What Do We Know About Adolescent Health: Findings From the National Longitudinal Study of Adolescent Health
July 17, 1998
Executive Summary
COSSA’s 1998 congressional breakfast seminar series came to an end on July 17th with nearly 80 people gathered to hear four social scientists discuss findings and results from the National Longitudinal Study of Adolescent Health (Add Health). Among the audience members were Representatives Ben Gilman (R-NY) and David Price (D-NC). The seminar speakers included Peter Bearman, Professor of Sociology at Columbia University; Kathleen Mullan Harris, Professor and Associate Chair of the Sociology Department at the University of North Carolina, Chapel Hill; James Jaccard, Professor of Psychology at the University of Albany, State University of New York; and J. Richard Udry, Kenan Professor of Sociology at the University of North Carolina, Chapel Hill.
After a brief welcome by COSSA Executive Director Howard Silver, Udry, the study’s principal investigator, discussed some background of the study and described the research design. Udry noted that the Add Health study was mandated to the National Institutes of Health’s (NIH) National Institute on Child Health and Human Development (NICHD) by Congress in the NIH Revitalization Act of 1993. Add Health, according to Udry, “is a study of a nationally representative sample of adolescents from grades 7 through 12 in the United States,” designed to “explain the causes of adolescent health and health behavior.” He further noted that Add Health covers “all the main health conditions and health behaviors of current concern,” including: depression, eating disorders, violent behavior, and drug, tobacco, and alcohol use. Udry explained, thought, that much of the research effort is focused on risk behaviors, because “the main determinant of poor health among adolescents is their behaviors.”
Udry proceeded to discuss the study’s research design. He noted that Add Health survey has been conducted in three phases. The first phase included selecting a random sample of 80 high schools throughout the study, from which 90,000 students from grades 7 through 12 were given in-school questionnaires to provide information about themselves and their “friendship networks and to measure a variety of health conditions.” In addition, Udry noted that roughly 140 school administrators were given questionnaires. During the second phase, researchers conducted in-home interviews with approximately 16,000 students and their parents. Finally, the researchers repeated the in-home interviews a year after the first in-home interview. He noted that from the in-school interviews, researchers “built a large number of special samples,” including samples blacks, Puerto Ricans, Chinese, and Cuban adolescents.
Avoiding Teen Pregnancy
James Jaccard followed Udry and focused his discussion on unintended teen pregnancy, which he said has “tremendous social, emotional, and economic costs.” Jaccard noted that for the last ten years he has been developing family-based approaches that promote communications between parents and their children to address the problem of unintended teen pregnancies. He mentioned several advantages to the family-based approach, saying
that “messages given to adolescents can be done in the context of the moral values and the moral codes of the family’s values.” Also, unlike school-based programs, parents can tailor their communications to the maturity level of their children.
Jaccard said that even if parents cannot provide technical information to their children, they can serve as a motivational force for their children. In fact, he said that the notion that parents have little, if any, impact on the actions of their teenage children is wrong. Add Health data show that parents doindeed have an effect on their children’s behaviors, he said.
Despite Add Health evidence that parents “do make a difference,” Jaccard said that “all is not rosy.” The Add Health data, for example, show a “tendency for parents to underestimate the sexual activity of their children.” Thus, Jaccard noted that “we need to strengthen communications between parents and adolescents.” He said that “we need to develop ways of teaching parents how to effectively communicate with their children on these [sexual] matters.” And, “we need to understand better why parents do not always talk to their kids.”
School Attachment/Sexual Networks
Peter Bearman focused his discussion on two projects that he has been working on: 1) school attachment (how kids feel about their school on multiple dimensions); and 2) the structure of sexual networks among adolescents.
Bearman first discussed school attachment. He noted at the outset that school attachment “is important for school success.” The Add Health data that he has analyzed shows that “as racial heterogeneity increases, school attachment decreases.” He cautioned, though, that his research results should not be interpreted as an argument for racially segregated schools. He said that schools can take steps to increase school attachment, such “integration of extracurricular activities to organize social relations of black students and white students.”
Bearman next discussed the research he has conducted on adolescent sexual networks. One analysis that he discussed centered on “Jefferson High School,” an all white, rural school of 850 students. Using extensive interview data, Bearman constructed a complete structure of all romantic and sexual relations at the high school. He found that 531 of the 850 students “are in some form of relationship with another student in the school.” He added that “286 students are tied together in a long string by sexual or romantic relationships.” Bearman noted that the structure of the relationship of the students at Jefferson High School is “really designed for extremely efficient and extremely widespread transmission of sexually transmitted diseases.” The structure, however, according to Bearman, is very fragile — use of contraceptives or abstinence by just a few students can “break apart” the whole structure.
Bearman cautioned that the sexual network study was based on only one school. Thus, he said it was “hard to know how generalizable it is.” He said, however, that “if it is generalizable, we’ll have radical implications for how we think about intervention.”
Father Involvement/Immigrant Adolescents
Kathleen Mullan Harris was the seminar’s last speaker. Like Bearman, she focused her discussion on two projects. The first project she discussed examined the impact of “family structure and father involvement on risk behavior among adolescents.” The study, Harris noted, examined the patterns of fathering involvement of resident biological fathers and non-resident biological fathers in 20,745 families. The study considered adolescent risk behavior with respect to several different types of family structures, including: two biological parents, biological mother/step father, biological father/step mother, two step parents, single mom, single dad, and other.
The AddHealth data, according to Harris, indicate a general pattern that “adolescents in two biological parent families are least likely to engage in risk behavior, while adolescents in single father families are most likely.” This pattern is particularly noticeable, according to Harris, with two specific risk behaviors: adolescents having sex, and adolescents using 3 or more substances (cigarettes, alcohol,marijuana, chewing tobacco, and hard drugs). She noted that high father involvement can protect youth from engaging in risk behavior. This is true, Harris said, for both resident and non-resident fathers’ involvement, even though the effects “are not as strong” for non-resident, highly involved fathers.
Harris proceeded to discuss a study she is conducting that examines the health status and risk behaviors of adolescents in immigrant families. The AddHealth data, according to Harris, indicate that “the longer the time spent in the U.S. and the younger the age that immigrant children arrive in the U.S. the greater the number of health problems they experience.” She said that this is also true for risk behaviors (sexual behavior, juvenile delinquency, violent behavior, substance abuse). The longer the time spent in the U.S., the “greater the socialization in American schools and neighborhoods, the greater the likelihood that immigrant children will engage in risk behavior.”
Copies of the full transcripts of the Adolescent Health seminar will soon be available. If you are interested in obtaining a copy, contact COSSA.