PreK=-3rd Data Resource Center: The First Six Years of Schooling and Beyond

National Longitudinal Survey of Youth, 1979: Child Surveys Resource Guide

Introduction

About the Guide

This resource guide provides a brief overview of the children of the National Longitudinal Survey of Youth, 1979 and instructions for creating an extract dataset which you can download to your own computer. It also offers guidance in obtaining access to additional data from the main sample of mothers and for other family members, as well as guidance in accessing restricted-use versions of the data.

About the Data

Data about the children of the National Longitudinal Survey of Youth, 1979, which can be found in the NLSY79 Child Surveys, provides detailed data on the cognitive, socio-emotional, physiological, and demographic characteristics of children born to a cohort of women who have participated in the NLSY79 surveys. Sponsored by the U.S. Bureau of Labor Statistics (BLS), the original NLSY79 survey consisted of a sample of individuals ages 14-22 first interviewed in 1979, annually through 1994 and then biennially to obtain information on labor force characteristics as well as a variety of other demographic, social, and financial indicators.

Beginning in 1986, the National Institute of Child Health and Human Development (NICHD) sponsored a set of supplemental surveys to gather information about the children of female NLSY79 participants. Data collection for the NLSY79 Child Surveys includes a maternal interview and child assessment performed by the interviewer for all biological children of NLSY79 mothers. Children of any age were assessed prior to 1994, while only children under age 15 were assessed starting in 1994. Since 1988, children ages 10 to 14 have also answered self-administered, confidential questions on sensitive topics such as family, friends, school, attitudes, and deviant behavior. For children who have reached age 15 by the end of the survey year, the assessment portion of the interview is omitted for these children are instead viewed as young adults and therefore complete an interview similar to the general NLSY79 computer-assisted interview.

In 1994-1995 a Child School Survey also was conducted, collecting detailed information from Principals and other administrative staff regarding the school and the performance of individual students. Many school survey variables are available in the public-use version of the NLSY79 Child Surveys file, while others are available on a restricted-use basis only.

The Center for Human Resource Research (CHRR) at The Ohio State University and National Opinion Research Center (NORC) at the University of Chicago are responsible for the design, collection, and distribution of the NLSY79 data and documentation.

Acknowledgements

This resource guide was prepared by Donald J. Hernandez, Department of Sociology, University at Albany, State University of New York. It was developed for the PreK-3rd Data Resource Center: The First Six Years of Schooling and Beyond, a website hosted by ICPSR with support from the Foundation for Child Development.

Sample

Data collection for the original 1979 cohort of NLSY79 participants included a nationally representative sample of 12,686 individuals born between 1957 and 1964, ages 14 to 22 at the time of the first interview. The NLSY79 Child Surveys contain information on all biological children of the NLSY79 female participants. When data collection for the NLSY79 Child Surveys began in 1986, all children who had been born up to that time (between 1970 and 1986) were included in the sample. The subsequent data collection in 1988 increased the size of the sample by adding any children born between the 1986 and 1988 survey dates. In this way, the sample size for the NLSY79 Child Surveys will continue to increase at each collection point until the original NLSY79 cohort has completed childbearing. The following table provides an example of the original NLSY79 cohort size and the subsequent number of children interviewed in the NLSY79 Child Surveys in 2004.

Number of women in 2004 NLSY79 Number and % of NLSY79 women who are mothers Number of children born to NLSY79 women as of 2004 Number and % of NLSY79 children interviewed in 2004
3,984 3,365 (84%) 8,267 7,538 (91%)

Data Elements

The NLSY79 Child Surveys collect a wide variety of information on the educational, psychological, social, and demographic characteristics of the children of NLSY79 participants. In addition to the information collected in the child interviews, NLSY79 allows the user to link child information with an extensive collection of information about the mother, as well as work history and geographical code information. This is made possible through a unique identifier or ID (MPUBID) that can be used to attach each child's observation to the mother's file in the NLSY79 main Youth dataset. In addition, a second unique ID number exists for each child in the NLSY79 Child Surveys. This second ID (CPUBID) allows users who make two or more separate extract data files on different occasions to merge or concatenate these data files. Variables within the NLSY79 Child Surveys are classified as belonging to one or more major data topics or categories. These categories are presented in the table below along with examples of the types of data included in each area of interest. Note: In all cases these are partial variable lists. Additional information below will direct you to complete lists of available variables.

Data category Sample variables available
Child demographic and family background characteristics Sex, race, date of birth, birth order, sibling identifier, characteristics of mother, usual living situation, location of father's residence, frequency of contact with father.
Prenatal and child postnatal health history Mother's prenatal care, alcohol use, smoking, drug use, gestation and birth weight, infant feeding practices, child illnesses and injuries in the first year of life.
Child health Physical characteristics, health conditions affecting school attendance or activities, use of medicine or medical equipment, healthcare usage, presence and number of accidents, injuries and illnesses, last routine checkup and health insurance coverage.
Child home environment Home Observation for Measurement of the Environment (HOME) Inventory, measures the nature and quality of the child's home environment. This is collected from the mother and by the interviewer at the time of the interview. Data include responsiveness and interaction with the mother, organization of the environment and for older children, parental modeling of maturity.
Child cognitive development The following assessments are administered to children below age 15 (to all ages previous to 1994). Children above age 14 are not assessed starting in 1994.
  • Peabody Picture Vocabulary Test - Revised

  • McCarthy Scale of Children's Abilities: Verbal Memory Subscale

  • WISC-R: Digit Span Subscale

  • PIAT: Math, Reading Recognition, and Reading Comprehension Assessments

  • Memory for Location Assessment

  • Body Parts Assessment

Child motor, social and emotional development For each child below age 15, mothers complete the following assessments:
  • Behavior Problems Index

  • Temperament Scale

  • Self-Perception Profile for Children

  • Motor and Social Development Scale

Educational Experiences School attendance, grades skipped or repeated, type of school attended, quality of school, child's performance in class (all children); characteristics of their school, parental involvement in their education, educational expectations (children 10+ yrs); degrees received, college expectations (children 14+).
Preteen and teen behaviors and attitudes Child-parent interaction, child home responsibilities, attitudes toward school, time use, computer use, employment, religious attendance, alcohol and drug use, sexual activity, dating and friendship patterns.
Young adult survey (children 15+ yrs) Job history, military experience, training investments, physical well-being, dating and marital history, fertility, household composition, family income and assets, substance use, childcare, education.
Mother-specific information In addition to mother information available on the main data file, the Child datafiles include constructed variables on: maternal family background, household composition, educational background of the child's parents and other household members, maternal heath history and work history.

Create Extract File

Accessing NLSY79 Child Data

Researchers who are interested in analyzing public-use data from the NLSY79 Child Surveys can download data via a Web-based extraction program (titled Web Investigator) provided by the Center for Human Resource Research (CHRR).

The following instructions will guide you through the process of selecting and extracting data from the NLSY79 Child Surveys, including options available at each stage of the process, and links to additional information.

Step 1: Navigate to the Web Investigator

Access the Web Investigator.

Step 2: Create a User Account

From the Web Investigator home page, new users are required to provide their email address to create an account to access the data. After providing the required information, the user must activate the new account by responding to an email that will be sent immediately to the email address you provide. The creation of a Web Investigator account allows the user to extract data and generate tables.

Step 3: Web Investigator Introduction

The Web Investigator home page includes the following links of interest across the top of the page:

  • Web Investigator directs the user to a page to begin the data search and extraction process.

  • Custom Weights allows the user to create customized weights to analyze studies which cross survey years.

  • Documentation allows the user to select for documentation on each of the NLSY79 components as well as links to the BLS and NLS Web pages.

  • Support & Contacts provides contact information for users of the NLSY79 and space for emailing questions or comments.

Detailed information about each of these options may be found in the NLS Web Investigator User's Manual.

To continue, select Web Investigator from the top of the home page.

Step 4: Select and Filter Variables to Create Extract Dataset

The page to which you have been directed provides several options for selecting variables to be included in your extract dataset. Begin by selecting Choose a Cohort. To extract NLSY79 Child Surveys data, select "NLSY79 Child/Young Adult (1986-2004)-revised 2008-03-04."

Within the Add a Filter section beneath the dataset title, are a set of filters that can be applied to limit the number of variables in the extract dataset. Users who have targeted specific variables or topics of interest can search for those variables in several ways. Note that more than one filter can be used to define a search.

Filter options include:

  • Word in Title: Allows the user to select a term from the attached dropdown menu and a list of variables with that term in the variable title is returned.

  • Search Variable Title: Allows the user to enter an alpha-numeric string in the attached data entry box (e.g. income). A list of variables with that keyword in their titles is returned.

  • Area of Interest: Allows the user to choose groups of variables.

  • Survey Year: Allows users to choose variables for data collected in a specific year.

  • Reference Number: Users who know the reference number of the set of variables they would like to extract can select the number from a drop down menu. All variables that fall within the specified range of reference numbers are returned.

  • Question Name: Allows users to select groups of questions across survey years.

  • Search Question Name: Allows a search according to the question name (often a mnemonic) used to designate a specific variable. This option provides for a more targeted search but is best used by users who have some experience with the data or the questionnaires.

  • Search Question Text: Allows a search according to the wording of the survey question itself.

  • Optional: Allows the user to view an individual case from the survey.

The best place for new users to begin is the Area of Interest drop-down menu, which provides many variable groupings from which to choose. These include assessment-related variables, items from the Child and Mother supplements, as well as the young adult questionnaire for children age 15 and over, for each of the survey years. In addition there are groupings of child background, child care, and family background variables. The category "child school survey" includes the 723 publicly-available variables from this data collection. After selecting one or more of the filter options, select Submit Filter Choice(s). The list of all variables meeting the filter requirements will be displayed at the bottom of the screen.

If there is not a list of variables you may notice the following message: '0 variable records.' This message indicates that either the filters applied to the dataset were too narrow or the search term or keyword used was not found. Attempt another filter search using broader criteria.

Step 5: Tagging Desired Variables

If the filter search has been successful, you will see a table at the bottom of the screen with the list of selected variables. This table provides several pieces of information in the following columns from left to right:

  • Unlabeled, left-most column: the number of that variable on the filtered list

  • Name: the reference number or name (mnemonic) of that variable

  • Tag: a check box to 'tag' or select that variable for extraction

  • Question: the question name designation on the questionnaire

  • Variable Title: the descriptive title describing the variable

  • Year: the survey year in which the question was asked

To include variables in the extract dataset, each variable must be tagged by placing a check in the appropriate box. When all desired variables have been checked the user may choose to perform another filter search or to click Review Tagged Variables before performing the extraction. If additional variables are needed the user must choose whether the new filter search should include the specifications provided in earlier searches. Web Investigator saves previous filter specifications within the section Filters Applied. To include earlier search criteria, check the box next to the appropriate filters. If the previous filters should not be included do not check this box and they will automatically be discarded.

It is important to note that for every data extraction the user MUST include the two identifiers unique to each record (CPUBID and MPUBID). These can be found by using the Area of Interest Filter to locate these key variables in the CHILD BACKGROUND area of interest. In the drop-down box, select Child Background, then select Submit Filter Choice(s). Then "tag" the first two variables (CPUBID and MPUBID).

To extract some or all of these data elements the user must select the desired variables. Above the list there are two helpful tabs: one which enables the user to tag (select) all of the data elements on the list, and another that untags the entire list of variables. To select individual variables, read down the list using the check box on the left-hand side of the table to mark each desired data element. Warning: Do not press the browser's 'Back' button while in the process of selecting new variables. Doing so will erase previous lists of tagged variables.

Step 6: Obtain More Information about the Variables

In some cases the user might want to preview a variable before choosing to extract it. Clicking on the variable name in the second column of the table opens an excerpt of the codebook or data dictionary that contains important information about the variable selected, including:

  • The variable name, question name and question title

  • The number of observations with non-missing values for that variable

  • For categorical variables, all valid values, with labels, and the number of respondents with those values in the dataset.

  • The number of observations with values of 'refused', 'don't know', or 'missing'

  • The minimum and maximum valid values for that variable

  • The actual minimum and maximum values for that variable in the dataset; the mean value for that variable (numerical variables only).

  • The lead in question or question asked immediately preceding this variable

  • The next question to be asked in the survey according to the questionnaire and skip patterns.

Step 7: Extract Tagged Variables

After tagging all desired variables and reviewing the list of selected variables (this is a good time to check to ensure that you have included the two necessary IDs), it is time to extract the variables.

Under the list of tagged variables click the button that says Extract Tagged Variables.

The following page allows the user to provide Universe Restrictors and to specify Extraction Options. Universe restrictors are not required. To select the entire universe, scroll down to Extraction Options. Universe restrictors can be used to limit the sample of respondents included in the extracted dataset. Limitations can be created based only on variables included in the extraction. For example, if the researcher intends to study only females in the sample, then under 'Condition 1' the user should select the variable for respondent's sex and then complete the following two fields to include the equals sign (=) and the number 2. This is the correct value to select for females in the sample. The user may apply up to two universe restrictors to the sample.

Options available under Extraction Options allow the user to choose the desired format of the extract data file. There are three options: Delimited ASCII data file, Comma Delimited ASCII data file and STATA dictionary file. If Delimited ASCII data file is chosen, the user must also specify either SAS or SPSS syntax statements along with the data. The user then chooses whether to download the codebook portions for the extracted variables. After making these selections, click Submit Request.

A table appears showing the status of the request. After submitting this request the user is brought to a page that includes a table of recent data extraction requests. The table includes the date of this request, the size of the request, and in the middle column a note that the request is running. Note: Click Submit Request only once. To determine whether the extraction is running consult the middle column of this table, the word 'Running' should appear next to the request number, and will continue to appear until the request is fully processed. The length of time required to extract and download each request will vary significantly depending on the number of variables chosen and the size of the sample.

When the download is complete, a zip file name will appear in the 'Name' column of the table. Click on the zip file and follow the directions provided by the data compression application on your computer. Save the files on your personal computer. Web Investigator retains data requests for up to four days after download.

Learn More

NLSY79 Restricted-use Data

Restricted-use versions of NLSY79 data, including Geocode data, ZIP Code and Census tract files and additional variables from the NLSY79 School Surveys are available only to approved researchers. Access to Geocode data is granted through an application process that allows for the controlled use of data at a secure location at the user's home institution. Data from the Zip Code, Census tract files, and NLSY79 School Surveys is available for analysis only at the BLS national office in Washington, DC. Information about these application processes, including links to the application forms can be found at their website.

Data on Siblings, Mothers, Fathers, Aunts, and Cousins

As previously noted, information is collected regarding all biological children of women who participated in the NLSY79 Surveys. Therefore, for any given child, the child dataset includes a complete record for each of the child's biological siblings who were ever interviewed. The child dataset provides the identification code of any other siblings so users can link information across children who are related.

It should also be noted that, while the child files include substantial information regarding each child's mother, additional information is also available in the NLSY79 Youth Surveys data file. The variables included in the main Youth file for individuals who are mothers of children in the child file can be viewed and extracted using the Web Investigator by selecting "NLSY79 (1979-2004)-revised 2007-03-09." Procedures described to create an extract file for children can also be used to create an extract file that includes sample persons who are mothers. This extract file can be merged into an existing extract dataset already created by the user. Variables that may be especially useful for studies of children include the following variable groups that can be selected through the Area of Interest Filter: Attitude, Birth Record, Child Care, Family Background, Fertility and Relationship History/Created. In addition to the father characteristics included in the NLSY79 child dataset, "spouse" variables from the mothers' main Youth record can be analyzed as "father" variables for children.

Because the original NLSY79 survey included all individuals living in sample households ages 14 to 22, the Youth Surveys sample includes a large number of sisters and their children. These respondents are the aunts and cousins of individuals in the child file. With appropriate matching procedures, it is possible to merge datasets with information about aunts or cousins into an existing extract dataset already created by the user.

Guidance for creating and merging datasets with variables for siblings, mothers, fathers, aunts, and cousins can be obtained by selecting the Support & Contacts link in the Web Investigator.

Additional Resources