In This Chapter

Chapter 7.
National Longitudinal Surveys

Sampling
Each of the four original NLS samples was designed to represent the civilian noninstitutionalized population of the United States at the time of the initial survey. Each age-sex cohort is represented by a multistage probability sample drawn by the U.S. Census Bureau from 1,900 primary sampling units (PSUs) that had originally been selected from the Nation's counties and cities for the experimental Monthly Labor Survey conducted between early 1964 and late 1966. A PSU consists of Standard Metropolitan Statistical Areas (SMSAs), counties (or parishes in Louisiana), parts of counties (parishes), and independent cities. A total of 235 sample areas comprising 485 counties and independent cities was chosen to represent every State and the District of Columbia. From the sample areas, 235 strata consisting of one or more PSUs that were relatively homogeneous according to socioeconomic characteristics were created. Within each stratum, a single PSU was selected to represent the stratum. Finally, within each PSU, a probability sample of housing units was selected to represent the civilian noninstitutionalized population.

Because one of the survey requirements was to provide separate reliable statistics for blacks, households in predominantly black enumeration districts (EDs) were selected at a rate between 3 and 4 times that for the households in predominantly white EDs. The sample was designed to provide approximately 5,000 interviews for each of the four cohorts — about 1,500 blacks and 3,500 whites.

For the NLSY79 cohort, the following three independent probability samples, which were designed to represent the entire population of youths born between 1957 and 1964 and living in the United States in 1978, were drawn: (1) A cross-sectional sample designed to be representative of the noninstitutionalized civilian segment of young people aged 14 to 21 years as of December 31, 1978; (2) a supplemental sample designed to overrepresent civilian Hispanic, black, and economically disadvantaged non-Hispanic, nonblack youths; and (3) a military sample designed to represent the population aged 17 to 21 as of December 31, 1978, and serving in the military as of September 30, 1978.

All sample selection was done through a multistage stratified area probability sample of dwelling units and group-quarters units, except for individuals on active military duty. A screening interview was administered in approximately 75,000 dwellings and group quarters distributed among 1,818 sample segments in 202 PSUs (inclusive of most of the 50 States and the District of Columbia). As part of this screening interview, information was obtained that would allow the identification of persons eligible for membership in the sample.

Members on active military duty as of September 30, 1978, were sampled from rosters provided by the Department of Defense. Sample selection was accomplished in two stages. In the first stage, a sample of approximately 200 "military units" was selected, with probabilities proportional to the number of persons aged 17 to 21 years within the unit. Within selected units, persons aged 17 to 21 years were subsampled with probabilities inversely proportional to the first-stage selection probability. Females were oversampled at a rate approximately 6 times that of males, in order to produce approximately equal numbers of males and females. Within each sex, the sample was stratified on the basis of military service (Army, Navy, Air Force, and Marine Corps) and geographic location.

The NLSY97 cohort consists of two independently selected, stratified, multistage area probability samples designed to represent noninstitutionalized youths born between 1980 and 1984 and living in the United States in 1997: A cross-sectional sample and an oversample of black and Hispanic youths. For each probability sample, 100 PSUs were selected to represent all 50 States and the District of Columbia. For the cross-sectional sample, the probability of selection for each PSU was proportional to the 1990 census count of its housing units. For the supplemental sample, probabilities of selection for PSUs and segments were based on 1990 census counts of blacks and Hispanics aged 17 years and younger. In addition, areas that contain high percentages of black and Hispanic residents were oversampled. Only eligible black and Hispanic youths were screened into the NLSY97 sample from the supplemental areas.

Each of the four original cohorts was interviewed initially between 1966 and 1968. Respondents for the NLSY79 were first interviewed in 1979, and interviewing for the NLSY97 began in 1997. About 90 percent of the individuals designated for interviewing responded to each of the first-year interviews: 5,020 (91 percent) of the older men and 5,225 (92 percent) of the young men were interviewed in 1966; 5,083 (94 percent) of the mature women were interviewed in 1967; 5,159 (93 percent) of the designated young women were interviewed in 1968; 12,686 (90 percent) of the NLSY79 were interviewed in 1979; and, finally, 8,984 (92 percent) of the NLSY97 cohort were interviewed in 1997. Completion rates for the initial and latest survey years of each of the cohorts are summarized in table 1.

Next: Questionnaire Design