Different Pathways To Offending and Violence: An Examination Of The Differences Among Youths With Varying Histories Of Contact With The Juvenile Justice System

1
Different Pathways To Offending and Violence An
Examination Of The Differences Among Youths With
Varying Histories Of Contact With The Juvenile
Justice System¹Mitchell Sartin, ²Daniel L.
Ullman, and ¹David J. Hansen, ¹University of
Nebraska-Lincoln and ²Adolescent and Family
Services Unit - Lincoln Regional Center
Introduction Highly publicized occurrences of
violence and increased media attention to
juvenile crime and delinquency have contributed
to an upsurge in research and funding to address
what is seen by many as a growing problem.
Juvenile problem behaviors result in tremendous
costs to society in terms of capital and
resources. In fact, a minority of juveniles
commit a disproportionate percentage of offenses.
Research on conduct problems and subsequent
delinquency in adolescence has tended to
implicate several factors. For instance, parental
substance abuse, family conflict, child abuse,
and various socio-cognitive variables have been
found to be related to the development of conduct
problems. It is believed that an understanding of
how these variables interact in different
pathways across differing groups of youths could
permit enhanced prevention and intervention of
conduct problems and subsequent delinquency.
The current study is an attempt to
differentiate between youthful offenders who
experience minimal contact with the juvenile
justice system and those who experience repeated
and prolonged contact with the juvenile justice
system. Research has shown that a small
percentage of youth accounts for a large
percentage of the crimes and costs. In addition,
some delinquency in adolescence has even been
hypothesized by some to be normative. The current
study is an effort to identify the
non-normative delinquents, so that prevention
and intervention programs could be more
targeted. Prevention and intervention programs
are costly and largely ineffective. Perhaps part
of the problem lies in trying to utilize one
approach for vastly differing groups. The
identification of different pathways would allow
for better-targeted treatments. More thorough and
effective interventions could be implemented if
those who are at high risk to be continually
involved with the juvenile justice system could
be identified. Moreover, given the tendency for
the same youths to engage in multiple problem
behaviors (e.g., legal offenses, violence,
substance abuse, and other risky behaviors),
prevention and early intervention could provide
large savings in terms of capital and
resources.
Results In an attempt to find subgroups of
juvenile delinquents, particularly groups with
high numbers of offenses and a history of
violence, cluster analysis was utilized. Because
preliminary results (univariate and multivariate)
and previous literature suggest significant
differences across gender, males and females were
clustered separately. For both the males and
females, the variables used in the cluster
analysis were Z-scores of violence rating and a
function of number of offenses. That is, instead
of inserting merely number of offenses, a
separate variable (the square root of the
dividend resulting from dividing number of
offenses by age at time of admission) was
utilized. This coding was utilized for two
reasons. First, in order to understand the
implications of an individuals number of
offenses it is essential to factor in the
individuals age. Second, the square root was
utilized because the difference between no
offenses and 1 offense is more important than the
difference between 22 offenses and 23 offenses.
The profiles from the resulting clusters are
contained Figure 1 (males) and Figure 2
(females). Several variables with significant
differences across clusters are also presented in
Figure 1 and Figure 2. It appears that the very
high offenses, high violence groups across gender
are the most easily differentiated. For instance,
in comparison to other clusters of the same
gender, both of the very high offenses, low
violence clusters have the lowest age of first
legal charge and lowest written achievement
scores. Conclusions The results of the cluster
analysis give support to the idea that there are
subgroups of delinquents who differ in the
tendency to engage in delinquent behavior. There
appear to be groups of delinquents who have
infrequent contact with the legal system, groups
with moderate levels of contact with the legal
system, and groups who have frequent contact with
the legal system due to a pervasive tendency to
engage in delinquent behavior. Further, based on
the results, the subgroups of delinquents have
differing histories, abilities, and problems
(e.g., age of first legal charge, achievement
test scores, prevalence of conduct disorder). If
subgroups can be reliably determined, the
implication is that interventions can be more
targeted. That is, it would seem worthwhile to
target more potent interventions for those who
have had frequent contact with the legal system
or those who are likely to have frequent contact
with the legal system. Future research should
focus on ways to increase the reliability with
which subgroups can be determined. Moreover, it
will be important to understand how the pathways
to delinquency differ for the various subgroups.
Methods Participants Participants in the study
are 726 youths (57 male and 43 female), between
ages 11 and 18, who have undergone evaluation at
the Adolescent and Family Services Unit of the
Lincoln Regional Center after being referred by
the juvenile court. Demographic information for
this sample is presented in Table 1.
Measures The youths completed the Millon
Adolescent Clinical Inventory (MACI), a Wechsler
intelligence scale (i.e., Wechsler Adult
Intelligence Scale-III (WAIS-III), Wechsler
Intelligence Scale for Children-III (WISC-III),
or Wechsler Abbreviated Scale of Intelligence
(WASI)), and the Wechsler Individual Achievement
Test (WIAT). Procedures The youths were given a
clinical interview by a licensed clinical
psychologist and a chemical dependency evaluation
by an alcohol and drug counselor. Several
variables which have previously been found to be
related to conduct problems were coded from the
file and the clinical interview by individuals
working at the Adolescent and Family Services
unit of the Lincoln Regional Center parental
legal history, parental substance abuse, age of
first legal charge, and abuse history. These
variables were coded based on the files and
substantiation by police, court, Child Protective
Services (CPS), or other institutional records.
For this study a few additional variables were
coded from the files number of offenses (based
on number of non-status offenses) and violence
rating. A reliability check by an independent
rater was performed for the coding of these two
variables on 111 cases ( 15 of the cases). The
correlations between the two raters for number of
offenses and violence rating were .984 (p lt .01)
and .979 (p lt .01), respectively.
Table 2. Descriptive Statistics for Variables
Included in the Cluster Profile
Table 1. Demographics of the Study Participants