The geographic indicator variables are created to describe the respondent's area of residence. A software package that assigns latitude and longitude information to respondent addresses is used. Latitude and longitude are then used to link respondent addresses to standard geographic information such as state, county, and metropolitan statistical area. See the NLSY97 Geocode Codebook Supplement introduction for more information about the use of programs used to assign latitude and longitude.

In addition, variables were created in round 1 to describe the respondent's residence as of his or her 12th birthday; another counts the number of residences in which the respondent has lived from his or her 12th birthday until the survey date (all rounds). See Youth Residential History for more details on these variables.

Note: Incarcerated respondents will have a -4 (valid skip) on CV_MIGRATE and a -3 on CV_DISTANCE_MOM/DAD while they are incarcerated as well as the interviews immediately before and after the incarceration spell.

Geocode Data File

The NLSY97 Geocode Data File provides a variety of statistics for the counties where respondents lived at each interview date. From rounds 1-5, most of these data are based on the 1994 edition of the U.S. Census Bureau's County and City Data Book (the most recent edition available). The first group of geocode variables lists basic demographic information for respondents' counties. These data include land area in square miles; population by race, age, and gender; and birth and death rates. Another variable reports the percent of persons in that county who lived in a different house and/or state in 1990 compared to their residence in 1985, providing information about migration rates for the respondent's area. Also included are the FIPS (Federal Information Processing Standards) codes for state and county (see Attachment 100 of the NLSY97 Geocode Codebook Supplement).

Factors that might influence the respondent's education and employment outcomes are the focus of several other geocode variables. These provide the number of serious crimes, households with children, female householders with no spouse present, persons with high school or college degrees, and families below the poverty level. A pair of variables summarizes medical availability for each county, reporting the number of active nonfederal physicians and community hospital beds.

Geocode variables associated with economic and labor force issues include the size of the county's civilian labor force, the percent employed in various industries, and the percent of workers age 16 and older with jobs outside their county of residence. Income variables include per capita money income for the respondent's county, per capita personal income, and median family money income. County and City Data Book information is not included in the geocode data set for rounds 6 and later. Information on obtaining the most recent County and City Data Book data is available at https://www.census.gov/library/publications/2010/compendia/databooks/ccdb07.html. The unemployment rates for the respondent's metropolitan areas or for the portions of the state not in an MSA (for non-MSA respondents) are calculated from other sources for all rounds.

Other geocode variables involve colleges attended by respondents. Survey staff use information from the Integrated Postsecondary Education Data Systems (IPEDS) to provide users with a code identifying each college attended by the respondent and its location. These codes can be used to associate the NLSY97 respondent's college with various characteristics of the institution contained in the IPEDS database. More information is located in Attachment 102 of the NLSY97 Geocode Codebook Supplement.

In addition, to support research on respondent mobility, survey staff created series of geocode variables for the distance between respondent addresses at each interview round. The data do not provide a location for the respondent's residence; rather, these variables provide distances between the various places the respondent has lived. Staff also created indicators of the quality of the geographic data. More information is located in Attachment 103 of the NLSY97 Geocode Codebook Supplement.

Due to the detailed nature of the information found on the Geocode Data File, use of this data set is restricted to those meeting confidentiality requirements. The geocode data set is released only to those who satisfactorily complete the Bureau of Labor Statistics geocode agreement procedure. The application for this geocode agreement is now available online at www.bls.gov/nls/geocodeapp.htm. Researchers interested in the geocode data can view a complete set of codebook pages in Attachment 104 of the NLSY97 Geocode Codebook Supplement. These codebook pages show frequency distributions and answer categories for the restricted-use geocode variables and may help users to determine whether the NLSY97 geocode data and sample sizes will be appropriate for their research interests.

Note: Incarcerated respondents will have a -4 (valid skip) or -3 (invalid skip) on many restricted geocode variables due to increased privacy concerns for these individuals. Details are provided in the NLSY97 Geocode Codebook Supplement. The majority of respondents still have childhood information available.