Inter/Intra Cohort Analyses

In Table 1 two birth cohorts are described as an example of a possible analysis that follows young adults through their NLSY79 years.  Since these two cohorts have lived all of their lives within the time frame of the NLSY79, data from their childhood, adolescence and early adulthood can be used to describe and define their lives in many ways.  The older of the two cohorts is made up of children born in 1980 or 1981 who were 25 or 26 years old as of the end of the 2006 calendar year, were born to mothers in their teens and early twenties, and who have essentially completed the transition from adolescence to adulthood by 2006. The younger young adult cohort members were born in 1986 and 1987 to mothers throughout their twenties, and were ages 19 and 20 by the end of 2006. The information presented in this table indicates how data obtained from both the child and young adult surveys during various survey rounds can be used to describe these children's first two decades or more from several disciplinary perspectives. This table also helps to demonstrate how key variables should be drawn from the appropriate survey rounds. For example, although the collection of comprehensive pre- and postnatal information for all children born to female respondents did not begin until 1984, retrospective information for all children was collected subsequently in 1985 and 1986. This means that considerable information about maternal behaviors during pregnancy, birth-related information, and health- and nutrition-related activities during the first year of life is available for almost all of these young adults.  However, information on the various child assessments noted in this table should be taken from specific survey rounds that correspond to when these children were the correct ages to receive the assessments: children born in 1980 were eligible to receive the PIAT math and reading assessments at the first child survey in 1986, but were too old to receive these assessments in 1996 and later rounds; conversely, children born in 1986 were only eligible to be interviewed as Young Adults for the first time in 2002. 

Table 1. Selected Data Elements Available over the Life Course for Two Young Adult Birth Cohorts

If we focus on the older of the two young adult cohorts in Table 1, child data collection began in 1986, the first year of the child surveys, and continued through 1994, the last biennial survey point prior to their reaching age 15.  In terms of cognitive testing, virtually all of these older children completed the PIAT mathematics and reading assessments five times?in 1986, 1988, 1990, 1992 and 1994.  They completed the Peabody Picture Vocabulary Test in 1986 and either 1990 or 1992.  At each survey point they were age-eligible, their mothers completed the 28-item Behavior Problems scale and the NLSY79 abbreviated version of the Caldwell and Bradley HOME scale.  Between 1990 and 1994 when they were in the 10 to 14 year old age period, these children additionally provided a range of self-reported information. When combined with maternal and child assessment data for the same age period, these data enable researchers to examine many aspects of cognitive, socio-emotional and behavioral development during critical late childhood/early adolescent years.

Beginning in 1996, all members of the 1980/81 birth cohorts had attained age 15 and were eligible to complete the Young Adult survey.  By 2006, most of these youth therefore have young adult data from six survey points.  (Those who had attained age 21 as of the 1998 survey date were not eligible for interview in that year but were once again eligible for interview in 2000-2008. For these young adults, questions were asked in 2000 to capture critical behavioral information covering the 1996-2000 gap.)  The Young Adult surveys include detailed data collection on schooling, employment, and family event histories, as well as a rich attitudinal profiling of these youth over time, and permit a careful descriptive and analytic exploration of the linkages between the teen and early adult years.  As these data have been collected over many years, researchers are able to examine connections that exist between early childhood, adolescence, and early adulthood, plus explore how these kinds of dynamic processes connect with attitudes and behaviors from the child's family of origin as we highlighted earlier in the section on "Life Cycle Profiles for the NLSY79 Children."  The one constraint for this specific cohort is that since all the mothers of these youth were in their teens and early twenties when they gave birth, results from these types of explorations necessarily pertain to children who were born to younger mothers, with all that that may imply for social and economic disadvantage.

The younger cohort highlighted in Table 1 offers a look at a second group of young adults who have lived their whole lives within the NLSY79 life span.  Born in 1986 or 1987, these youth were either 19 or 20 years old by the end of 2006.  An important difference between this younger cohort and their older counterpart highlighted above is that these youth born in 1986/1987 were all born to a somewhat more mainstream group of mothers ages 21-30.  Additionally, almost all of their data were collected on an ongoing basis as opposed to retrospectively.  Virtually all of these youth have now completed three young adult interviews.

Because these younger cohort members were born more recently, a relatively continuous family and household profile is available, both for them and for their mothers. This includes contemporaneous information about their residence structure, paternal presence-absence and possible visitation patterns, and virtual life-long health profiles.  Also, beginning at very early ages, this child cohort has been administered the full range of cognitive and socio-emotional assessments.  They completed the Child Self-Administered Supplement beginning at age 10 in 1996 and the Young Adult survey in 2002.  By 2006 these young adult respondents are moving into the early adult ages, completing their schooling, attaining family responsibilities, and becoming more fully integrated into the work force.

We could have selected other birth cohorts equally well for this profiling.  One of the points illustrated by our choice of cohorts is that comparisons can be made in terms of how different birth cohorts may progress from the childhood years, through adolescence and on into young adulthood.  In our example, our younger cohort members were born to women in their twenties whereas the older cohort members were born to teenage mothers and women in their early twenties.  How children born to young mothers fare when compared to children born to older mothers has been a focus of much research using these data. 

Information presented in the section on "Life Cycle Profiles for the NLSY79 Children," particularly the table, also shows how researchers may also explore attitudes and behaviors across generations. For example, many of the behavioral sequences--including early sexual activity and childbearing, religious involvement, substance use or other non-normative activities--have been asked one or more times of both mothers and their children.  The research possibilities for exploring intergenerational transmission of behaviors to one or more siblings in the child generation are therefore considerable. Further, there are also instances where mothers and children are asked essentially mirror images of the same questions so that the responses of both generations can be compared.  This includes questions on parental monitoring, school satisfaction, schooling behaviors, and knowledge about the child's friends.  It is therefore possible to explore whether mothers and children are on the same wavelength regarding how they are viewing child or mother activities, or knowledge level. Finally, there are a limited number of questions asked of young adults who are now parents themselves about how they are raising their own children. The answers they give can be compared with answers given by their mothers a generation earlier to similar questions.