Behavioral Regulation Problems in Childhood Predict Adult Financial Status, Health, and Criminal Activity: A Conceptual Replication and Extension of Moffitt et al. (2011) Using Cohorts from the U.S. and the U.K.

• In Event: Early Self-Regulation and Long-Term Outcomes: Why Does Childhood Self-Regulation Predict Adult Success?

Abstract

In a highly influential study cited over 5,000 times, Moffitt et al. (2011) found that children exhibiting behavioral regulation problems had greater financial struggles, poorer health, and a greater likelihood of criminality once they reached adulthood. As compelling as these findings are, some aspects of the study bear replication and extension. We used data from longitudinal cohort studies in the U.S. (the Study of Early Childcare and Youth Development (SECCYD); n = 1,168, 51% male, 81% non-Hispanic White) and the U.K. (National Child Development Study; n = 16,506, 51% male, 98% White) to conduct two conceptual replications and extensions of Moffit et al. (2011), examining the links between behavioral regulation problems (i.e., impulsive aggression, hyperactivity/ impulsivity, and attention problems) in childhood and adult financial status, health, and criminal activity.
For the U.S. cohort, we carefully reviewed all measures from the SECCYD waves that most closely matched the ages of the Dunedin, namely ages 4, 5, 7, 9, and 11. We found appropriate matches for 37 of the 41 items reported as being used to assess children’s behavior in Moffitt et al. (2011). We drew on items from the Child Behavior Checklist (Achenbach, 1991), the Child Behavior Questionnaire (Rothbart et al., 2001), and an index of disruptive behaviors reflecting symptoms of Attention Deficit/Hyperactivity Disorder (American Psychiatric Association, 1994). For the U.K. cohort, we used items and scales from the Rutter Behavioral Scale (Rutter, 1967) and the Bristol Social Adjustment Guide (Shepherd, 2013), given at age 7 and age 11. Following Moffitt and colleagues (2011), we extracted the first axis of a principal components analysis to create an index of behavioral regulation problems across childhood. We then used regression models to predict adult outcomes, using multiple imputation in Stata 16 to adjust for non-random attrition.
We modeled our analyses on those in Moffitt et al. (2011), largely reproducing their pattern of findings that behavioral regulation problems in childhood predict a range of adult outcomes including educational attainment (βU.S. = -.22, βU.K. = -.13) and spending time in jail (ORU.S. = 1.74, ORU.K. = 1.48) after controlling for parental occupation, childhood IQ, child gender, race/ethnicity, and study site/region of birth. Associations with outcomes in education, work, and finances diminished in the presence of additional control variables for children’s home environment and achievement but associations for other outcomes were more robust (see Table 1). We also found that behavioral regulation problems across distinct periods of childhood were associated with adult outcomes (table not shown). Distinct components of behavioral regulation problems showed some specificity in predicting outcomes in the U.S. cohort, with attention problems predicting lower educational attainment and hyperactivity/impulsivity predicting ever spending time in jail (see Table 2). Together with the findings from Moffitt et al. (2011), our study makes clear that childhood behavioral regulation problems are associated with a range of outcomes in adulthood for cohorts born in the 1950s, 1970s, and 1990s across three countries.
 

Authors

• Andrew E. Koepp, University of Texas at Austin
  Presenting Author

• Tyler Watts, Teachers College, Columbia University

• Elizabeth T. Gershoff, University of Texas at Austin

• Sammy Ahmed, University of Rhode Island

• Pamela Davis-Kean, University of Michigan - Ann Arbor

• Greg J Duncan, University of California, Irvine

• Megan Kuhfeld, NWEA

• Deborah Lowe Vandell, University of California - Irvine