Mason et al.
ABSTRACT
Background: This study examined adolescent delinquency and alcohol use in relation to
young adult crime, alcohol use disorders (AUDs), and risky sex. Analyses further examined
the influences of late childhood involvement in these problem behavior outcomes, with
mediation through teen delinquency and alcohol use, and examined differences in the
pathways for youth from low‐ compared to middle‐income backgrounds.
Method: Multiple‐group latent growth curve modeling was conducted using data collected
from a sample of 808 youth followed from age 10 to age 24. Self‐report assessments
included delinquent involvement, alcohol use, and sexual activity in late childhood;
delinquency and alcohol use in adolescence; and crime, AUDs, and risky sex in early
adulthood.
Results: Late childhood delinquent involvement was associated with young adult crime,
AUDs, and risky sex indirectly through adolescent delinquency, and had a persistent direct
effect on crime. Adolescent delinquency also mediated the relation between early sex onset
and crime. Early alcohol use predicted a higher level of, and a faster rate of increase in,
adolescent drinking, which predicted, in turn, young adult AUDs and risky sex. Significant
group differences indicated stronger associations between adolescent delinquency and each
young adult outcome for youth from low‐ compared to those from middle‐income
backgrounds.
Conclusions: Early intervention may help prevent the development of crime, AUDs, and
risky sex behaviors, especially among disadvantaged youth.
Hypothesis
One hypothesis is that positive associations of adolescent delinquency and alcohol
use with young adult crime, AUDs, and risky sex will be stronger for youth from low-
income backgrounds compared to those from middle-income backgrounds. Here, the
assumption is that low-income youth lack access to resources and supports that may
help buffer the negative consequences of adolescent delinquency and alcohol use.
An alternative hypothesis is that youth from middle-income backgrounds have more
to lose in terms of social resources and capital; thus, their involvement in these
problem behaviors will be more strongly linked to adverse long-term outcomes by
making it difficult to return to adaptive functioning.
A third hypothesis is that the predictive relationships represent general developmental
processes that operate in the same manner for all youth, regardless of their income
background, resulting in no differences across subgroups.
To test these hypotheses, multiple group analyses of data collected from a diverse
sample of youth are conducted.
Measures
Childhood problem behaviors
o Early delinquent involvement (age 10)
o Early alcohol use (age 14)
o Early sex onset (age 14)
Adolescent delinquency and alcohol use (ages 14, 15, 16 and 18)
o Delinquency
o Alcohol use
Young adult outcomes (ages 21 and 24)
o Crime
o Alcohol Use Disorders (AUDs)
o Risky sex
Demographic characteristics
o Income background (ages 10, 11 and 12)
o Gender