none of the risk factors present are sexually active while 88% of the teens who have 8 or more risk factors present are sexually active.


Let's take a moment to consider what this research and ecological risk-focused model mean for developing effective community programs. There are a number of preventive strategies that follow from an ecological risk-focused approach. Most of these strategies are relevant not only for teen pregnancy prevention but also for preventing other teen problems. We'd like to review several of the most important ones.

First, an ecological, risk-factor approach to program development suggests the need to target multiple risk factors at multiple levels of the social ecology. Program implementors should ask themselves, "What risk or protective factors will my program address? Is there scientific evidence that these factors increase or decrease the likelihood of a particular behavior?" Often, untested assumptions have driven programs, some of which have turned out to be good bets and others unfortunate guesses (BrooksGunn & Furstenberg, 1989).

Just as important is the need for programs to address multiple risk factors. As the growing body of research evidence indicates, for teen pregnancy there is no single cause, but many, and these risk factors are not confined to any one part of the adolescent's world. Furthermore, adolescents are not a monolithic group. No two kids are alike; consequently, some risk factors may be more important for one kid than another. Thus, effective programs will make an effort to address these multiple risk factors. Prevention programs exist to address many of these risk and protective factors. What does not exist is this - communities do not assess which risk factors are in place, which protective factors are missing, and then target prevention programs to the gaps that exist. It is unlikely, especially in the case of disadvantaged, minority youth, that enhancing interpersonal and decision-making skills alone will be effective; efforts will also be needed to address the problem that youth have in entering the legitimate roles of adulthood, to have a substantial or lasting effect on sexual activity.

Figure 24: Multiple Approaches

If a community is interested in preventing early sexual experimentation, they may need to mount a comprehensive, multidimensional approach, rather than investing all their effort in a single solution. For example, through church or school-based sex education programs, youth can learn assertiveness and decisionmaking skills. Schools can take steps to increase student interest in school and to improve academic performance. At the level of the family, parenting workshops might be held and newsletters on childrearing and communicating about sexuality might be sent to parents. At the community level, adult role models can be provided for at-risk youth, youth can be given positive means of demonstrating their independence and maturity, and efforts can be made to counteract the media glorification of sexuality. the vast majority of programs offering family planning or sex education, according to Hamburg (Brooks-Gunn & Furstenberg, 1986). Few attempts are made to provide competing goals (such as school achievement, employment prospects, and life planning). Little effort is made to provide alternative means of reaching their goals (intimacy without sexual intercourse or with regular contraceptive use) or to alter peer norms (making early sexual activity less desirable). For example, older teens teaching younger youth how to say no sends the message that popular older youth say no; that can be a powerful message. The key is addressing multiple risk factors in diverse parts of the child's world targeted specifically to the needs of the local community.

STEVE: Second, comprehensive prevention efforts should involve cooperation and collaboration. The magnitude of addressing risk factors at multiple levels requires collaboration. No single organization has the resources to do the whole job. The fragmented system of youth and family programming that now exists in social and prevention services has not and cannot be expected to solve the complex problems faced by youth. The well-being of youth requires a comprehensive, community-wide response.

STEVE: Third, be sure you know what the real problem or issues are facing local youth, what factors underlie them, and who is most likely to be affected. In other words, it's important to address the real problems not the perceived problems.

In Wisconsin, we have addressed this issue of real versus perceived problems by developing several community needs assessment programs. One such program, which many of you have been involved in, is the Teen Assessment Project. For those of you who don't know, TAP involves surveying local youth in the community about their mental health, worries and concerns, perceptions of the community, aspirations for the future, and frequency of both positive and problematic behavior such as drug use and sexual activity.

This local data can be a powerful way to convince members of a community about the relevance of the issue. As some of you already know, it can be an uphill battle trying to convince parents and community leaders of the many problems and challenges facing youth today. For instance, many parents don't want to acknowledge that their children are sexual beings who may very likely be sexually active.

Local data can also provide us with insight into the causes of the problems identified. Such information is invaluable as we work at developing appropriate and timely solutions. I should note that we are in the process of developing a parent needs assessment which will provide insight into such issues as local parents' beliefs about nonmarital intercourse, their support or lack of support for sex education in the school, and community norms about issues such as curfews and age of dating. Knowledge of community norms and practices will make educational programs for parents and youth much more effective.

KAREN: Fourth, have a sense of what programs and resources already exist in the community. Sometimes, in our zest to help youth and their families, we implement a program that duplicates a program already in place. In Milwaukee, for example, we found that there are over 35 organizations delivering parent education programs. Another parent education program would be foolish.

In order to address this issue, we are in the process of developing a checklist of community resources relevant to teenage sexuality. This tool will be used by the community to take an accounting of the relevant programs and resources that already exist in the community. New programs can be targeted to the gaps that exist.

STEVE: Fifth, be sure you have realistic and well-defined outcomes. If we haven't clearly defined the problem or identified our expected outcomes, we won't know if we have been successful in our efforts.

It is important to emphasize that for behaviors like sexual activity, our goal is probably not prevention in the literal sense, but rather delay or responsible behavior. Most people would agree that we want our young people to eventually develop the ability to be a responsible partner in an intimate, sexually-fulfilling relationship. So the goal of a pregnancy prevention program is probably not to prevent sexual activity, but to delay it or minimize the dangers associated with it. For example, realistic goals might include preventing pregnancy and venereal disease.

KAREN: Sixth, consider how the implementation of a particular program might affect other programs and institutions in the community - both positively and negatively. An ecological model recognizes that when one context or setting is altered, it will almost always have an effect on other systems or settings. Even our most well-intentioned efforts sometimes have unexpected consequences. Sexuality educator, Marian Howard, relayed that she recommended contraceptive use for young teens who were sexually active. As a result, some of these teens had multiple partners beginning at a young age, and later when they wanted children, found themselves infertile. This doesn't mean that Howard shouldn't have recommended contraceptives - not at all, but she could have better prepared these young adolescents for the potential consequences of having multiple partners from an early age.

STEVE: Seventh, whenever possible, involve the target audience in the planning and implementation of the program. No matter how sound the academic base, teen pregnancy prevention programming is more likely to fall if there is not broad-based community awareness, involvement, and commitment. This has been proven over and over again in human service programs for the last 25 years. Thus, while we as experts may believe that a particular course of action is desirable, it is more important that the community agree on a strategy that they think is realistic and are willing to support. In my work in Wisconsin I find that some communities prefer a community-wide solution (like a school-based clinic), while others prefer an approach that focuses on individual family responsibility (like training parents on how to talk with and educate their children about sex and sexuality).

KAREN: Let me give you an example of how involving the target audience can improve the end result. A high school principal relayed to me that he was involved in building a new school. He planted a lawn around the building and then waited to see where students walked and killed off the grass. These patches determined the location of the sidewalks, thereby preventing a potential problem. In the area of teen sexuality, youth involvement is critical, and boys are as central to the solution as girls (Hayes, 1987).

As we begin to design programs that address teen pregnancy, sensitivity to the cultural climate of the community becomes more critical. Such programs can be highly controversial. Even talking publicly about sexuality may be seen as an activist, non-neutral act by some community members. However, when we invite members of the community to be a part of the planning or implementation process, we not only learn more about their values and beliefs, which help us to tailor our efforts, but we are more likely to win their support and commitment. In Youth Futures, for example, we consciously include in the community coalition, a spokesperson for conservative elements in the community that may oppose strategies such as sex education in the school.

STEVE: Eighth, when replicating a model program, be sure that it is developmentally and culturally appropriate to the audience at which it is targeted. There is a tendency to assume that the problems faced by young people and their families, the causes of these problems, and their solutions, are uniform across all communities and all youth. Even when we do identify a model program, we need to recognize the limitations of using that program with other audiences and with other communities with different cultural and religious compositions, economic conditions, and histories.

We also need to take into account age or developmental differences. For example, because the influence of peers reaches its peak in early adolescence, peer education programs may be most effective for younger adolescents (Howard & McCabe, 1990). Similarly, pregnancy prevention programs aimed at upper level high school students may be a waste of time given that many of these students are already sexually active. If we start our prevention programs too late, teen behavior may have already begun to crystallize and therefore be much more immune to change. Evidence suggests that even our most successful prevention programs are often not successful after adolescents have become sexually active.

Developmental psychologist Larry Steinberg recommends that prevention programs for adolescents younger than 13 focus on the immediate, concrete results of being involved in risky behavior. Programming for adolescents beyond 13 years of age can build on adolescents' emerging ability to consider the long-term consequences of their actions. For example, 12-year-olds may not understand arguments against getting pregnant that stress that they will not be able to complete their education or adequately provide for their child. Examples that might be more relevant to this age group include: your clothes won't fit, you'll be too fat to sit behind your desk, you'll have another person sleeping in your room, and your friends and teachers may look at you differently.

For younger teens, especially, pregnancy and parenthood are often distant, intangible abstractions. Relating sexual decisions to lifelong consequences is difficult. Early adolescents who find it difficult to link current behavior to future consequences are often unreliable users of contraception. For them, efforts aimed at discouraging the initiation of sexual activity may be a more appropriate means of reducing unintended pregnancy. Sexuality educator, Marian Howard, explained that she used to use a decisionmaking approach with early adolescents, encouraging them to individually consider when sex was right for them. Some twelveyear -olds would weight the potential costs of sexual activity against the perceived benefits and decide that sex is right for them now. Marian Howard adapted her program and now tells young teens they are not ready for sexual activity until they are ready for the consequences.

KAREN: Finally, it's important to know the literature in the area in which you're working. There is a wealth of research-based information and theoretical reviews that are relevant to teen sexuality and pregnancy prevention. It would be a crime to not take advantage of them. Moreover, the state of the art in program development is changing at a rapid pace. As professionals we have a responsibility to keep abreast of the latest developments in our field.


KAREN: This risk-focused ecological model suggests that preventing teen problems and promoting healthy development are complex tasks. An ecological perspective is useful for understanding the factors that contribute to these complex problems, as well as strategies for addressing them.

I bet that each of you has experienced this ecological principle in your home or office. How many of you have tried to raise a plant that for one reason or another was not doing well? Have you ever had a plant that started out green and healthy but just didn't seem to thrive, its growth seemed stunted, and its future was uncertain at best? What did you do?

Did you focus solely on the plant? Did you try to stretch its leaves? Did you give it a pep talk about growing into a healthy, mature plant? Did you push and pull it in the direction you wanted it to grow?

Or did you attend to the growing conditions, the climate around the plant? Did you feel the soil to see if it was wet enough? Did you check to see if it was getting the proper amount of sunlight? Did you think back to the last time you fertilized it? Did you try to determine the special needs of this ailing plant, recognizing that not every plant needs the same remedy; a cactus, for example, needs a different environment than a violet. Conceivably, some ailing plants may need better soil. For others, the soil may be fine but the critical factor is the amount of sunlight. A third may simply need more water. Yet the plant with the dimmest prospect for survival is the one that needs all three.

STEVE: Just like plants, when things aren't going well for our young people, we need to pay attention to the climate and conditions surrounding them that may hinder their development. As the natural ecology affects the well-being of plants, the human ecology affects the well-being of youth.

There are no simple solutions to the problems of youth. But if we want to promote healthy development and encourage wise and responsible decisions by our young people, we need to put our energies into enhancing and creating the many conditions that can nurture and support them.

Thank you!

Now I'd like us to spend the rest of the time applying some of these principles to social issues we are currently facing in our own communities.


Adams, G., Adams-Taylor, S., & Pittman, K. (1989). Adolescent pregnancy and parenthood: A review of the problem, solutions and resources. FAMILY RELATIONS, 38, 223-229.

Berrueta-Clement, J. R., Schweinhart, L. J., Barnett, W. S., Epstein, A. S., & Weikart, D. P. (1984). CHANGED LIVES: THE EFFECTS OF THE PERRY PRESCHOOL PROGRAM ON YOUTHS THROUGH AGE 19. Ypsilanti, MI: The High/Scope Press.

Bronfenbrenner, U. (1986). A generation in jeopardy: America's hidden family policy. Testimony presented at a hearing of the Senate Committee on Rules and Administration on a resolution to establish a Select Committee on Families, Youth, and Children.

Brooks-Gunn, J., & Furstenberg, F. F. (1989). Adolescent sexual behavior. AMERICAN PSYCHOLOGIST, 44(2), 249-257.

Butler, J. R., & Burton, L. M. (1990). Rethinking teenage childbearing: Is sexual abuse a missing link? FAMILY RELATIONS, 39 73-80.

The Center for Population Options (1990). TEENAGE PREGNANCY AND TOO-EARLY CHILDBEARING: PUBLIC COSTS, PERSONAL CONSEQUENCES. Washington, DC: Center for Population Options.


Chilman, C. S. (1982). Adolescent childbearing in the United States: Apparent causes and consequences. In T. M. Field, A. Huston, H. C. Quay, L. Troll & G. Finley (Eds.), REVIEW OF HUMAN DEVELOPMENT, 418-431. New York: John Wiley & Sons.

Chilman, C. S. (1979). Teenage pregnancy: A research review. SOCIAL WORK, 37, 492-498.


Dryfoos, J. G. (1990). Community schools: New institutional arrangements for preventing high-risk behavior. FAMILY LIFE EDUCATOR, Summer, 4-9.

Finklehor, D. (1986). A SOURCEBOOK ON CHILD SEXUAL ABUSE. Beverly Hills: Sage Publications.

Flick, L. H. (1986). Paths to adolescent parenthood: Implications for prevention. PUBLIC HEALTH REPORTS, 101(2), 132-147.

Forrest, J. D., & Singh, S. (1990). The sexual and reproductive behavior of American women, 1982-1988. FAMILY PLANNING PERSPECTIVES, 22(5), 206-214.

Garbarino, J., & Kostelney, K. (In preparation). Neighborhood and community influences on parenting. To appear in T. Luster & L. Okagaki (Eds.), PARENTING: AN ECOLOGICAL PERSPECTIVE. Hillsdale, NJ: Lawrence Erlbaum Associates.

Garbarino, J., & Sherman, D. (1980). High risk neighborhoods and high risk families: The human ecology of child maltreatment. CHILD DEVELOPMENT, 51, 188-198.

Garmezy, N. (1983). Stressors of childhood. In N. Garmezy & R. Rutter (Eds.), STRESS, COPING, AND DEVELOPMENT IN CHILDREN, 43-84. New York: McGraw-Hill Book Company.

Hanson, S. L., Myers, D. E., & Ginsburg, A. L. (1987). The role of responsibility and knowledge in reducing teenage out-of-wedlock childbearing. JOURNAL OF MARRIAGE AND THE FAMILY, 49, 241-256.

Harris, Brown, & Bifulco, D. (1987). Loss of parent in childhood and adult psychiatric disorder: The role of social position and premarital pregnancy. PSYCHOLOGICAL MEDICINE, 17, 163-183.

Hawkins, D. (n.d.). Risk-focused prevention: Prospects and strategies. Invited lecture at the Coordinating Council on Juvenile Justice and Delinquency Prevention.

Hayes, C. (1987). RISKING THE FUTURE (Volume 1). Washington, DC: National Academy Press.

Higgins, P. S. (1988). Teenage pregnancy: An intractable problem? A literature review. St. Paul, MN: Amherst H. Wilder Foundation.

Howard, M., & McCabe, J. B. (1990). Helping teenagers postpone sexual involvement. FAMILY PLANNING PERSPECTIVES, 22(l), 21-26.

Jessor, R., & Moore, K. (1989). FACTS AT A GLANCE. Flint, MI: Charles Stewart Mott Foundation.

Moore, K. A., Peterson, J. L., & Furstenberg, F. F. (1986). Parental attitudes and the occurrence of early sexual activity. JOURNAL OF MARRIAGE AND THE FAMILY, 48, 777-782.


Roosa, M. W., & Christopher, F. S. (1990). Evaluation of an abstinence-only adolescent pregnancy prevention program: A replication. FAMILY RELATIONS, 39, 363-367.

Rosenbaum, E., & Kandel, D. B. (1990). Early onset of adolescent sexual behavior and drug involvement. JOURNAL OF MARRIAGE AND THE FAMILY, 52, 783-798.

Rutter, M. (1979). Protective factors in children's responses to stress and disadvantage. In M. W. Kent & J. E. Rolf (Eds.), PRIMARY PREVENTION OF PSYCHOPATHOLOGY: SOCIAL COMPETENCE IN CHILDREN, 3, 49-74. Hanover, NH: University Press of New England.

Rutter, M. (1983). Stress, coping, and development: Some issues and some questions. In N. Garmezy & M. Rutter (Eds.), STRESS, COPING, AND DEVELOPMENT IN CHILDREN, 1-41. New York: McGraw Hill Book Company.

Rutter, M. (1987). Psychosocial resilience and protective mechanisms. AMERICAN JOURNAL OF ORTHOPSYCHIATRY, 57(3), 316-331.

Segal, J. (1983). Utilization of stress and coping research: Issues of public education and public policy. In N. Garmezy & M. Rutter (Eds.), STRESS, COPING, AND DEVELOPMENT IN CHILDREN, 239- 252. New York: McGraw-Hill Book Company.

Small, S. A., & Silverberg, S. (1991). Parental monitoring and adolescent health-compromising behavior. Paper presented at the biennial meeting of the Society for Research in Child Development, Seattle, WA.

Steinberg, L. (1991). Adolescent transition and substance abuse prevention. Monograph commissioned and distributed by the U.S. Office of Substance Abuse Prevention, Washington, DC.

Swift, M. S., & Healey, K. N. (1986). Translating research into practice. In M. Kessler & S. E. Goldston (Eds.), A DECADE OF PROGRESS IN PRIMARY PREVENTION, 205-234. Hanover: University Press of New England.

Werner, E. E. (1990). Protective factors and individual resilience. In S. J. Meisels & J. P. Shonkoff (Eds.), HANDBOOK OF EARLY CHILDHOOD INTERVENTION, 97-116. Cambridge, England: Cambridge University Press.

Wilson, W. J. (1987). THE TRULY DISADVANTAGED. Chicago: University of Chicago Press.