National Longitudinal Survey Of Youth

BIBLIOGRAPHY

The National Longitudinal Survey of Youth (NLSY) is a set of panel surveys in which the same respondents are interviewed periodically about various aspects of their life experiences. Sponsored and largely funded by the U.S. Bureau of Labor Statistics, the NLSY data do not primarily contribute to official government statistics; instead the microdata are available to researchers for basic research on the life course. By 2006 the bibliography of publications from these data included over four thousand items on topics such as transition from school to work, job mobility, youth unemployment, educational attainment and the returns to education, welfare recipiency, the impact of training, and retirement decisions. The Center for Human Resource Research at Ohio State University carries out much of the content planning, and the National Opinion Research Center at the University of Chicago carries out the interviewing, either in person or by telephone.

The precursors of the NLSY are two cohorts of the National Longitudinal Survey of Labor Market Experience (NLS), 5,000 each in the Young Men cohort, aged fourteen to twenty-four in 1966, and the Young Women, aged fourteen to twenty-four in 1968. The young men were interviewed biennially until 1981; the young women continue to be interviewed as of 2006.

Interviewing of a new cohort comprising a 12,686-member probability sample of the age-specific U.S. population of young men and young women aged fourteen to twenty-two started in 1979 with broadly expanded survey instruments. Interviews with the NLSY79 cohort were annual through 1994 and biennial thereafter. This sample originally included military members (dropped in 1985) and an oversample of disadvantaged youths (dropped in 1991). So the eligible sample after 1991 consisted of just fewer than 10,000 respondents. The twenty-second round of interviewing took place in 2006. Interviewing of a further cohort of 8,984 young men and young women aged twelve to sixteen at the time started in 1997 and continues biennially as NLSY97.

All children born to the women in NLSY79 become members of the cohort called Children of the NLSY79, from which data have been collected biennially since 1986; as these children reach age fifteen they graduate into a separate young adult sample, begun in 1994 and also fielded biennially. The child sample and the young adult sample do not represent probability samples of their age groups in the United States, as children born to women who immigrated after 1979 and children born to women not in the age cohort fourteen to twenty-two in 1979 are excluded from the sampling frame; but they do represent a sample of all children born to women in the age cohort.

Data (without identifiers) are available as public use files, either for analysis online or for downloading, at the Web site supported by the Bureau of Labor Statistics. NLSY questions concentrate on labor force behavior but also ask about education and training, income and assets, health conditions, alcohol and substance abuse, and marital and fertility histories. There are also linked data of scores on the Armed Services Vocational Aptitude Battery, which includes measures of cognitive development. Also linked are data on the secondary schools attended by respondents. In the child sample, mothers are asked about pregnancy and prenatal care; children are given cognitive assessments at ages four and fourteen and are interviewed about family life between ages ten and fourteen. Members of the young adult sample are administered interviews similar to the NLSY79 interview itself.

Because of the longitudinal nature of the data, analyses are possible on a wide variety of topics in economics, sociology, and other areas. For example, one can trace not only entrances into and exits from the labor force, but also marital history, family formation, and the impact of such changes on labor force participation. The cognitive test information is useful for controlling for varying inputs in investigations of individual attainment. The longitudinal nature of the data also assures that the first criterion of causality—that the putative cause be temporally prior to the putative effect—has been met, in a way that cannot be achieved in a cross-sectional survey.

As in all longitudinal data, attrition from the sample can detract from the value of the data. In the original cohorts, if a respondent missed two consecutive interviews, that person was dropped from further follow-up. Hence the sample became less representative as time went on. In the 1979 cohort, however, efforts are made to contact all respondents at each round, with some success in recovering lost respondents (a response rate of 80 percent of the living original respondents for the 2002 round [Pierret 2005]). Further, a major innovation of the 1979 and 1997 cohorts is the elicitation of life histories. Respondents not only report about their current status on such variables as employment and marital status, but also report and date any changes in these statuses since the previous interview. Thus if a respondent misses one or more biennial interviews, longitudinal data for that respondent are nevertheless available from the retrospective reports given in the life history. This further ameliorates the problem of attrition, although possibly at some cost in accuracy if respondents’ memories are faulty.

SEE ALSO National Assessment of Educational Progress; National Family Health Surveys; National Longitudinal Study of Adolescent Health; Panel Study of Income Dynamics; Research, Longitudinal; Sample Attrition; Survey; Surveys, Sample

BIBLIOGRAPHY

National Longitudinal Surveys of Youth. Bibliography of work using NLSY data, http://www.nlsbibliography.org. Overview, Bureau of Labor Statistics, http://www.bls.gov/nls/nlsview.htm. Detailed documentation, http://www.nlsinfo.org/web-investigator/docs.php.

Pierret, Charles. 2005. The National Longitudinal Survey of Youth: 1979 Cohort at 25. Monthly Labor Review 128 (2): 3–7. http://www.bls.gov/opub/mlr/2005/02/art1exc.htm.

Judith M. Tanur