NLSY79 Child and Young Adult

User note: NLSY urban/rural residence variable

Users are cautioned that we have discovered an error in the NLSY urban/rural residence variable. This error affects variables in the following rounds:

• NLSY79: rounds 21-22 (2004-2006)
• NLSY79 young adult: rounds 21-22 (2004-2006)

The revised NLSY79 urban/rural variables will be released on this errata page as soon as they are created. We suggest that researchers using this variable obtain the revised data before concluding their research.

The error stems from a change in the Census Bureau's definition of an urban area. The 1990 Census criteria used in creating the NLSY urban/rural residence variable used the population of a place to determine the correct classification. People who lived in urbanized areas or places with a population of 2,500 or more were considered urban; everyone else was rural. The 2000 Census criteria changed the method of determining whether a particular point was urban or rural to one that relied on population density within an area. Areas of higher population density are called Urbanized Areas (UA) and Urban Clusters (UC). Residence in either is now considered urban.

From 2003 (the first year the new definition could be implemented), the NLS geocoders used a hybrid approach that considered respondents living in either an Urbanized Area (but NOT an Urban Cluster) or a place with a population of 2,500 or more to be urban. Otherwise the code is rural. A preliminary estimate of the differences between using this hybrid code and the 2000 Census definition indicates that 6% to 7% of respondents may be affected.

The Census Bureau website provides additional explanation about the Difference Between Urbanized Area Criteria from the 1990 Census and Census 2000.

Round 24 (2010) NLSY79 Child/Young Adult release

YA incorrect variable titles

POSTED September 2012

The following variables were inadvertently assigned incorrect titles. Below are the corrected titles:

NLSY79 Child survey: BPI Externalizing/Internalizing scales

POSTED September 2012

The NLSY79 Child survey dataset provides users with two sets of scores based on the Behavior Problems Index items, administered to mothers of children age 4-14. One set of scores is based on binary recoded items (Often or Sometimes true, 1,2 = 1; Not true, 3=0). The second set, based on trichotomous inputs, is constructed from items that are recoded from 1 (Often True), 2 (Sometimes true), 3 (Not true) to 0, 1, 2 with the following recoding: 3=0, 2=1, 1=2 before summing. These "trichotomous-based" BPI scores are designed to measure a child's tendency to internalize or externalize behaviors. The composition of both types of scales can be found in Appendix D-1 of the NLSY79 Child/YA Data Users Guide.

A file that documents the norms for the external/internal BPI scores can be found here. In the next updated NLSY79 Child data users guide, the external/internal norms will be included as a table in Appendix D. The norms for the dichotomous-based BPI scores are available in Appendix D-2 of the NLSY79 Child/YA Data Users Guide.

R24-2010 Child Supplement interviewer ID

POSTED January 2013

The following variable, from the R24-2010 NLSY79 Child Supplement, was inadvertently excluded from the 1986-2010 NLSY79 Child/YA public data release:

C51327.01 CSIR-IDNUM-REV INT REMARKS: INTERVIEWER ASSIGNMENT IDENTIFICATION NUMBER FOR CASE 2010

First cohabitation/first marriage transposition

POSTED May 2013

The data for first cohabitation and first marriage, in the YA FERTILITY AND RELATIONSHIP DATA - CREATED area of interest, are transposed. The corrected data will be available on the next data release.

Below is a crosswalk indicating which variables are transposed:

Errata last updated 11/22/2019

Newest errata

Data addition: Relationship of person(s) providing financial assistance

POSTED 2/22/2016

In 2006, a new series of questions about receiving help with living expenses from anyone other than a spouse/partner was added to the income section of the Young Adult questionnaire and has been asked in all subsequent rounds. One question in this series, Q15-74C, has been inadvertently been omitted from public release. Q15-74C asks respondents how the person(s) helping was/were related to them. In 2006, this question only allowed the selection of one relationship.  From 2008 on, this question has allowed for multiple relationships to be selected. These data will be included in our next data release.

Previous errata

Region of Residence variable now available for 1994 survey year

POSTED 9/10/2015

Region of Residence has been constructed for all YA years starting in 1996, but this variable was not previously created for 1994. Region of Residence has now been constructed for 1994 and will be available in the merged data as of our next data release. 

Updates for two retrospective child care variables

POSTED 6/17/2015

1. CCR0013 # months used 1st childcare arrangement in 2nd year of life
   
   • A file containing updated NLSY79 Child retrospective childcare data for CCR0013 is now available for public use. This file contains valid values for 106 cases in which CCR0013 (# months used 1st childcare arrangement in 2nd year of life) was previously missing in the database. One additional case changed value from 1 to 12.
2. CCR0022 # months used 1st childcare arrangement in 3rd year of life
   
   • A file containing NLSY79 Child retrospective childcare data for CCR0022 has been posted for public use. This file contains valid values for 129 cases in which CCR0022 (# months used 1st childcare arrangement in 3rd year of life) data were previously missing in the database. One additional case changed value from 1 to 6.

[1/15/2015 note:  An updated data release on 1/14/2015 corrected the 1/9/2015 errata posted below. As is our usual policy, we will keep this errata notice posted for any users who may have downloaded data prior to the updated data release]

Round 25 (2012) NLSY79 Child/YA: 2012 Family background and maternal HH Composition data update

POSTED 1/9/2015

In the current public release (December 26, 2014), a number of Child/YA cases were inadvertently assigned missing values for the following family background variables: HGCDOI2012 and ATTSCH2012.

As a result, the counts for the following Maternal HH Composition variables do not reflect the mother's status in 2012 for the child cases with missing values on ATTSCH and HGCDOI:

Note: In the next data release, these items will be updated in the public merged file and this errata will no longer be needed. Users may also wish to consider the alternative "Mother's Highest Grade Completed" constructed variable as of May 1 (HGCREV2012), which contains the full number of valid cases for all children born to interviewed mothers in 2012. HGCREV2012 is a child-based version of HGCREV12 from the mother's main Youth record, which is created to maintain consistency and prevent regression in years of schooling when respondents return to school in a new field:

Newest errata

Revised Variable CEXCLUDED2000

POSTED 6/18/2019

The variable C00115.13 [CEXCLUDED2000], which alerts users to minority oversample cases that were not fielded in 2000, has been updated to include both CHILD and YOUNG ADULT cases who were not fielded in that round.

Previous errata

Updates to 1996 HOME-SF scores

POSTED 10/18/2017

Several inconsistencies in the scoring of the Home Observation Measurement of the Environment-Short Form (HOME-SF) in 1996 have been identified for Section C (children ages 6-9) and Section D (children ages 10-14). These inconsistencies affected raw scores, standard scores, and percentile scores for Cognitive Stimulation subscale (Sections C and D), Emotional Support subscale (Section D), and the overall HOME-SF scale (Sections C and D). Details about the specific scoring issues and resulting updates can be found below.

Note that because standard scores and percentiles are generated by age, standard scores and percentile scores changed for some cases whose raw scores did not change. Spearman rank-order correlations between corrected and uncorrected standard scores and percentiles were .99 for Cognitive Stimulation, Emotional Support, and overall HOME scales.

1. Cognitive Stimulation Scores for Sections C and D
   
   • The item "How often has any family member taken or arranged to take your child to any type of museum within the past year?" was inadvertently left out of the computation of the Cognitive Stimulation subscale for sections C and D in 1996. As a result, scores were 1 point too low for those who reported taking the child to a museum once or twice or more within the past year. Values were updated for the Cognitive Stimulation raw score (COGNC1996: 1,186 updates; COGND1996: 1,373 updates), standard score (COGNZ1996: 3,198 updates), and percentile score (COGNP1996: 2,554 updates).
2. Emotional Support Scores for Section D
   
   • In 1996, the item "How many times in the past week have you had to spank your child?" was inadvertently counted as missing for children whose mothers reported that they had spanked their child 2 or more times. As a result, Emotional Support raw scores for these cases were prorated, even though there were no missing items. Values were updated for Emotional Support raw score (EMOTD1996, 81 updates), standard score (EMOTZ1996: 1,479 updates), and percentile score (EMOTP1996: 397 updates).
3. Total HOME Scores for Sections C and D
   
   • Total HOME scores for 1996 were also affected by the changes to the Cognitive Stimulation and Emotional Support subscales noted above. Values were updated for the overall raw HOME score in Section C (HOMEC1996: 1,224 updates) and Section D (HOMED1996: 1,434 updates), HOME standard score (HOMEZ1996: 3,308 updates), HOME percentile score (HOMEP1996: 2,379 updates), and the number of missing items for the D section of the HOME (HOMEFD1996: 88 updates).

These data will be corrected in Investigator for the next full release of data.

Newest Errata

Note about sampling weights

POSTED 10/7/2019

The NLSY79 Child and Young Adult 1994-2016 data set includes revisions to several Young Adult sampling weights as well as the replacement of two revised Child sampling weights.

Young Adult sampling weights

Beginning in 2010, young adults over age 30 are only interviewed every four years. The interviewed sample is selected by age as of December 31 of the survey year, so that approximately half of the older young adults are eligible each round. Since 2010, young adults age 31-32, 35-36, 39-40, 43-44, etc. as of December 31 of the target year have not be fielded.

The algorithm creating the round-specific sampling weights did not adequately account for this change in fielding, leading older YAs in the age groups that were fielded (33-34, 37-38, 41-42, etc.) to receive disproportionately high weights. The algorithm has been readjusted and the round-specific sampling weights for the interviewed YAs over age 30 have been replaced in the following variables:

• Y26159.00    [YA10WEIGHT]
• Y29663.00    [YA12WEIGHT]
• Y33318.00    [YA14WEIGHT]

Child sampling weights

An error in the code creating the revised round-specific sampling weights that were released in 2002 led to the 1986 and 1988 Child Sampling weights to be incorrectly calculated. The following weights have been replaced for all affected children:

• C05812.01 [CSAMWT1986_REV]
• C08007.01 [CSAMWT1988_REV]

Other Errata

Incorrect codebook notes

POSTED 9/24/2019

The codebook notes on the following variables should read “2 IMPLIED DECIMALS."

Uncorrectable data errors

Round 22 (2006) NLSY79 Young Adult

An error in the construction of the information sheet flag indicating whether or not the father of the respondent is deceased led respondents whose last Young Adult interview was in 2004 to have this flag indicate incorrectly that their father was deceased. These respondents were subsequently skipped over questions concerning their contact with and closeness to their fathers. Users should note that respondents who were first-time Young adults in 2006 were unaffected by this problem and were branched correctly into questions concerning their biological fathers, as were older Young Adults who were noninterviews in 2004 but interviewed in 2006. Users should also note that some of the affected questions would only have been asked of respondents under age 21; users are encouraged to consult the questionnaire to determine skip patterns surrounding these questions. The affected questions, which will show a higher than expected level of missing values, are as follows:

Newest Errata

Errata for cross-round child care variables

POSTED 2/10/21

The child-based child care variables that are assigned to the CHILD CARE area of interest (child reference numbers C03564.-C03590.) provide a cumulative updated profile of the child care experiences in the first three years of life for the NLSY79 children. Retrospective questions on child care during the first three years of life for each child were collected in the NLSY79 surveys in 1986, 1988, 1992, and 1994-2014. While child care information was not collected in the 1990 main Youth survey round, it was updated in 1992 or in subsequent rounds through 2014 for mothers not interviewed in 1992. These variables were reviewed for the 1986-2018 public release. The initial variables for whether or not a child was in childcare during a specific year of life (C03564., C03573., and C03582.) were expanded to include information to help users understand why information is unavailable for that year. The revised variables contain the following categories:

• 1 YES
• 0 NO
• 2 Child did not live with mother in this year of life
• 3 Child adopted out
• 4 Child deceased
• 5 Mother of child in dropped military oversample
• 6 Mother of child in dropped poor white oversample
• 7 Mother was a non-interview when these questions were asked
• 8 Mother of child deceased before answering these questions
• 9 Child reported or mother interviewed after these questions were discontinued

In the process of assigning these new codes, some errors in the original variables were identified and corrected. These corrections affected less than 5% of children.

SPPC Data for YA2020

Data from the Self-Perception Profile for Children (SPPC) questions in R29 were inadvertently left off of the NLSY79 Child and Young Adult 1986-2020 data release.

Download the variables: YA 2020 Self-Perception Profile for Children raw items (ZIP).

This attachment contains codebook pages for the restricted-use Young Adult geocode variables. These codebook pages provide the answer categories, frequency distributions, and other related information for the available geocode variables. They are intended to help users determine whether the geocode data would be appropriate for their research needs, including considerations of content, sample size, and other factors.

These confidential files are available for use only at the BLS National Office in Washington, DC, and at Federal Statistical Research Data Centers (FSRDCs) on statistical research projects approved by BLS. Access to data is subject to the availability of space and resources. Information about applying to use the zip code and Census tract data is available on the BLS Restricted Data Access page.

Click to view the Young Adult geocode codebook pages (TXT).

There are two public identification codes generated for all biological children of NLSY79 mothers, regardless of age: CPUBID and MPUBID.

• CPUBID (C00001.00) is the identification code for each child in NLSY79 Child and Young Adult data. It is created by appending a 2-digit code (01 to 11) to the end of the mother’s public ID (R00001.00 in the NLSY79 data). This two-digit code generally, but not always, indicates the child’s birth order. For example, the three biological children of NLSY79 main Youth respondent with MPUBID=58 will have CPUBIDs of 5801, 5802, and 5803. CPUBID is provided for all children, regardless of age and interview status.
• MPUBID (C00002.00) is the public identification code of each biological child’s mother. MPUBID can be used to link information about NLSY79 children and young adults to information about NLSY79 mothers in the main Youth database. MPUBID in the NLSY79 Child/Young Adult database corresponds to CASEID (R00001.00) in the NLSY79 main youth database. Additional details can be found in the tutorial: Linking NLSY79 Mothers and Their Children.

Other identification variables in the Child and Young Adult database

A Young Adult identification code, CASEID (Y00001.00), is available in the YA COMMON KEY VARIABLES area of interest. The Young Adult CASEID is only available for children who have been interviewed as young adults, whereas CPUBID is a comprehensive ID code with values for all children (including young adults), regardless of age or interview status. The YA COMMON KEY VARIABLES area of interest also contains MOTHERID (Y06508.00), which provides the mother’s public ID for all children ever interviewed as Young Adults. The YA-specific variables CASEID and MOTHERID are included for the convenience of users limiting their samples to Young Adult respondents.

A series of identification codes is included that identifies all of the siblings of each child as well as the interviewed spouse and sisters of each child's mother (if they were part of the original NLSY79 sample selected in 1979). The sibling identification codes (CSIBID01-10), the mother's spouse ID (SPOUSEID) and the mother's sisters IDs (SISTID1-SISTID4) are assigned to the CHILD BACKGROUND area of interest.

Identification codes are found in the Child Supplement, Mother Supplement, and Young Adult Instrument. To search for them in the NLS Investigator, use the Area of Interest criteria search and search under Child Background and YA Common Key Variables.

• Table 1. 2010 assessment scores on the NLSY79 Child file
• Table 2. 2012 assessment scores on the NLSY79 Child file
• Table 3. 2014 assessment scores on the NLSY79 Child file

/*******************************************************************

* The sample SAS program reads a child-based file extracted from the NLSY79 Child

   & Young Adult Data.
   It then reads a main Youth file extracted from the NLSY79 main Youth data set.  

   The two files are sorted by case ID and merged to create a child-based file with 

   additional mother characteristics attached to each child case ID.
* NOTE: Users who start with data in a system file format can skip the steps that 

   read the ASCII files.
*******************************************************************/
options nocenter;
 
filename chdfile 'Child file specification';
filename momfile 'Mother file specification'; 
data one(drop=C0000200);
infile chdfile;   
input   @1 C0000100  7.
           C0000200  5.
           C1526000  2.
           C1531500  2.
           C1531600  2. ;
 
kidflag=1;      /* Use flag to restrict final child-based file after merge. */
momid=C0000200; /* Use same var name for mom ID in both files to be merged. */
 
label momid       = "ID Code Of Mother Of Child";
label C0000100 = "ID Code Of Child";
label C1526000 = "Child Conditn Reqires Attention fr Dr  96";
label C1531500 = "Child Health Covered By Insurance     96";
label C1531600 = "Child Health Covered By Medicaid      96";
label kidflag       = "Observation From Child Dataset"; run;
 
proc sort;  by C0000100; run;   /* Sort by child ID; mother ID is embedded in child ID */
proc format;
  value yesnof 1 = 'Yes' 0 = 'No';  run;
 
proc contents;
  title1 'NLSY79 Child: Sample Merge program-Mother vars from main Youth file';
  title2 'Extract child-based child health insurance variables';  run;
 
 proc means;  run;
 proc freq;
   tables C1526000 C1531500 C1531600;
   format C1526000 C1531500 C1531600 yesnof.;  run;
*****************************************************************************;
data two(drop=R0000100);
infile momfile;      
input @1 R0000100  5. 
         R0214800  2. 
         R5625500  2. 
         R5625600  2. 
         R5625601  2. 
         R5625602  2. 
         R5625603  2. 
         R5625604  2. 
         R5625605  2. 
         R5625606  2.;
 
momid=R0000100;   /* Assign same ID name in both child & mother-based files */  
 
label momid    = "NLSY79 Identification Code (1-12686)  79";
label R0214800 = "Sex Of R                              79";
label R5625500 = "Children Have any Health/Hospitl Plan 96";
label R5625600 = "Hlth Plan-Current Employer Policy        96";
label R5625601 = "Hlth Plan-Previous Employer Policy      96";
label R5625602 = "Hlth Plan-Spouse/Partnr Curr Employer 96";
label R5625603 = "Hlth Plan-Spouse/Partnr Prev Employer 96";
label R5625604 = "Hlth Plan-Direct Purchase frm Med Co. 96";
label R5625605 = "Hlth Plan-Medicaid/Pub Assist/Welfare  96";
label R5625606 = "Hlth Plan - Other Source                        96";
 
if  R0214800=2;   /* Restrict main Youth file to females */ run;
 
proc sort; by momid; run;
 
proc contents;
 title 'NLSY79 Females: Extract child health insurance vars';  run;
 
proc means;  run;
****************************************************************************;
data three;
 merge one(in=kids) two(in=females); by momid;
if kids=1; /* Merge child-based and mother-based files: output as child-based file. */
run;
 
 proc means;
  title 'Check final merge of NLSY79 Child & Mother files';
 run;
 
 proc freq;
  tables C1526000 C1531500 C1531600
         R5625500 R5625600 R5625601 R5625602 R5625603 R5625604 R5625605 R5625606
         C1531500*(C1526000 R5625500);
         format R5625500 C1526000 C1531500 yesnof.;
  run;
 
  proc print data=three(obs=20);