Family Background

Family Background

Family background is a key element of the NLS because parent and sibling attributes have a significant impact on a respondent's future life experiences. Data are available on the NLSY79 respondents' parents and siblings as well as on the respondents' early characteristics. 

Parent Information

The majority of parental information was collected in the 1979 survey (see R00061.-R00089.). Users are cautioned that a significant amount of data are legitimately missing in this section. For example, 28 respondents stated that they never knew their mother and 230 stated they never knew their father. These individuals were skipped over all parent questions in 1979. Data are also missing for 121 additional respondents who did not know in what State or country their mother was born and 269 who did not know their father's birthplace. Respondents able to answer questions about their parents were asked:

  • Mother's and Father's Birthplace: The raw data show 11,310 mothers and 11,161 fathers were born in the United States, while 1,328 mothers and 1,223 fathers were born in another country. The survey also requested information on the specific State or foreign country in which the parent was born.
  • Highest Grade Completed: Almost half of mothers (46 percent) and fathers (49 percent) either have less than a high school diploma or have "don't knows" or "invalid skips" reported.
  • Mother and Father Living or Dead: In 1979, only 324 respondents stated their mother had died but 1,006 respondents stated their father was no longer living.
  • How Much of the 1978 Calendar Year Did Mother and Father Work for Pay? Respondents reported that 59 percent of the mothers and 81 percent of the fathers worked for pay during all or part of 1978. An additional question asked if the parent worked more or less than 35 hours a week.
  • Mother's and Father's 3-Digit Occupation: The most often cited occupations for a mother were Clerical and Kindred worker (1,760) while for a father they were Craftsman, Foreman, and Kindred worker (2,310).
  • Does Respondent Live With Parents? There are two separate variables that asked with whom the respondent lived.  The one-third of respondents who did not live with their mother and three-eighths not living with their father were asked the distance they lived from that parent.
  • Do Parents Live Together? If the respondent was not living with either parent, the interviewer asked if their mother and father lived in the same household

Parental Age: Many researchers are interested in knowing how old a parent is when a child is born. This subsection explains how to calculate the age of a respondent's parent at the respondent's birth by subtracting the respondent's age in 1979 from their parent's age in 1979.  Much of the age information comes from questions asked in 1987 and 1988 that determined when most parents were born. In 1987, respondents were asked for the day, month, and year that their natural parents were born. If they did not know, respondents were asked for that parent's age in years. In 1988, the questions were repeated, but only respondents who did not answer the 1987 series were asked to provide this information. The program in Table 1, which calculates the father's age in 1979, provides three separate models for creating a variable to calculate parents' ages. The computer code is listed in the second column, with an explanation in the third column. This program, and the similar one for mothers, results in an age for almost 77 percent of fathers and almost 87 percent of mothers.

Other Parent Information: Researchers can find additional information about parents in the household record if the parents lived with the respondent during any of the survey years. Parents can be found in the household record by searching the relationship fields. The relationship code for fathers in the household is "4," while mothers are coded "5." The household record in each year contains information on the person's age, highest grade completed, sex, and work history in the past calendar year.

Table 1. Computer Code to Create Father's Age in 1979

A if (R2303200 > 0) then dad_age = R2303200 - 8 1) If age in 1987 exists,
set age to - 8.
The subtraction results in the
father's age in 1979 NOT
 1987. End algorithm.
if (R2505400 > 0) then dad_age = R2505400 - 9 2) If age in 1988 exists, set age
to age - 9. End algorithm.
B if (R2303100 ^= 66) and (R2303100 > 0) then dad_age = 79 - R2303100 1) Try birth year from the
1987 survey. Note 66
means the respondent never
knew the parent.
if (R2505300 ^= 66) and (R2505300 > 0) then dad_age = 79 - R2505300 2) Try birth year from the
1988 questionnaire.
C

if ((R0175800 = 4) and (R0175900 > 0)) then Dad_age = R0175900;
if ((R0176700 = 4) and (R0176800 > 0)) then Dad_age = R0176800;
if ((R0177600 = 4) and (R0177700 > 0)) then Dad_age = R0177700;
if ((R0178500 = 4) and (R0178600 > 0)) then Dad_age = R0178600;
if ((R0179400 = 4) and (R0179500 > 0)) then Dad_age = R0179500;
if ((R0180300 = 4) and (R0180400 > 0)) then Dad_age = R0180400;
if ((R0181200 = 4) and (R0181300 > 0)) then Dad_age = R0181300;
if ((R0182100 = 4) and (R0182200 > 0)) then Dad_age = R0182200;
if ((R0183000 = 4) and (R0183100 > 0)) then Dad_age = R0183100;
if ((R0183900 = 4) and (R0184000 > 0)) then Dad_age = R0184000;
if ((R0184800 = 4) and (R0184900 > 0)) then Dad_age = R0184900;

Look at the household record to
see if the father lived in the
household in 1979.  If the father  
lived in the household, his age
should be listed.  Fathers 
 are coded as "4" on the
household record.

Siblings

The majority of sibling information was collected in two phases: One set of questions was asked in 1979, while a second set of questions was asked in 1993. In 2006 respondents were asked to confirm (from the 1993 data) whether their siblings share the same biological mother and father. 

Users are cautioned that some sibling data is missing. The first sibling question in 1979 (R00090.) asked if respondents were certain or uncertain about who their brothers and sisters are. This question shows that 1,814 out of the 12,686 respondents were uncertain of the identities of their siblings. Respondents who were unsure were instructed to "think of whomever you consider as your brothers and sisters" as the valid set of siblings. Hence, half-brothers and sisters for some, but not all, respondents will be included in the 1979 set of questions. The 1979 questions (R00090.-R00095.) capture the following information:

  • Number of Siblings: The raw data show that the modal number of siblings is two. Respondents' answers to this question range from zero siblings to 29.
  • Number of Siblings Attending School: The raw data show that the modal number of siblings in regular school is one. Respondents' answers to this question range from no siblings in school to 16.
  • Number of Siblings Older Than R: Most respondents had 1 sibling older than themselves.
  • Age of Oldest Sibling: The age of a respondent's oldest sibling (older than the respondent) ranges from 14, just above the minimum age for inclusion in the survey, to 52 years old.
  • Highest Grade Completed By Oldest Sibling: As had the parents, the typical (modal) oldest sibling completed 12 years of schooling.

Understanding siblings is important because brothers and sisters often provide influential behavioral examples for younger siblings. Beyond the 1979 data, a special sibling supplement is available. This module was funded in 1993 by the Bureau of Labor Statistics in an effort to assess the general representativeness of the siblings contained in the  multiple respondent records of the original NLSY79 sample. The module, which is located on the data set as reference numbers R41251. to R41345., contains information on up to 13 siblings for each respondent. For each of 12 siblings, the interview gathered:

  • Number of years younger or older than respondent
  • Sex
  • Highest grade completed
  • Number of children
  • Age of sibling at birth of first child

For the 140 respondents who have more than 12 siblings, an additional set of questions gathered data on the characteristics of each respondent's youngest sibling. Researchers using this data set should read a special report on the supplement's data quality, available from NLS User Services. The report, entitled The Collection of Sibling Attributes: Some Data Quality Issues, shows that "response rates are highest for items which are easier to recall and which do not change in 'value' over time as siblings leave the parental household and reduce daily contact" (Haurin 1994). Additionally, the report finds that response levels drop substantially when a respondent has more than four brothers or sisters.

In 1994 a special 14-question module was added to the "Marital History" section of the questionnaire. This module confirmed information on respondents who, during the 1993 sibling supplement, stated that they were either a twin or triplet. Questions in this module also provide additional information on the respondent's twin (triplet) sibling(s). These items are contained in variables R45215.-R45228.

Respondent Background

The NLSY79 contains a variety of information on a respondent's background characteristics. Researchers interested in the race and ethnicity of a respondent are encouraged to read the topical section entitled Race, Ethnicity & Immigration in this guide. Researchers interested in a respondent's education should refer to the Educational Attainment & School Enrollment section. The current topic focuses on three sets of background information: religion, home life at age 14, and residence history.

Religion: Questions about religious affiliation were asked of NLSY79 respondents in 1979, 1982, 2000, and 2012. The 1979 questions asked the respondents in what religion they were raised and their present religion (1979, 1982, 2000, and 2012). In addition to religious affiliation questions, the survey also asked the frequency with which a respondent attended religious services. Respondents were asked if they never attend, attend several times a year, about once a month, three times a month, about once a week, or more than once a week. Many respondents who were classified as "other" religions in 1979 were reclassified in 1982 as "general Protestant."

The sequence of questions on religion was also asked about the respondent's spouse in 1982, and 2000-2014. Additionally, in 1988, and 1992-2014, female respondents were asked how often they argue about religious matters with their husband/partner (see, for example, R27085., R38831., and R49587.).

Home Life at Age 14: The 1979 round of the NLSY79 contained a section asking respondents to describe aspects of their life at age 14. The first questions determined whether the respondent lived in the United States or outside its borders at age 14. Respondents living in the United States were asked whether they lived in a rural or urban area. If they lived outside the United States, they were asked in what country. After establishing the respondent's location, the survey then asked about the adults the respondent lived with during this time. These data provide information on the household structure during the respondent's teenage years. The survey additionally asks about the work characteristics and occupations of adults in the household.

Finally, the family background at age 14 section included three general literacy questions. The first question asked if the respondent or anyone else in his or her family regularly received magazines during the time period in question, while the second asked about the receipt of newspapers. Fifty-six percent of NLSY79 respondents stated they or a family member received magazines; more than 76 percent received newspapers. The third question asked about library cards.  Seventy percent of all respondents reported that either they or someone in their household held a library card.

Residence History: Respondents' family backgrounds were also addressed in 1979 and 1988 through questions about childhood residence. The 1979 question asked, "With whom were you living when you were 14 years old?" In 1988, a much broader set of questions was funded by the U.S. Department of Health and Human Services to supplement the 1979 data. This supplement, the Childhood Residence Calendar, enlarged the 1979 question's focus by creating a retrospective year-by-year history of each respondent's childhood from birth to age 18. The supplement focused on long-term changes by asking respondents to report living arrangements that lasted at least four months.

The key 1988 residence question is R27379. This question asked each respondent if they lived with both biological parents from birth to age 18. Respondents who stated yes were skipped over the residence section while respondents stating no were asked to fill out the residence history. In 1988, every interviewed respondent completed this question; there are no missing responses.

Information was obtained on ages at which a respondent lived with either a biological, step, or adopted mother or father. For those individuals not residing with any parent-type adult at a given age, follow-up questions detailed other multiple forms of living arrangements such as residence with grandparents, other relatives, foster care, and group or institutional arrangements. Auxiliary questions documented the age at which the respondent stopped living with a parent, which parent-type this happened to be, the reason for the change, and the frequency of visitation with the absent parent within one year after the change. The 1988 responses also provide detailed information on the reason and length of time respondents spent in alternative living arrangements. This enables a researcher to identify how often a respondent changed residence due to divorce or parental death. Users should note that if a respondent lived with at least one parent, the survey prevented them from reporting that they also lived in an alternative arrangement. For example, if a respondent lived with both her mother and grandmother, she would be marked as living with one parent even though potentially up to three adults are present in the household. 

A report providing a description and evaluating the quality of this data is available from CHRR (Haurin, 1991). Haurin (1991) compared the 1979 responses with the retrospective survey and found similar answers (see Table 4 in the Haurin study). The data show that, in 1988, about 1.6 percent more of the sample indicated they lived with two parents when they were 14 years old than had reported this arrangement in 1979. The difference was larger among minorities than nonblack/non-Hispanics.

Comparison to Other NLS Cohorts: Due to its design, the entire NLSY79 main survey provides family background information for the Children of the NLSY79.

Data provided by the respondent about his or her parents are available for all other cohorts. Cohort respondents have provided information about the country of birth and life status of their parents, as well as the educational attainment and occupation of their parents during their teenage years. NLSY97 respondents and the Mature and Young Women also provided information about their parents' health and income and about transfers of time and money to and from their parents.

The NLSY97 survey identifies siblings on the household roster giving age and relationship to the respondent. Parent information was collected from the responding parent in the round 1 Parent Questionnaire and included questions concerning parental background, attitudes, employment, health, income, marital status, and religion. Information also includes respondents' religious preferences, beliefs and practices, and church attendance in various rounds. In 1976, 1977, and 1978 respectively, a full collection of information about siblings was included in the surveys of Young Men, Mature Women, and Young Women of the Original Cohorts. Geographic data for NLSY79 respondents fall into two categories: information on the main public file and more detailed information released on a restricted-access geocode CD. 

References

Haurin, R. Jean. Childhood Residence Patterns: Evidence From The National Longitudinal Surveys of Work Experience of Youth. Report to the U.S. Department of Health and Human Services. Columbus, OH: CHRR, The Ohio State University, 1991.

Haurin, R. Jean. "The Collection of Sibling Attributes: Some Data Quality Issues." Columbus, OH: CHRR, The Ohio State University, 1994.

Survey Instruments and Documentation Interested users should see Section 1 of the NLSY79 questionnaire for parental background information concerning home life at age 14 (1979). Section 2 of the 1987 questionnaire contains parental age information. Section 2 of the 1993 questionnaire contains the sibling module. Additional childhood information was collected in Section 16 of the 1988 questionnaire.
Areas of Interest Most of the variables described above may be found within the "Family Background" areas of interest.