Linking Children, YAs & Mothers

Linking Children, Young Adults, and Mothers

The Child/Young Adult data files contain many constructed variables drawn from multiple sources, including both cross-sectional and longitudinal information in both the child and mother records.  Users of the Child/Young Adult data who wish to construct variables not found in the Child/ Young Adult files may access the mothers' records in the main Youth files in order to obtain the necessary inputs.  A useful variable for linking the mother's longitudinal record to the child is C00052.00, which defines the first survey year (of the mother) following the child's date of birth. The question name for this variable is FSTYRAFT. (NOTE: Children born prior to the 1979 survey date are assigned 1979 as their first post-birth survey point.)

In general, a researcher interested in creating a variable that references a particular post-birth survey point for all children would need to follow a two-step procedure.  For example, to create a variable that defines the first post-birth survey maternal school enrollment status for all children from information collected at the mother's survey dates, one would take the following steps:

  1. Find Question Name = FSTYRAFT (reference # C00052.00) to define the appropriate survey year following the child's date of birth
  2. Use the value for FSTYRAFT to choose a maternal enrollment status variable, depending on the date of the birth of the child

If the child's first post-birth survey year FSTYRAFT is 1982, the maternal enrollment status variable as of the 1982 interview (Qname = ATTSCH1982, reference # C00570.00) would be selected.  Similarly the appropriate enrollment status information could be combined across the relevant surveys into one created "first post-birth survey enrollment status" variable.

For more refined linking procedures, the user may need to access maternal interview dates as well as the dates of many relevant events, such as school leaving, job changes, and marital changes.  Variables that are keyed to the child's date of birth or to the mother's date of interview can be used to (1) link social, economic, and demographic maternal behaviors with a child's birth or age at a particular point or to (2) define the survey point closest to a particular child's birth or attained age.  The first of these two objectives can be facilitated by using variable CBIRWEEK (ref # C02700.00), which defines the birth of every child in terms of the number of weeks that have elapsed between January 1, 1978 and the current maternal interview date.  This variable can be used to link with the NLSY79 main Youth Work History data that constructs a week-by-week profile of the respondent's employment experience.

The pre- and postnatal variables (C03200.00-C03392.10) are already defined in an appropriate life cycle context, since they reference either the pregnancy leading up to a particular birth or an event/behavior during the first year following a birth.  A missing value in this series of constructed child variables means that at least one input from the main Youth file was missing (refer to the note in the Child codebook for C02700.00).

Linking Data Files across Cohorts

The child identification code (CPUBID, C00001.00) allows the user to link children with their siblings on the NLSY79 Child files and with their mothers on the main NLSY79 files.  The child ID is composed of the first 5 digits of the mother's ID plus a 2-digit code (01-11) generally but not always indicating the child's order of birth.  Thus, children with the same first 5 digits in their IDs are siblings.  Since the child ID contains the first 5 digits of the mother ID, either the child ID or the mother ID (MPUBID, C00002.00) enables the users wishing to create a mother-based file to extract maternal characteristics and write them out only for mothers.

Child ID vs. Young Adult ID.  Users will note that identification codes are provided for all children as CPUBID (C00001.00) in the CHILD BACKGROUND area of interest and again for young adults as CASEID (Y00001.00) in YA COMMON KEY VARIABLES.  The only distinction between these two variables is that the young adult CASEID is only available for children who have been interviewed as young adults.  The variable called CPUBID is a comprehensive ID code with values for all children (including young adults), regardless of age or interview status.  The YA CASEID is included for the convenience of users accessing only the Young Adult data.

Linking Children and Young Adults to Mothers.  Linking NLSY79 mothers to their children can be done by using the case identification code for both the mother (R00001. in the main youth file) and the child (C00001.00). To merge the files, save the Child and Young Adult case ID (C00001.00, CPUBID or the Young Adult case ID, Y00001.00, CASEID) and mother case ID (C00002.00, MPUBID) from an extract of the Child and Young Adult file and save the mother case ID (R00001.00) from an extract from the main NLSY79 file.  The mother's case ID will be the same in both files. The CPUBID child ID is provided for all children, regardless of whether they are under age 15 or classified as young adults in any survey year. Appendices E (SPSS) and F (SAS) in the codebook supplement offer sample programs designed to assist users in merging files.

Important Information

In the CHILD BACKGROUND Area of Interest users will find mother ID labeled MPUBID, C00002.00 for all children regardless of age. When each round of the Young Adult is made available as a single-round preliminary release, this same variable, created only for young adults, appears as MOTHERID in the YA COMMON KEY VARIABLES Area of Interest.

NLS Tutorial: Linking Mothers and Children. We have produced a tutorial that gives detailed instructions on how to link a data file of young adult children extracted from the Child/Young Adult database with a file from the main Youth file for mothers. This example with step-by-step instructions will assist users still uncertain about the linking process.