The National Longitudinal Study of Adolescent Health (Add Health)

Description of the Study and Data Collection

Kathleen Mullan Harris, a fellow at the Carolina Population Center, is the principal investigator of the Add Health study, which is funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). Elder is a collaborator on the project for the second and third wave of data collection. Go to the official Add Health website for more details than presented here.

The Add Health study was launched as a national school-based survey of adolescent behavior during 1994. In the first wave, the in-school survey of adolescents from grades 7-12 was carried out in 140 schools during the spring of 1994. This survey was followed up in 1995 by an in-home interview of the study youth and a principal caregiver (typically mother). The adolescents were interviewed once again in the home a year later, 1996. Approximately 10,000 adolescents participated in all three waves. The important next wave of the study occurred during 2001-2002. By these dates, all respondents had left high school for education, work, or other options. The age range extends from 18 to 24.

This is a nationally representative sample of youth who were in grades 7 through 12 as of 1994. Data have been collected from the adolescents themselves, and from their parents, siblings, friends, romantic partners, and school administrators. The National Quality Education Database provided the sampling frame with its list of all high schools in the United States. To qualify, a high school had to include an 11th grade and an enrollment of more than 30 students. From this frame 80 schools were selected. The sample was then stratified by region, suburban/urban/rural, school type (public or private), ethnic mix, and size. More than 70 percent of the schools agreed to participate and each was paired with a middle school (based on its contributions to the high school student body).

The first wave of data collection yielded more than 90,000 students from grades 7 through 12. Nearly four out of five schools agreed to provide a roster of students, from which the first in-home interview sample was drawn, the second wave in the project. From the rosters, a sample of 16,000 students was randomly selected for a 90-minute interview in the home. Approximately 200 students were recruited from schools in each school pair, regardless of size. This procedure resulted in a self-weighting sample. A total of 12,105 adolescents in grades 7 through 12 were interviewed at home. The second in-home interview with target children, comparable to the first, was carried out a year later in 1996, between April and 1996.

Special supplements were also drawn from the in-school roster. They include African-American youth from higher status families, Chinese youth, Cuban and Puerto Rican adolescents, and siblings of varying relationships — twins, full siblings, half siblings, unrelated adolescents who live together. In addition to these subsamples, the collection of student network information required the sampling of all youth who were in attendance at 14 schools (N = approx. 3,000) — a saturated subsample. This procedure enabled each member of a peer network to report on his or her own values, goals, and activities. With these data, the study can investigate peer network cultures and their influence on life choices, activities, and pathways.

Four subsamples have particular relevance to this proposed study. The largest subsample is defined as the core in the Adolescent Health sample; the nationally representative sample of adolescents who participated in the first two in-home interviews (N = 9,273). An additional 1,400 students come from the saturated schools. Two other subsamples were added to the project sample in order to test the generalizability of models: Black youth with a college-educated parent (N = 1,200); and sibling pairs (about 3,000 pairs). Another generality subsample is defined by Hispanic youth — Mexican, Central American, Cuban, Puerto Rican. By excluding youth who could be in multiple groups (such as sibs), the total number of cases in the project sample is approximately 15,000.

Research Program

The young adult transition is typically viewed in terms of behavioral continuities that neglect divergent pathways. Some youth surmount childhood hardships (a resilient path to adulthood), while others tend to lose the advantage of their childhood (become more vulnerable). Little is known about these pathways and their explanatory factors. Using Add Health data, this proposed five-year study investigates key potential influences that shape trajectories of resilience and vulnerability from adolescence to young adulthood, with attention to potential sources of variation — such as gender, social contexts, race/ethnicity, and siblings.

Objectives 1 and 2 focus on potential factors that influence the likelihood of these trajectories among youth from disadvantaged and advantaged origins, such as personal resources, family dynamics, influences beyond the family (religion, peers, mentors), mastery experiences in school and work, and role transitions. Objective 3 asks whether these trajectories and their causal processes generalize across age and gender, neighborhoods and schools, race/ethnicity, and siblings.

Data for the study come from three waves of Add Health and its sample of approximately 20,000 youth in grades 7 to 12 in 1994: Wave I (1994-1995), the in-school survey and in-home interviews with the adolescent and caregiver; Wave II (1996), interview with the adolescent; and Wave III (2001-2002). The data analysis will employ both variable-based and person-centered strategies, depending on the issues at hand, including the estimation of resilient and vulnerable youth from the residuals of prediction equations, latent growth curve models, and methods for estimating genetic and environmental effects on siblings.

Selected Citations

Ge, Xiaojia, Glen H. Elder, Jr., Mark Regnerus, and Christine Cox. 2001. “Pubertal Transitions, Perceptions of Being Overweight, and Adolescents’ Psychological Maladjustment: Gender and Ethnic Differences.” Social Psychology Quarterly 64(4): 363-375.

Johnson, Monica Kirkpatrick, Robert Crosnoe, and Glen H. Elder, Jr. 2001. “Students’ Attachment and Academic Engagement: The Role of Race and Ethnicity.” Sociology of Education 74(4): 318-340.

Crosnoe, Robert, and Glen H. Elder, Jr. 2002. “Adolescent Twins and Emotional Distress: The Inter-Related Influence of Non-Shared Environment and Social Structure.” Child Development 73(6): 1761-1774.

Crosnoe, Robert, Shannon Cavanagh, and Glen H. Elder, Jr. 2003. “Adolescent Friendships as Academic Resources: The Intersection of Social Relationships, Social Structure, and Institutional Context.” Sociological Perspectives 46(3): 331-352.

Crosnoe, Robert, and Glen H. Elder, Jr. 2004. “Family Dynamics, Supportive Relationships, and Educational Resilience during Adolescence.” Journal of Family Issues 25: 571-602.

Crosnoe, Robert, Monica Kirkpatrick Johnson, and Glen H. Elder, Jr. 2004. “School Size and the Interpersonal Side of Education: An Examination of Gender and Organizational Context.” Social Science Quarterly 85(5):1259-1274.