Abstract

Youth development programs represent an alternative way to reduce teenage pregnancy.  These programs do not focus on sexuality, as do traditional pregnancy prevention programs.  Instead, they strive to improve adolescents' life skills, belief in their future, opportunities, or "life options" more generally.  That is, they address motivation to avoid early childbearing.  Research on the antecedents of adolescent sexual behaviors and pregnancy rates, as well as the experience in this country and other countries, suggest that youth development approaches may be effective.

This paper reviews eight studies that have evaluated youth development programs.  Although the rigor of these studies varies greatly, they collectively suggest that some youth development programs may effectively reduce adolescent pregnancy or birth rates.

Introduction

Traditional sexuality education programs and family planning programs are designed to provide youths with the knowledge, attitudes, and skills to avoid sexual intercourse or to use contraception properly, and to provide reasonable access to effective methods of contraception.  Although these attributes are critical for avoidance of unprotected sex, they are not sufficient.  Without motivation, youths are not likely to use their knowledge and skills to avoid pregnancy, nor are they likely to seek access to contraceptives.

Recognizing that motivation to avoid pregnancy is critical to avoiding pregnancy, many professionals concerned with youths and with adolescent pregnancy are beginning to adopt a youths development approach to adolescent pregnancy.  A youth development approach is more holistic than are traditional approaches that focus upon improving adolescent sexual knowledge, attitudes, norms, and skills, and even on improving adolescent access to contraceptives.  Indeed, one of the underlying principles of youth development is to help prepare young people for adult life, not just to keep them problem-free (Community Network for Youth Development, 1994).  A youth development framework provides mechanisms for youths to fulfill their basic needs, including a sense of safety and structure, a sense of belonging and group membership, a sense of self-worth and contribution, a sense of independence and control over one's life, a sense of closeness and relationships with peers and nurturing adults, and a sense of competence.  Once these needs are fulfilled, youths can more effectively build competencies necessary to become successful and productive adults, and they may become more motivated to avoid early childbearing (Pittman & Cahill, 1991).  Fulfilling these needs cannot be done quickly or sporadically; thus, some youth development programs strive to change multiple facets of adolescents' lives over a continuous and prolonged period of time.

Youth development approaches to reducing adolescent pregnancy are not new.  In the early 1980s (and perhaps before), professionals in the field recognized the importance of programs that improved life options (Dryfoos, 1983).  The life options approach to reducing adolescent pregnancy was given considerable support in the 1987 study titled Risking the Future and prepared by the National Research Council of the National Academy of Sciences (Hayes, 1987).

More recently, there has been a rapid growth of interest in youth development programs.  For example, some national nonprofit organizations concerned with youths have recognized youths development as a promising approach to reducing teen pregnancy (for example, the National Campaign to Prevent Teen Pregnancy) (Kirby, 1997).  Additionally, some national foundations are funding the development and evaluation of large evaluations of youth development approaches (for example, the Children's Aid Society in New York City) (Philliber, 1996).  Further, some states are redirecting adolescent pregnancy prevention funds to youth development programs.  In California, for example, the Department of Education has a multimillion dollar initiative to develop and implement youth development programs to prevent teen pregnancy; a newly created Office of Community Challenge Grants has a large initiative to develop and implement community-based youth development programs to reduce pregnancy; and,finally, the Office of Family Planning has a male initiative that includes youth development approaches for males.  Finally, some federal agencies are funding youth development programs and syntheses of their effects (for example, the National Institute for Child Health and Human Development).

There are at least three broad reasons why youth development programs have gained such popularity.  Professionals in the field have increasingly recognized that (1) effective sexuality education programs and access to condoms and other forms of contraception are important, but further improvements in these programs will have only modest impact on teen pregnancy; (2) several types of evidence suggest that youth development programs may be effective; and (3) youth development programs may address other risk behaviors that plague many adolescents.

The remainder of this paper presents and discusses three kinds of evidence suggesting that youth development programs may reduce adolescent pregnancy, including (1) research on the antecedents of teenage pregnancy, (2) the declines in childbearing among adolescents in the United States and young women in developing countries during the past five decades, and (3) the evaluation results (often preliminary) of several youth development programs.

Antecedents of Adolescent Sexual and Contraceptive Behaviors and Pregnancy
 

During the past three decades researchers have conducted hundreds of studies of the antecedents of adolescent sexual behavior, pregnancy, and childbearing, which have been summarized in numerous reports (see, Kirby, 1997; Miller, 1995; Moore et al., 1995; Santelli & Beilenson, 1992).

The results of the research in this area demonstrate that there are a multitude of antecedents that are related to one or more sexual or contraceptive behaviors, pregnancy, and childbearing, including characteristics of the adolescents themselves, of their peers and sexual partners, of their families, and of their communities and states (Kirby, 1997).  No single one of these antecedents is highly related to behavior; rather, each of many antecedents is weakly (or occasionally moderately) related to behavior.

The antecedents can be divided roughly into three groups.  The first group includes some of the antecedents that are most strongly related to sexual behaviors--the biological antecedents, such as gender, age, testosterone level, and pubertal timing.  They are both causally and moderately related to adolescent sexual behavior.  For all practical purposes, however, these cannot be modified by social programs.

The second group includes various attitudes and beliefs directly related to sexual behavior--for example, beliefs, personal values, perceived norms, and intentions regarding sexual behaviors, pregnancy, and childbearing.  Although the results differ with the study, most of these antecedents are weakly or moderately related to actual sexual and contraceptive behaviors.  These are the antecedents that are most commonly addressed by sexuality education programs.

Finally, the third group includes a remarkably large proportion of the remaining risk factors, those involving some aspect or manifestation of social disorganization:  for example, violent crime, poverty, unemployment, family marital disruption, parents' lack of education, poor child-rearing practices, inappropriate sexual pressure and even abuse, engagement in other problem behaviors and deviance, poor educational performance, and low expectations for the future.  Depending on the study and level of analysis, some of these antecedents are strongly related to sexual behavior, pregnancy, or childbearing, while others are only weakly related.  It is this third group of risk factors that are addressed, in part, by youth development programs.

The full list of antecedents in this third group can be used to paint a picture of the youths who are most likely to engage in unprotected sexual intercourse and to become pregnant (or to impregnate others).  Clearly, not all youths who engage in unprotected sex and become pregnant (or impregnate others) have most of the characteristics in the composite picture below; indeed many youths experiencing unintended pregnancy have only a few of the following characteristics.  Nevertheless, all of the characteristics in the following composite picture do increase the odds of unprotected sex and pregnancy.

Youths at greatest risk are more likely to live in communities with high residential turnover (Brewster et al., 1993), low levels of education (Brewster et al., 1993), high poverty rates (Brewster, 1994), high divorce rates (Brewster et al., 1993) and high rates of adolescent non-marital births (Brewster et al., 1993).  Similarly, the parents of these youths have lower levels of education (Moore et al., 1995), are poorer (Brewster et al., 1993; Miller et al., 1994), are more likely to have experienced a divorce or separation or to be single (Miller et al., 1994; Wu & Martinson, 1993), and their mothers are more likely to have given birth as adolescents (Sonenstein et al., 1992).  In addition, their parents' childrearing practices are poorer (Feldman & Brown, 1993; Jaccard, Dittus, & Grodon, 1996), and the adolescents receive less support or supervision from their parents (Ensminger, 1990).  The youths themselves invest less effort in school (Ohannessian & Crockett, 1993), do more poorly in school (Brewster et al., 1993; Ohannessian & Crockett, 1993; Robbins, Kaplan, & Martin, 1985), and have lower expectations for their future (Whitley & Schofield, 1986).  Even in elementary school they are more aggressive and less well liked by their peers (Underwood, Kupersmidt, & Coie, 1996).  They are more likely to use alcohol and drugs excessively, and to engage in other unconventional and unhealthful behaviors (Costa et al., 1995; Hercog-Baron et al., 1990; Serbin et al., 1991).  They experience sexual pressure or even abuse (Boyer & Fine, 1992; Miller, Monson, & Norton, 1995).  They begin dating when they are very young (Koyle, et al., 1989) and, if female, have a relationship with an older male (Alan Guttmacher Institute, 1994).  As a result of all these factors, they are more likely either to want to have a child or to be ambivalent about having a child, and these positive or ambivalent feelings toward early childbearing affect their sexual and contraceptive behaviors and their actual childbearing (Zabin, 1994).

Although 85% of pregnancies among 15- to 19-year-olds are unintended (AGI, 1994), many teenagers have ambivalent feelings about becoming pregnant.  These ambivalent feelings have been explored in several highly acclaimed ethnographic studies.  For example, in a study of inner city adolescent females who gave birth, Leon Dash discovered that many of them had intentionally decided to become pregnant and have a baby.  They gave several reasons, such as to keep a boyfriend, to share the love of a child, and to add more meaning to their lives (Dash, 1989).  Anderson (1994) found that among inner-city youths, young males engage in sex and father children to prove their sexual prowess, while girls engage in sex to catch a young man and fulfill dreams (often unrealistic) of a better future.

Zabin's study of high-risk adolescents 17 or younger who attended two Baltimore clinics for  pregnancy tests indicates that ambivalence toward pregnancy is more widespread.  About half of those adolescent females were ambivalent about becoming pregnant and a few wanted to become pregnant (Zabin, 1994).  Furthermore, the young females who were ambivalent about early childbearing were just as likely to become mothers during the following two years as the young females who wanted to become pregnant.  This suggests that motivation to avoid pregnancy must be rather strong in order for adolescents consistently and effectively to use contraception if they are having sex.

In the past decade, researchers have begun to look more closely at youths who do not engage in risky behaviors despite being exposed to the same risks (for example, Rutter, 1985; Werner & Smith, 1992).  This research suggests that there are factors that contribute to a sense of resiliency among these youths, and that these factors "protect" them from turning to risk-taking behaviors.  One of the key protective factors of resilient youths is a sense of purpose and future.

In sum, the composite picture above of youths who are disproportionately likely to engage in unprotected sex, the risk factors on which that picture is based, the ethnographic studies that provide a more detailed understanding of motivation (or lack of motivation) to avoid pregnancy, and the resiliency literature all strongly suggest that programs focusing exclusively upon sexual knowledge, attitudes, skills, and contraceptive access are not likely to have a large impact on adolescent pregnancy.  The data also suggest that youth development programs, which address several nonsexual aspects of adolescents' lives--especially aspects which, in turn, affect motivation to avoid early childbearing--may reduce sexual risk-taking behavior and adolescent pregnancy.
 

The Historical Experience of Developing Countries and the United States
 

There are dramatic statistics suggesting that improving young women's education and life options reduces their pregnancy and birth rates.  In many countries throughout the world, as young women's educational levels have increased, their employment opportunities also increased and their fertility rates declined (Population Reference Bureau, 1995).  Women who obtain more formal education are more likely to delay both marriage and childbearing than are their peers with less formal education (McCauley & Salter, 1995).  And UNICEF has estimated that one extra year of schooling for girls reduces subsequent fertility rates by 5 to 10% (UNICEF, 1996).

In the United States, there has been a period when adolescent pregnancy and birth rates declined markedly.  The birth rates for 15- to 19-year olds declined from 90.3 per 1,000 15-19 year old women in 1955 to 53.0 in 1980 (Hayes, 1987).  Comparable figures for pregnancies are not available because of limited data on abortions during that time.

Although the causes of this decline in adolescents birth rates (and presumably pregnancy rates) are not fully understood, it is known that during this period, increasingly large percentages of young women in the United States decided to forgo early marriage and childbearing in pursuit of higher education and more challenging professional careers.  For example, in 1960 about 38% of adolescent females aged 16 to 24 who graduated from high school during the preceding 12 months entered college, while in 1980 this percentage had increased to 52% (U.S. Bureau of the Census and U.S. Department of Labor, 1996).

This explanation for the decline in the U.S. adolescent birth rate is also supported by the research on protective factors discussed above.  Women who had higher educational aspirations were more likely to avoid teen pregnancy (Plotnick & Butler, 1991), and for every additional year of schooling, African-American women delayed their first births by more than a year, while white and Latino women delayed their childbearing for nearly a year (Kahn & Anderson, 1992).

Youth Development Approaches with Results on Impact
 

This review of youth development programs includes eight studies meeting two criteria:  (1) the program evaluated was a youth development program broadly defined, and did not focus primarily on sexuality, and (2) the study provided at least minimal evidence for the impact of the program on sexual or contraceptive values, behavior, pregnancy rate, or birth rate.  Because few youth development programs have been evaluated, this review includes some programs with very weak evaluation designs and preliminary data. The primary components of the eight programs are summarized in table 1.  The programs varied in the number and type of components defined as essential program elements.

Key Components of Selected Youth Development Programs*

 Table 1

*Some of these programs may have provided most of these components. However, in this table, a program is credited with a component only if that program has identified that component as a key component of the program.

For example, the Youths Incentive Entitlement Pilot Projects (YIEPP) had only one key component, namely, the provision of employment, whereas the program implemented by the Children's Aid Society in New York City features multiple key components (discussion of contraception, provision of medical and reproductive health services, provision of academic support, help in getting employment and actual provision of employment).  None of the studies conducted thus far have examined in any rigorous way which of the components are most effective in changing risk behaviors.  Each of the eight programs is described more fully below.

Best Friends.  The Best Friends program is an intensive, long-term program for girls in grades 5 to 12.  It is designed to foster self-respect and to promote responsible behavior, especially abstinence from sexual intercourse, alcohol, and drugs.  Because one of its most important goals is for all girls in the program to abstain from sexual intercourse until marriage, girls must remain sexually abstinent to remain in the program.  The theme of sexual abstinence is emphasized in the program's multiple components:  group discussions held every three weeks during the school day on topics important to adolescents (such as, relationships, physical fitness, and sexually transmitted diseases, including HIV); role-model presentations by women from the community who discuss important decisions they made in their lives; weekly meetings between each girl and her mentor (a female member of the school faculty); weekly fitness and nutrition classes after school; special cultural events and community service projects; and end-of-year recognition ceremonies honoring each girl for her achievements during the year.

A comparison of girls in Best Friends in Washington, D.C. and other girls in Washington D.C., revealed that girls in the program were less likely to have participated in sex or to become pregnant.  The strength of this evidence, however, is reduced by the small sample size, lack of a good comparison group, use of different questions to measure sexual activity, and unknown validity of answers to questions about sexual experience among members of the program group.

Community of Caring.  The Community of Caring program is a comprehensive K to 12 school-based program that focuses on five important values:  caring, trust, respect, responsibility, and family.  The program is based on the belief that if youths understand these values, accept them, and base their behavior on them, they will then engage in more pro-social behavior and less risk-taking behavior--especially in less unprotected sexual intercourse.  To inculcate these values in youths, the Community of Caring program works with schools, communities, and families.  Within schools, the program strives to change school climate and structure.  For example, teachers modify the way they teach by including lessons and activities that reinforce the five core values, and by systematically rewarding students for actions that reflect the core values.  In addition, the schools sponsor student forums and other activities that reinforce the values.  In the community, the Community of Caring program works as a partner with businesses and community groups--students perform community service, and community organizations support the program.  To involve families, the program sponsors parent forums and other activities aimed at helping parents support the core values in the home environment.

The impact of this program has not yet been well evaluated.  However, data from one evaluation suggests that it increased student acceptance of abstinence from sex, increased actual abstinence from alcohol, and improved grades (Balicki et al., 1991).  The methods employed in that study were limited by a small number of participating schools, lack of random assignment of schools, and failure to measure sexual and contraceptive behaviors.

Youth Incentive Entitlement Pilot Projects (YIEPP).  YIEPP targeted youths from low-income households who were still enrolled in high school (Olsen & Farkas, 1990).  It was part of a national experiment begun in the late 1970s to guarantee jobs to economically disadvantaged youths.  All youths still enrolled in school and residing in the participating communities were offered part-time jobs during the school year and a full-time jobs in the summer, with the requirement that they remain in school.  Both types of job were minimum wage.

The evaluation compared eight matched community sites with and without YIEPP.  Notably, it evaluated the impact not on the participating youths, but upon all the youths in the participating communities.  This was deemed appropriate because the program increased the employment opportunities for all youths in the participating communities, not just those who participated in the program.  Using sophisticated statistical techniques, the study concluded that increasing economic opportunity decreased adolescent birth rates among blacks (the only group studied).

Teen Outreach Program (Allen, Philliber, & Hoggson, 1990; Philliber and Allen, 1992; Allen, Philliber, Herrling, and Kuperminc, in press).  The primary aim of the Teen Outreach Program (TOP) is to foster the positive development of adolescents. TOP uses both volunteer work and small group/classroom discussion.  Specifically, youths involved in the program (1) participate in individual and group service projects that impact their communities; (2) engage in discussions that enhance their personal growth and development by exploring their values as well as their relationships with family and peers and by developing valuable life skills such as communications, decision making, and goal setting; and (3) use a process of reflective discussion to connect learning from the school and TOP classrooms to the service work and their lives.  Designed as a nine-month program, participants typically meet in the small groups for approximately an hour each week, and perform a minimum of 20 hours of volunteer work during the program year.

The evaluations of this program have provided some of the most consistent and strongest evidence that a youth development program has reduced teen pregnancy.  The program has been evaluated for a number of years using common data collected from TOP sites across the nation.  From 1984 to 1995, sites identified a local comparison group using random assignment procedures when feasible, and using similar students when random assignment was not feasible.  A total of 9,116 students participated in this evaluation.  Because of the limitations of the comparison group, TOP was also evaluated using a more rigorous design in which all students were randomly assigned to receive TOP or to be part of a control group; from 1991 to 1995, a total of 695 students in 25 sites nationwide participated in this more rigorous evaluation.   Although there exist limitations of the evaluation studies, they consistently suggest that the program reduced pregnancy rates during the year in which the youths participated.  Data also indicate that the program was effective in reducing course failure, school suspension, and school dropout rates.  Several suggestions have been given for these possible positive results:  the participants developed ongoing relationships with caring program facilitators; both supervision and alternative activities reduced the opportunity for participants to engage in problem behaviors; and the volunteer experiences improved self-esteem and encouraged participants to think about their future.

Quantum Opportunities Project (QOP) (Hahn, 1994).  The QOP project is a multiyear program starting in the 9th grade and continuing through the high school years.  This program was designed to "foster achievement of academic and social competence among high school students from families receiving public assistance."  The program included educational activities (such as tutoring, computer-based instruction, homework assistance); service activities (such as community service projects, assistance at public events, regular jobs); and development activities (curriculum on life and family skills, and college and job planning).  Participants received small stipends and bonus payments for participation and completion of activities, and matching funds for approved activities after high school.  Specifically, the program guaranteed up to 250 hours of education, 250 hours of service and 250 hours of development activities each year from 9th to 12th grade.  Students received hourly stipends ranging from $1.00 to $1.33 per hour.  A $100 bonus was provided after completion of 100 hours of programming.  An equal amount was placed in an interest-bearing account for participants, which was available for use for an approved activity after high school (such as college or training).  Incentives also were used for program staff.

The program was evaluated using a longitudinal, experimental design involving multiple sites. The sample included 250 youths (50 from each site) who were selected randomly from lists of 8th grade students in families receiving public assistance.  Within each site, half the students were randomly assigned to the treatment group and half were assigned to the control group.  Each site was then required to recruit the youths for participation in the program.  Participants completed study questionnaires six times over a period of four years (fall 1989 to fall 1993).  A total of 219 youths completed the baseline survey in the fall of 1989; 170 students (88 experimental and 82 control) completed the final questionnaire in fall 1993.  According to Hahn and his colleagues (1994), there was no evidence of response bias attributable to attrition.

The evaluation revealed several positive results.  For example, program participants had a significantly lower birth rate than had the control group.  Similarly, members of the experimental group were more likely to have graduated from high school and be enrolled in a post-secondary school than were members of the control group.  Further, a significantly greater proportion of youths in the program compared with youths in the control group agreed with the statements "I am hopeful about the future" and "my life has been a success."

Summer Training and Education Program (STEP) (Walker et al., 1992).  The STEP program targeted 14- and 15-year-olds who resided in poor urban areas and were seriously behind academically.  During each of two successive summers, the program provided 90 hours of work experience, 90 hours of academic support focusing on remedial reading and math and using innovative instructional methods, and 18 hours of life skills education that included sexuality education.  During the school year between the two summers, the program provided 5 to 15 hours of other support with limited one-on-one adult contact, recreation, and other noneducation activities.

The evaluation of STEP was one of the most rigorous in the field of pregnancy prevention.  About 4,800 youths in a summer job program were randomly assigned to treatment and control groups and tracked for five years.  Thus, the evaluation measured the impact of the academic support, the life skills education, and the additional support.  Despite the large sample size and the powerful design, these components did not have a consistent and statistically significant impact on sexual activity, use of contraception, or pregnancy.   Math and reading scores did improve during the first summer, but this positive outcome vanished during the following academic years.

I Have A Future (Greene, Smith, & Peters, 1995).  The program called "I Have a Future" targeted African American adolescents aged 10 to 17 living in public housing projects.  It was established in 1987 by Meharry Medical College as a model program to reduce risk behaviors as well as unemployment, and to increase life options for youths.  The program includes a service component and numerous experiential modules.  As part of the service component, program staff implement specific activities for youths and their families and provide ongoing case management.  The experiential modules include Family Life Education, the Entrepreneur Program (which has a peer educator/counselor component), Pre-Employment Training, Prosocial Skills (substance use prevention), CHARM (Choosing How to Adorn and Refine Myself), MATURE (Males Adorning, Thinking, and Using Refined Energies), and Conflict Resolution.  The program is grounded in an African value system that promotes seven principles (the Nguzo Saba), including unity, self determination, collective work and responsibility, cooperative economics, purpose, creativity, and faith.  These principles, first introduced by Maulana Kerenga in the 1960s, are incorporated in all experiential modules, and form the basis for discussions regarding community responsibility and respect for personal rights and the rights of others.

The program was evaluated using a longitudinal, quasi-experimental design involving four public housing sites in Nashville, Tennessee.  Two of the sites received the program and the other two served as matched controls.  Participants completed a self-report questionnaire three times during the study: before program implementation (pretest), approximately 13 months after the pretest, and approximately 17 months after the first posttest.  Although the study has several limitations (such as selection bias, attrition, change in data collection procedures over time), the data indicate that the program had a positive impact on participants' acceptance of the seven principles (the Nguzo Saba).  The data also suggest that youths who reported more life options engaged in fewer delinquent behaviors.  Finally, the study found that youths who participated in numerous sessions had much lower pregnancy rates than had youths in the comparison group.

Children's Aid Society in New York City (Carrera & Dempsey, 1994; Kaye & Philliber, 1995).  One of the most comprehensive programs is that developed and implemented by the Children's Aid Society in New York City.  It is based upon the philosophy that it is important to influence multiple facets of youths' lives over a continuous and prolonged period of time.  It includes the following components: Family life and sex education for both teens and parents cover standard topics such as reproduction, contraception, and STDs including AIDS, as well as gender and family roles love, intimacy, body image and roles and responsibilities in relationships.  Medical and health services include reproductive health services, annual physical exams, and mental health and counseling services.  Dance and dramatic productions help teens and their parents explore issues such as racism, family roles, and topics related to self-esteem.  Opportunities to participate in a variety of sports provide opportunity to master new skills.  The job club provides two hours a week to learn about careers and build work skills, and then job opportunities are provided to youth.  Staff provide tutoring and help with homework.  And finally, college admission is guaranteed to many of the program participants who complete high school.  In the process of providing these services, staff and volunteers develop ongoing relationships with the youth in the program, and staff attempt to teach values through role modeling, rather than through lecturing.

Summary and Conclusions
 

The results of the research on antecedents of adolescent sexual and contraceptive behavior, pregnancy, and births rates strongly demonstrate that many factors other than knowledge, beliefs, attitudes, skills, and contraceptive access are related to these behaviors.  Taken together, this body of research suggests that to have a more marked impact on adolescent childbearing, these risk factors must be addressed.

Although studies of the change in birth rates in developing countries and the decline in the teen pregnancy rate in this country are only correlational in nature, they provide some evidence that the pursuit of higher education and careers may enhance adolescents' motivation to avoid childbearing, and may actually reduce their birth rates.

There are few studies of youths development programs with experimental or quasi-experimental designs, and several of their results are very limited methodologically.  Their results, however, are encouraging.  All but one of the studies suggest that youth development programs that focus on education, employment, and/or life options more generally may markedly reduce adolescent pregnancy rates.  Given the cost of adolescent pregnancy, this approach holds promise and should be explored more fully.  Nonetheless, because the STEP project was not successful in reducing sexual risk-taking behavior or pregnancy, additional research is needed to determine which components are most critical to reducing pregnancy rates.

Some of these youth development programs are relatively costly on a per-youth basis.  Because they have other positive outcomes, however, such as increasing education, improving employment, reducing substance use, and reducing incarceration, they may be very cost-effective in the long run.  Clearly, these types of program warrant further development and rigorous evaluation.

References

Alan Guttmacher Institute. (1994).  Sex and America's Teenagers. New York:  Alan Guttmacher Institute.

Allen J.P., Philliber S., & Hoggson N. (1990). School-Based Prevention of Teenage Pregnancy and School Dropout:  Process Evaluation of the National Replication of the Teen Outreach Program.  American Journal of Community Psychology, 18(4), pp. 505-524.

Anderson E. (1994). Sex Codes Among Inner-City Youth.  In M. Smith (ed.) Sexuality, Poverty, and the Inner City (pp. 1-36).  Menlo Park, CA:  Henry J. Kaiser Family Foundation.

Balicki, B.J.,  Goldenberg, D., Keel, K.S., Burnette, J., & Yates, T. (1991).  An evaluation of the Community of Caring-In-Schools Initiative.  Final Report.  Columbia, Maryland:  The Center for Health Policy Studies.

Boyer, D., & Fine, D. (1992). Sexual Abuse as a Factor in Adolescent Pregnancy and Child Maltreatment.  Family Planning Perspectives, 24(1), pp. 4-11.

Brewster, K. (1994). Race Differences in Sexual Activity Among Adolescent Women: The Role of Neighborhood Characteristics.  American Sociological Review, 59, pp. 408-424.

Brewster, K., Billy, J.O., & Grady, W.R. (1993).  Social Context and Adolescent Behavior: The Impact of Community on the Transition to Sexual Activity.  Social Forces, 71, pp. 713-740.

Carrera, M., & Dempsey. (1994). Carrera/Dempsey replication programs:  1993-94 summary of client characteristics and outcomes.  Accord, NY:  Philliber Associates.

Community Network for Youth Development. (1994). Background Paper: Youth development.  Unpublished paper. Redwood City, CA: Community Network for Youth Development.

Costa, F.M., Jessor, R., Donovan, J.E., & Fortenberry, J.D. (1995).  Early Initiation of Sexual Intercourse: The Influence of Psychosocial Unconventionality.  Journal of Research on Adolescence, 5(1), 91-121.

Dash, L. (1989).  When Children Want Children.  New York:  Penguin Books.

Dryfoos, J. (1983). Review of Interventions in the Field of Prevention of Adolescent Pregnancy.  Preliminary report to the Rockefeller Foundation, New York.

Esminger, M.E. (1990). Sexual Activity and Problem Behaviors Among Black, Urban Adolescents.  Child Development, 61, pp. 2032-2046.

Feldman, S., & Brown, N. (1993).  Family Influences on Adolescent Male Sexuality: The Mediational Role of Self-Restraint.  Social Development, 2(1), pp. 15-35.

Greene. L.W., Smith, M.S., & Peters S.R. (1995, June). "I Have a Future" Comprehensive Adolescent Health Promotion:  Cultural Considerations in Program Implementation and Design.  Journal of Health Care for the Poor and Underserved, pp. 267-281.

Hahn, A., Leavitt, T, & Aaron, P. (1994).  Evaluation of the Quantum Opportunities Program (QOP):  Did the program work?.  Waltham, MA:  Center for Human Resources, Brandeis University.

Hanson, S.L., Morrison, D.R. & Ginsberg, A.L. (1989).  The Antecedents of Teenage Fatherhood.  Demography, 26(5), pp. 579-596.

Hayes, C.D. (ed). (1987).  Risking the Future:  Adolescent Sexuality, Pregnancy, and Childbearing.  Washington, D.C.:  National Academy Press.

Hercog-Baron, R., Harris, K.M., Armstrong, K., Furstenberg, F., & Shea, J. (1990). Factors Differentiating Effective Use of Contraception Among Adolescents.  Advances in Adolescent Mental Health, Vol. 4.  Jessica Kingsley Publishing, Inc.

Institute of Medicine. (1995). The Best Intentions. Washington, D.C.: National Academy Press.

Jaccard, J., Dittus, P.J., & Gordon, V.V. (1996). Maternal Correlates of Adolescent Sexual and Contraceptive Behavior.  Family Planning Perspectives, 28(4). pp. 159-165, 185.

Kahn, J.R., & Anderson, K.E. (1992). Intergenerational Patterns of Teenage Fertility.  Demography, 29(1), pp. 39-57.

Kaye, J., & Philliber, S. (1995).  Comprehensive Adolescent Pregnancy Prevention Programs:  Are They Working?  New York:  The Children's Aid Society.

Kirby, D. (1997).  No Easy Answers.  Washington DC: National Campaign to Prevent Teen Pregnancy.

Koyle, P, Jensen, L., Olsen, J., & Cundick, B. (1989). Comparison of Sexual Behaviors Among Adolescents Having an Early, Middle, and Late First Intercourse Experience.  Youth and Society, 20(4), pp. 461-475.

McCauley, A.P., & Salter, C. (1995).  Meeting the Needs of Young Adults.  Population Reports, Series J, No 41.  Baltimore: Johns Hopkins School of Public Health, Population Information Program.

Miller, B.C. (1995). Risk Factors for Adolescent Nonmarital Childbearing.  In Report to Congress on Out-of-Wedlock Childbearing (DHHS Publication No. 95-1257).

Miller, B.C., Monson, B.H., & Norton, M.C. (1995). The Effects of Forced Sexual Intercourse on White Female Adolescents.  Child Abuse & Neglect. 19(10), pp. 1289-1301.

Miller, B.C., Norton, M.C., Curtis, T., Hill, E.J., Schvaneveldt, P. & Young, M.H. (1994). The Timing of Sexual Intercourse Among Adolescents: Family, Peer, and Other Antecedents.  Paper presented at the Society for Research on Adolescence.

Moore, K., Sugland, B., Blumenthal, C., Glei, D. & Snyder, N. (1995). Adolescent Pregnancy Prevention Programs:  Interventions and Evaluations.  Washington, D.C.:  Child Trends, Inc.

Ohannessian, C., & Crockett, L. (1993).  A Longitudinal Investigation of the Relationship Between Educational Investment and Adolescent Sexual Activity.  Journal of Adolescent Research, 8(2), pp. 167-182.

Olsen, R.J., & Farkas, G. (1990).  The Effects of Economic Opportunity and Family Background on Adolescent Cohabitation and Childbearing Among Low-Income Blacks.  Journal of Labor Economics, 8, pp. 341-362.

Philliber, S.  Personal communication.  November 22, 1996.

Philliber, S., & Allen, J.P. (1992). Life Options and Community Service: Teen Outreach Program. In Miller, B.C., Card, J.J, Paikoff, R.L., & Peterson, J.L. (Eds.), Preventing Adolescent Pregnancy: Model Programs and Evaluations. (pp. 139-155). Newbury Park, CA: Sage Publications.

Pittman, K.J., & Cahill, M. Pushing the Boundaries of Education: The Implications of a Youth Development Approach to Education Policies, Structures, and Collaborations.  Undated, pp. 17-28.

Plotnick, R.D., & Butler, S.S. (1991). Attitudes and Adolescent Nonmarital Childbearing: Evidence From the National Longitudinal Survey of Youth 1991, Journal of Adolescent Research, 6(4), pp. 470-492.

Population Reference Bureau. (1995). New Perspectives on Populations: Lessons from Cairo.  Washington, D.C.: Population Reference Bureau.

Robbins, C., Kaplan, H.B., & Martin, S.S. (1985). Antecedents of Pregnancy Among Unmarried Adolescents.  Journal of Marriage and the Family, 47, pp. 567-583.

Rutter, M. (1985). Resilience in the Face of Adversity: Protective Factors and Resistance to Psychiatric Disorder. British Journal of Psychiatry, 147, 598-611.

Santelli, J.S., & Beilenson, P. (1992). Risk Factors for Adolescent Sexual Behavior, Fertility, and Sexually Transmitted Diseases.  Journal of School Health, 62(7), pp. 271-279.

Serbin, L.A., Peters, P.L., McAffer, V.J., & Schwartzman, A.E. (1991). Childhood Aggression and Withdrawal as Predictors of Adolescent Pregnancy, Early Parenthood, and Environmental Risk for the Next Generation.  Canadian Journal of Behavioral Science, 23(3), pp. 318-331.

Sonenstein, F.L., Pleck, J.H., & Ku, L.C. (1992).  Cost and Opportunity Factors Associated with Pregnancy Risk Among Adolescent Males.  Paper presented at the Annual Meeting of Population Association of America, Denver, Colorado, April 30, 1992.

Underwood, M.K., Kupersmidt, J.B., & Coie, J.D. (1996).  Childhood Peer Sociometric Status and Aggression as Predictors of Adolescent Childbearing.  Journal of Research on Adolescence, 6(2), pp. 201-223.

United Nations Children's Fund. (1996).  The State of the World's Children, Oxford:  Oxford University Press.

U.S. Bureau of the Census, & U.S. Department of Labor. (1996). College enrollment of high school Graduates, various years.

Walker, G., & Vilella-Velez, F. (1992)  Anatomy of a Demonstration.  Philadelphia, PA:  Public/Private Ventures.

Werner, E. & Smith, R. (1992). Overcoming the Odds: High-Risk Children from Birth to Adulthood. Ithaca, NY: Cornell University Press.

Whitley, B.E., & Schofield, J.W. (1986). A Meta-Analysis of Research on Adolescent Contraceptive Use.  Popular Environment, pp. 173-203.

Wu, L.L., & Martinson B.B. (1993). Family Structure and the Risk of a Premarital Birth.  American Sociological Review, 58, pp. 210-232.

Zabin, L.S. (1994). Subsequent Risk of Childbearing Among Adolescents with a Negative Pregnancy Test. Family Planning Perspectives, 26(5), pp. 212-217.