wamfiwt-a‘.

. «xi ~ v‘E'.

Km!»

5-3?» 4:»; .5qu

u:
£
' I:

 

 

trig

_ kgvi-flf.

—-- at." a
.fi
v .15:

43% ‘
Sign

«37-...

 

This is to certify that the
dissertation entitled

FAMILIAL RELATIONS AND INTERVENTION AMENABILITY:
PAVING A PATH TO DELINQUENCY DESISTANCE

presented by
JUSTIN W. PATCHIN

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Criminal Justice

 

 

 

 

J Major Préssor’s éignature

Date "

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

mvogmm
@ttvns

 

 

JAI‘E‘ 192 2093
MR 2 2 7009

 

 

Age 3 12014

“45 i3

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

FAMILIAL RELATIONS AND INTERVENTION AMENABILITY:
PAVING A PATH TO DELINQUENCY DESISTANCE

By

Justin W. Patchin

A DISSERTATION
Submitted to
Michigan State University
In partial fitlfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
School of Criminal Justice

2004

ABSTRACT

FAMILIAL RELATIONS AND INTERVENTION AMENABILITY:
PAVING A PATH TO DELINQUENCY DESISTANCE

By

Justin W. Patchin

Serious childhood delinquents are among the most challenging populations facing
criminal and juvenile justice officials. Intervention efforts aimed at young offenders are
lacking, and many existing programs have proven unsuccessful at rehabilitating this
difficult group. Failure to address the antisocial behavior of young offenders may result
in a population of inveterate adult criminals. This study examines the efficacy of an
intervention that targeted serious and violent young offenders who were between the ages
of 10-13 years old when arrested for a second offense. Because parents are an important
influence in adolescent development at this age, it is argued that interventions targeting
young populations must incorporate family-based programming that seeks to improve the
familial environment. To this end, it is maintained that youth involved in an intervention
at this developmental stage must be kept in their families if possible. The current
analysis investigates the way in which elements of the family (family bond and family
structure) and, the intervention (family-focused, community-based programming) are

associated with delinquency desistance.

Cepyright by
Justin W. Patchin
2004

To Jill, for standing by me through it all.

iv

ACKNOWLEDGEMENTS

I greatly appreciate the guidance I have received fi'om Dr. Tim Bynum, Dr. Christina
DeJong-Schwitzer, Dr. Cynthia Perez McCluskey, and Dr. Bill Davidson. Your hands
have helped to shape this work into something beyond what I had envisioned. I want to
extend a special thanks to my wife Jill and my parents Mike and Kathie for continuing to
support me during this “extended adolescence” which is graduate school. The love and
encouragement of my family has sustained me at difficult times. I have also been
fortunate to have developed many amazing friendships over the past five years. Instead
of following convention and listing you all here, I will save several trees and thank you
all in person the next time I see each of you. Finally, I want to thank God for being in my
life everyday. But for His grace, I may have been the subject rather than the author of

this work.

TABLE OF CONTENTS

 

 

 

 

 

 

LIST OF TABLES viii
LIST OF FIGURES x
CHAPTER ONE: INTRODUCTION 1
Introduction ..................................................................................................................... 1
Family Relations and Delinquency ................................................................................. 3
Serious, Violent, and Early Juvenile Offending ........................................................... 17
Discussion ..................................................................................................................... 25
CHAPTER TWO: THEORETICAL FRAMEWORK 27
Introduction ................................................................................................................... 27
Social Bond ................................................................................................................... 28
Social Development Model ........................................................................................... 35
Coercion Theory ........................................................................................................... 39
General Theory of Crime .............................................................................................. 42
Informal Social Control ................................................................................................ 47
Theory to Practice: Family-Focused Interventions ....................................................... 51
Discussion ..................................................................................................................... 53
CHAPTER THREE: DATA, METHODS, AND ANALYTIC STRATEGY ........... 55
Introduction ................................................................................................................... 55
Study Context ................................................................................................................ 55
Sample ........................................................................................................................... 6O
Hypotheses .................................................................................................................... 66
Analytic Strategy .......................................................................................................... 7O
Dependent Variables ................................................................................................. 70
Independent Variables .............................................................................................. 72
Control Variables ...................................................................................................... 75
Quantitative Methods and Models ............................................................................ 75
Summary ....................................................................................................................... 79
CHAPTER FOUR: FINDINGS 80
Descriptive Statistics ..................................................................................................... 80
Hypothesis 1: Family Bond at Intake and Delinquency Desistance ............................. 84
Hypothesis 2: Family Structure at Intake and Delinquency Desistance ....................... 88
Hypothesis 3: Family Based Programming and Delinquency Desistance .................... 90
Hypothesis 4: Change in Family Bond and Delinquency Desistance ........................... 95
Additional Analyses: Hierarchical Linear Modeling .................................................... 99
CHAPTER FIVE: CONCLUSION 108
Introduction ................................................................................................................. 108
Theoretical Implications ............................................................................................. 110
Policy Implications ..................................................................................................... 112

vi

Limitations .................................................................................................................. 1 15

 

Conclusion .................................................................................................................. 120
APPENDICES - 122
APPENDIX A: Survey Instrument ............................................................................. 123
APPENDIX B: Correlation Matrices. ......................................................................... 146
APPENDIX C: City and Target Area Characteristics. ............................................... 147
REFERENCES 148

 

vii

LIST OF TABLES

Table l. Qualifying Offense Information (83 youth) ....................................................... 63
Table 2. Type of Qualifying Offense (83 youth) ............................................................. 63
Table 3. Attrition Diagnostics ........................................................................................... 66
Table 4. Family Bond Component Variables and Factor Loadings .................................. 74
Table 5. Descriptive Characteristics of the Treatment and Comparison Groups ............. 81
Table 6. Family Relationships by Group .......................................................................... 82
Table 7. Self-Reported and Official Offending Characteristics ........................................ 83
Table 8. Bivariate Logistic Regression — Family Bond at Intake (Treatment Group) ...... 86

Table 9. Multivariate Logistic Regression — Family Bond at Intake (Treatment Group). 86

Table 10. Bivariate Logistic Regression — Family Bond at Interview 1 (Comparison

Group; N=83) .................................................................................................................... 87
Table 11. Multivariate Logistic Regression — Family Bond at Interview 1 (Comparison
Group; N=83) .................................................................................................................... 87
Table 12. Contingency Table Analysis: Family Structure at Intake and Future
Delinquency (Treatment Group; N=68) ............................................................................ 88
Table 13. Contingency Table Analysis: Family Structure at Intake and Future
Delinquency (Comparison Group; N=83) ........................................................................ 89
Table 14. Bivariate Logistic Regression — Family Programming (Treatment Group;
N=68) ................................................................................................................................ 93
Table 15. Multivariate Logistic Regression — Family Programming (Treatment Group;
N=68) ................................................................................................................................ 93
Table 16. Self-reported Delinquency over Time by Group ............................................. 94

Table 17. Bivariate Logistic Regression — Change in Family Bond (Treatment Group;
N=68) ................................................................................................................................ 97

Table 18. Multivariate Logistic Regression - Change in Family Bond (Treatment Group;
N=68) ................................................................................................................................ 97

Table 19. Bivariate Logistic Regression — Change in Family Bond (Comparison Group;
=83) ................................................................................................................................ 98

viii

Table 20. Multivariate Logistic Regression - Change in Family Bond (Comparison

Group; N=83) .................................................................................................................... 98
Table 21. HLM: Final Estimation of Fixed Effects - Family Bond Over Time (N=217)
......................................................................................................................................... 101
Table. 22. HLM: Final Estimation of Fixed Effects — Family Bond Over Time (N=217)
......................................................................................................................................... 1 02
Table 23. HLM: Final Estimation of Fixed Effects — Demographic Controls and Variety
of Delinquent Activities (N=217) ................................................................................... 103
Table 24. HLM: Final Estimation of Fixed Effects — Family Bond at Time 1 and Variety
of Delinquent Activities (N=217) ................................................................................... 104
Table 25. HLM: Final Estimation of Fixed Effects — Family Bond Change over Time and
Variety of Delinquent Activities (N=217) ...................................................................... 106
Table B1. Correlation Matrix (Treatment Group; N = 68) ............................................. 146
Table B2. Correlation Matrix (Comparison Group; N = 83) .......................................... 146
Table 2. Descriptive Statistics for the City and Target Area .......................................... 147

ix

LIST OF FIGURES

Figure 1. HLM Two-Level Model .................................................................................. 100

Figure 2. HLM: Discrepancy within Individuals over Time (Family Bond at Time 1). 105

CHAPTER ONE: INTRODUCTION

Introduction

For decades social scientists have attempted to identify and understand the
etiology of juvenile crime and to determine factors that contribute to the persistence or
desistance of antisocial behaviors. A better understanding of such factors would allow
practitioners to develop effective interventions to ameliorate the antecedent conditions of
aggressive and delinquent behavior. One such factor that has been identified as playing a
pivotal role in the development or cessation of deviant tendencies is the relationship
between children and their parents.

During the 1950’s, several researchers examined the role of families in
delinquency causation. While largely unpopular at the time (Laub & Sampson, 1991;
Sampson & Laub, 1993), Sheldon and Eleanor Glueck (1950; 1952; 1962) provided
much of the impetus for focusing on family relationships and delinquency causation.
Others at that time were also involved in research attempting to disentangle the
association between family relationships and crime (Nye, 1958; Reiss, 1951). Recently,
however, there has been increased attention directed toward the role parents play in
delinquency prevention (Anderson, 2002; Bank & Burraston, 2001; Cemkovick &
Giordano, 1987; Gerard & Buehler, 1999; Gonnan-Smith, Tolan, Zelli, & Huesmann,
1996; Gorman-Smith, Tolan, Loeber, & Henry, 1998; Heck & Walsh, 2000; McCord,
1991; Thomberry et at., 1999; Wright, Cullen, & Miller, 2001).

Concomitant with this interest in the relationship between family functioning and

crime in recent years has been a concern that interventions aimed at rehabilitating

delinquent youth have largely failed (Gottfredson & Hirschi, 1990; Whitehead & Lab,
1989). Particularly obstinate are serious and violent offenders who seem impervious to
efforts at intervention. The current study examines why some interventions are
unsuccessful, and argues that one problem is that they are not theoretically grounded. As
will be discussed in detail, several criminological theories consider the family an
important socializing institution. Despite this body of knowledge, few intervention
programs consider the familial environment within which the youth is embedded.
Indeed, it is further argued that, among early (preadolescent) offenders, the most salient
environment is the family.

In general, this study will attempt to further investigate the ways in which family
relationships can provide a supportive environment to serious juvenile offenders within
an intervention context. It is argued that the familial environment must be considered in
any comprehensive intervention aimed at preadolescent offenders. As such, it is
hypothesized that individuals who report having “strong” relationships with their parents
are likely to fare better in intervention programs than those who do not. Characteristics
of strong familial relationships include: regular communication, respect, trust, and
general positive emotional affect. It is hypothesized that youth who begin an intervention
experience already having a strong relationship with their parent(s) may not require
family-based programming to the extent that others do. A related hypothesis examines
the extent to which youth who receive family—based programming or otherwise improve
their relationship with their parent(s) during an intervention are more likely to desist

offending.

Specifically, this study seeks to determine whether serious young offenders who
were involved in a comprehensive, community—based intervention were differentially
successful at desisting offending based on specific family characteristics (family bond
and structure) and whether or not they received family-based programming. In
determining the importance of family relationships on intervention amenability, this
study will proceed in several stages. First, this chapter will provide a broad overview of
the extant research on families and delinquency, and review what is known about serious,
violent, and early juvenile offenders, the sample being targeted in the current analysis.
Chapter Two will discuss the theoretical foundations of a relationship between family
functioning and delinquency. Focusing primarily on control-based theories, several
popular perspectives will be examined insofar as they incorporate familial relations as a
primary component in their explanatory models. Next, Chapter Three will discuss. the
current sample and methods employed to test the relationship between family relations
and intervention arnenability among young offenders. Chapter Four will present the
results of the statistical tests performed to assess the importance of family relationships in
juvenile programming. Logistic regression and Hierarchical Linear Modeling will be
employed in this regard. Finally, Chapter Five will discuss the strengths and limitations
of the current study, and point to specific theoretical and policy implications to be drawn

from its results.

Family Relations and Delinquency
In all cultures, historical and modern, the family has been, and continues to be, the

central socializing institution that is responsible for instilling in youth a set of norms,

values, beliefs, and ideals (Loury, 1987). The failure of families to accomplish this task
can result in negative consequences for the individual and society at large. This section
will review the family and delinquency literature for the purpose of better understanding
the association between family relationships and juvenile delinquency. Explored are
several ways in which a negative family environment can be a risk factor for children to
engage in delinquent activities and ways that effective parenting can act as a buffer to
insulate at-risk youth from participating in deviance.

Dynamics within the family can influence whether a youth will engage in anti-

 

normative activities. For example, children of parents who are constantly quarrelling or
are otherwise incompatible with each other are more likely to develop negative traits such
as extreme restlessness and destructiveness—both common antecedents to delinquency
(Glueck & Glueck, 1962). Also, lack of emotional ties between parents or between
parents and children contribute to involvement in maladaptive behavior (Hirschi, 1969;
Glueck & Glueck, 1962). Needle, Su, Doherty, Lavee, and Brown (1988), for example,
identified family instability, deficient family cohesion, and lack of quality relationships
between parents and children as predictors of adolescent substance use. Loeber and
Stouthamer—Loeber (1986) identified four paradigms of how the family can negatively
influence adolescent behavior. These areas include: neglect, conflict, deviant behaviors
and attitudes, and disruption. The following sub-section will discuss each of these
paradigms in turn.
The Neglect Paradigm

Parents who fail to enforce their demands and lack control over the individual

child both inside and outside the home may foster delinquent tendencies (Nye, 195 8).

Neglectful parents either explicitly ignore the behaviors in which their children are

engaging, or are unprepared to effectively address the behavior. Specifically, parents
who are afraid to cause tension in the family often allow their children too much latitude
in conduct. Instead of punishing children for breaking curfew, for example, neglectful
parents regularly look past these seemingly trivial violations. Overly permissive parents
are usually indulgent to their children and give in to their every desire. These parents fail
to set limits to provide proper structure and regulate boundaries within which their

children should behave (Cobb, 2001). As such, permissive parents are loved by their

 

children because few constraints are placed on the child. There are, of course, negative
ramifications stemming fi'om this parental indifference. Due to this lack of familial
organization, children of permissive parents often have trouble developing fiiendships
and lack the ability to regulate their emotions (Cobb, 2001). These issues can manifest
themselves in frequent quarrels at school or involvement in other delinquent behavior
(i.e., shoplifting) as the youth attempts to gain fiiends.

Two common forms of neglect are lack of supervision and lack of involvement.
A large body of literature demonstrates the deleterious effects of improper parental
supervision (Glueck & Glueck, 1950; Hirschi, 1969; Patterson, 1980). Effective parents
are aware of their children’s activities, the fiiends they associate with, and the places they
hang out (Cobb, 2001). Parents who do not adequately supervise their children will be
unable to effectively control their behavior through positive reinforcement and
appropriate punishment. In their meta-analysis of longitudinal studies, Loeber and
Stouthamer-Loeber (1986:29) found that parental supervision was “. . .among the most

powerful predictors of juvenile conduct problems and delinquency.” In another study,

youth who reported that their parents knew where they were and who they were with
were less likely to engage in delinquent activities (Hirschi, 1969). It has also been
suggested that inadequate or neglectful supervision contributes to association with
deviant peers.

In addition, attachment to one’s parents can result in decreased delinquency
through a process known as “virtual supervision” (Hirschi, 1969). Virtual or indirect
supervision occurs when children who are strongly attached to their parent(s) consider the
response of their parents when participating in certain activities, even if a parent is not
immediately supervising them. Fearing a negative response from a parent who may learn
of proscribed activities, youth will often fail to engage in forbidden activities.

Parental involvement is also an important protective factor. On the other hand,
parents who are not actively involved in activities with their children may be increasing
their risk for delinquency. Of the twenty-nine studies reviewed by Loeber and
Stouthamer-Loeber (1986), twenty-two reported a significant association between lack of
parental involvement and delinquency. Moreover, this relationship remained whether
considering self-reported or official delinquency and a number of different outcome
measures (e.g., frequency and variety of offenses).

The Conflict Paradigm

Within the conflict paradigm, parents demonstrate overly harsh disciplinary
approaches ranging fiom emotional abuse to extreme physical assault. As a result of
inconsistent or harsh disciplining, children often rebel and escalate disruptive behavior.
A reciprocal relationship ensues whereby both parent and child view each other as

enemies. Frustrated adolescents may look toward their peer networks or other sources for

support in environments where parents are abusive. These relationships, then, lead to
higher rates of delinquency, substance abuse, and the development of inappropriate
sexual identities (Bank & Burraston, 2001; Hill, Howell, Hawkins, & Battin-Pearson,
1999; Wallace & Bachman, 1991; Warr, 2002).

An abusive home environment can be characterized in many different ways.
Youth can be physically, sexually, or emotionally abused and/or neglected. Even if not
physically injured, there is consensus that maltreatment can result in lasting emotional
and cognitive damage (Hunner & Walker, 1981). In a comprehensive review of literature
examining the link between child maltreatment and delinquency, Knutson (1995)
concluded that youth who are seriously maltreated are at risk to becoming antisocial
adolescents and aggressive adults (see also Besharov, 1987; McCord, 1989).
Interestingly, Ireland, Smith, and Thomberry (2002) found that maltreatment in
childhood only was not significantly related to delinquency and drug use in late
adolescence, however adolescent or persistent maltreatment was significantly related to
these negative outcomes. This finding stresses the importance of research from a
developmental perspective which identifies the timing and duration of life events as
important considerations.

A wealth of empirical information exists indicating that abusive discipline
practices are related to a “lack of parenting skills, social and economic disadvantage, and
parental psychopathology and substance abuse” (Bank & Burraston, 2001 :196). In this
way, an abusive environment often represents a constellation of negative stimuli that can
affect youth. Moreover, inconsistent discipline may indirectly affect the long-tenn

development of youth by increasing the likelihood of maltreatment, neglect, and sibling

conflict (Bank & Burraston, 2001; Patterson, 1980). High rates of sibling conflict, for
instance, has been identified as being significantly associated (r=.80) with antisocial
activities in adolescence and criminality in adulthood (Bank & Burraston, 2001). Sibling
antisocial behavior was also predictive of adolescent gang membership (Hill et a1, 1999).

As these studies indicate, parents who severely punish their children may put
them at a greater risk for firture delinquency (Miller & Knutson, 1997). Additionally,
some parents may punish harshly at one time but ignore the same behavior at another.
This disciplining style can actually teach a difficult child to become more oppositional as
undesirable behaviors are inconsistently punished (Wahler, 1987). Children become
unclear as to what behavior is acceptable and what is not. A proper balance, then, must
involve appropriate levels of discipline whereby the youth understands the purpose
behind such response (Patterson, 1980).

The myriad number of negative consequences associated with child abuse are
important and well documented, and, as such, the topic warrants more consideration than
is available here (for reviews, see Jonson, 1998; Knutson, 1995; Salrnelainen, 1996;
Veltrnan & Brown, 2001). As Bank and Burraston (2001:211) note, “when we maltreat
our kids, they are more likely to get hurt, to get into trouble with the law, and to harm
others.”

The Deviant Behavior and Attitudes Paradigm

Parents who are themselves deviant may implicitly or explicitly pass these
behaviors on to their progeny (Hirschi, 1969). While the causal process is not completely
understood, research indicates that delinquents are more likely than nondelinquents to

have been raised by criminal fathers (Glueck & Glueck, 1962; Hirschi, 1969). It is

unclear whether childhood deviance is learned from parents who are involved in
antisocial activities (Bandura, 1986; Burgess & Akers, 1966; Skinner, 1953); if there is
something about the environment from within which parents and children are raised
(Brooks-Gum, Duncan, Klebanov, & Sealand, 1993); or if there is some biological
component to crime (Wilson & Hermstein, 1985). Indeed, it is likely a combination of
these factors which affect the developmental outcomes of children. Whatever the cause,
criminological consensus concludes that parental deviancel is a strong correlate of

adolescent offending.

 

When one or both parents are involved in criminal behaviors or substance abuse,
similar propensities may be fostered among their children. Sampson and Laub (1993)
also found evidence that parental deviance pushes children towards deviant peers.
Similarly, youth who are reared by alcoholic fathers are more likely to develop
criminogenic traits such as hostility and unconventionality and, correspondingly, are at a
greater risk for delinquency (Glueck & Glueck, 1962). Pathology among mothers also
plays an important role in the development of deviant predilections. Indeed, histories of
criminal involvement and alcoholism of the mother was found to be more prevalent in
delinquent youth when compared to nondelinquent youth (Glueck & Glueck, 1962). In
fact, the negative effects of a pathologic mother appeared more influential in predicting
delinquency than pathology of the father (Glueck & Glueck, 1962). It has also been
found that younger children are more susceptible to the pathology of their mothers while
older children are often more affected by the behavior of their fathers (Connell &

Goodman, 2002).

 

' It is important to note that deviance (or deviant behavior) is not used interchangeably with other, more
specific terms, such as offending, delinquency, or crime. Instead, it is used to suggest a broader form of
maladaptive behavior that may or may not be regulated by law.

 

Growing up with parents who are openly involved in deviant activities can also
have detrimental effects for youth as they develop their own identity. A major step in the
development of an adolescent’s identity involves the differentiation of self from the ego
mass of family (Bowen, 1978). Familial conflict associated with parental involvement in
deviant activities can have at least two negative consequences for youth: (1) the child
may learn about deviant activities from his or her parents or at least perceive that they are
normal behaviors; or, (2) if a child recognizes that the behavior is counter to what he or
she has learned elsewhere (schools or friends) he or she may detach from the familial unit
too early. Either trajectory can lead to other potentially problematic outcomes. For
example, one negative manifestation involves retreating to the acceptance of a deviant
gang. To illustrate, Hill and colleagues (1999:302) found that gang members reported
“frequent conflict and abuse among their parents, child abuse, family member alcoholism
and drug addiction, and family trouble with the police.”

The relationship between parental deviance and the antisocial behavior of children
may suggest a biologically determined component to crime causation (Wilson &
Herrnstein, 1985). That is, some genetic predisposition towards criminality is passed
from parent to child. Indeed, there are considerable volumes of research suggestive of a
relationship between biology and crime (Mednick & Christainsen, 1977; Wilson &
Herrnstein, 1985). Sampson and Laub (1993) argued, instead, that deviant parents put
their children at risk for delinquency not because of some genetic trait, but because their
parenting is compromised due to their own illicit activities. For example, inebriated
parents cannot effectively supervise their children and may punish inconsistently or

harshly. Indeed, Sampson and Laub’s (1993) analyses supported the assertion that

10

 

parental deviance negatively affects their discipline and supervision abilities (see also
Patterson, 1980). Further, Sampson and Laub (1993:69) argued that “Parenting is
perhaps the most demanding of conventional roles, and we expect that deviance in the
adult world will manifest itself in disrupted styles of child socialization.” Finally, strong
parenting skills can buffer youth with at-risk traits, such as irritability or poor
temperament, from engaging in antisocial activities. As such, the biological argument is
largely irrelevant once proper parenting is considered (Sampson & Laub, 1993).

Gottfredson & Hirschi (1990) have a different interpretation of the relationship
between parental deviance and childhood delinquency. These theorists suggested that
adults who are deviant are also poor at child rearing. Both of these behaviors are
manifestations of the same latent trait: low self-control. Adults with low self-control,
therefore, often have children who have been improperly trained, and as a result, have
low self-control themselves. Similarly, Moffitt (1993:681) suggests that “. . .parents of
children who are difficult to manage often lack the necessary psychological and physical
resources to cope constructively with a difficult child.” Patterson (1982:11) concured:
“Parents of aggressive children are generally unskilled, but they are particularly unskilled
in their use of punishment for deviant behavior.” This argument is also similar to
Sampson and Laub’s (1993) claim that deviant adults simply lack effective parenting
skills.

Because there are few children who grow up away from their deviant parents, it is
often difficult to distinguish between the role of environment and biology in producing
delinquent children (Loeber, 1987). Despite these limitations, Rowe and Osgood

(1984:535) noted that “The most convincing evidence for genetic influence on antisocial

11

behavior comes from studies of adopted children who were separated at birth fiom their
biological parents.” These studies indicated that adoptive children whose biological
parents were criminal are more likely to become criminal than children of adoptive
parents who are criminal. Moreover, regardless of criminality of adoptive parent,
children whose biological parent was criminal are more likely to become criminal (see
also, Gottfredson & Hirschi, 1990:53-56). These findings have been interpreted to
suggest that there is a genetic component predisposing some youth to engage in deviant
activities.

The Disruption Paradigm

Disruption within the familial unit can stem from several sources. Arguing,
ailing, mentally ill, or altogether absent parents can be difficult for children to cope with.
Some of theses stressors fall within the other paradigms summarized above. The most
commonly studied form of disruption, however, is that resulting from the absence of one
or both biological parents.

Some researchers have suggested that there exists a “two-tiered system of
childrearing” (Sokol-Katz et al., 1997:200; Tienda & Angel, 1982). One system consists
of a two parent household, also termed, “intact homes.” Tire other system is comprised
of single-parent households where one individual (commonly the mother) is responsible
for all aspects of parenting, also referred to as “broken homes.” While the former system
is the traditional structure, the latter is becoming increasingly common. The effects of
these changes in terms of adolescent deve10pment are still being explored.

Because rising crime rates have generally occurred concomitantly with changes in

family composition in the previous two decades, concern has been raised that the two

12

phenomena are in some way related. Indeed, research into the effect of family disruption
on adolescent development is widespread (Johnson, 1986; Juby & Farrington, 2001;
Quensel et al., 2002; Rankin, 1983; Rebellion, 2002; Sampson & Laub, 1993; Wells &
Rankin, 1991), and investigators have suggested that family disruption is more common
in delinquent samples (Glueck & Gleuck, 1950). Additionally, the relationship between
broken homes and delinquency has been supported by official and self-report studies and
by several longitudinal studies (J uby & Farrington, 2001).

It has been estimated that approximately 40% of white children and 75% of
African American children will experience parental separation or divorce before the age
of 16 (Bray & Hetherington, 1993). Currently, about 28% of adolescents in the United
States live in a one parent household (Cobb, 2001). Of these, about 84% live with their
mothers (Cobb, 2001). A variety of studies have demonstrated the negative
consequences associated with families that have been disrupted by desertion, divorce, or
death (Glueck & Glueck, 1962; Thomberry et al., 1999). For example, youth from one-
parent families are more likely to use drugs (Wallace & Bachman, 1991) or engage in
delinquent activities (Wells & Rankin, 1991) than youth from two-parent families. More
specifically, Wells and Rankin (1991) found that youth residing in single-parent
households were involved in 10-15% more delinquent activities than adolescents from
intact homes. Rebellion (2002) presented evidence that Wells and Rankin’s estimation
may be low. Also, children who lived with one parent and his or her significant other
(who was not related to youth) were more likely to join a gang in adolescence (Hill et al.,
1999). Intuitively, children with only one parent often have more unsupervised time in

which they could become involved in delinquent activities. As discussed above, parental

l3

 

supervision (or lack thereof) has been highly predictive of delinquent behavior (V azsonyi
& F lannery, 1997).

The negative effects of a single parent household also vary depending on with
whom the youth is residing. Historically, much concern has been raised regarding
separation of children from their mothers. As one theorist remarked:

. . .on the basis of this varied evidence it appears that there is a very strong case

indeed for believing that prolonged separation of a-child from his mother (or

mother-substitute) during the first five years of life stands foremost among the
causes of delinquent character development and persistent misbehavior (Bowlby,

1946, as quoted in Hirschi, 1969:86).

It has also been argued that young men who do not have the support of a “father figure”
are more likely to engage in aberrant activities (Davies & Sinclair, 1971). In addition,
Cantor (1982) found that the effects of a broken home on delinquency were stronger for
boys than for girls.

Contrary to previous studies which found that attachment to mother was more
important than attachment to father in reducing delinquency (i.e., Johnson, 1987; Krohn
& Massey, 1980), Rankin and Kern (1994) argued that it is not the gender of the parent
that is important, but the ntunber of strong attachments. More specifically, children
attached to both parents are less likely to be delinquent than youth attached to only one
parent - even in families where both parents are still present (Rankin & Kern, 1994).
Hirschi (1969:103), for example, maintained that the importance rests in the bond the
youth maintains with the parent, and that “. . .the one parent family is virtually as efficient
a delinquency-controlling institution as the two-parent family. . ..” On the other hand,

some researchers have suggested that the broken home creates fewer opportunities for

strong attachment between the child and parent, and renders the child less likely to

14

internalize familial norms (Elliott, Huizinga, & Ageton, 1985). Moreover, Rankin and
Kern (1994) found that youth with attachment to two parents were less delinquent than
youth who maintained a strong bond to one parent. Sokol-Katz and colleagues
(1997:212) conclude that “it is possible to have a broken or reconstituted family that
provides attachment and belief better than do some intact families, resulting in lower
levels of delinquency.”

Others have indicated that the presence of a stepparent places a child at risk
because children are less likely to be strongly attached to nonbiological parents
(Gottfredson & Hirschi, 1990; McCarthy, Gersten, & Langner, 1982; Rankin, 1983;
Rebellion, 2002). Gottfredson and Hirschi (1990) argued that, all else being equal, a
single, biological parent can be sufficient. Indeed, it is the ability of that parent to
properly socialize his or her child such that self-control is developed that is important.
These theorists asserted that one parent can accomplish this task as efficiently as two.
Gottfredson and Hirschi (1990) also noted, however, that a single parent may find it more
difficult to supervise his or her child at all times; yet, despite this disadvantage, they can
succeed at properly socializing their children. In fact, Gottfredson and Hirschi further
pointed out that a parent is not the only adult that can effectively socialize children;
others from outside the home such as a grandparent or a mentor can also aid in the
development of self-control.

Rankin (1983) also found a relationship between relatively minor antisocial
behaviors (truancy, running away, vandalism, fighting) and family structure.
Specifically, Rankin (1983) noted that homes in which both biological parents were

absent were more likely to have a child who engaged in these activities compared to a

15

home with only one absent parent. As Reiss (1951:199) argued, “It is generally agreed
that children develop and maintain personal controls less readily when raised in a milieu
other than the family of procreation.”

Research cited on each of the family risk factors and correlates discussed above
demonstrates that each is potentially hazardous by themselves. Often, however,
adolescents are barraged with multiple risk factors over the course of their development.
For example, Hirschi (1969) found that both prosocial and antisocial behaviors were
highly correlated between parents. Similarly, Sampson and Laub (1993) found that
father’s deviance had a negative effect on mother’s parenting. These findings suggest
that youth who live with both parents may not necessarily be less at risk; if one parent
demonstrates poor parenting skills or is deviant, it is very likely that the other one acts
similarly or is in some other way negatively impacted. Reviewing much of the evidence
available at the time regarding the above risk factors, Loeber and Stouthamer—Loeber
(1986:92) conclude “. . .the risk of child problem behavior increases rapidly as the number
of handicaps in the family increases.”

Based on the preceding review, it is clear that the family is one of the most
important socializing agents in developing adolescents. Youth are especially influenced
by their parents and others within their family during their pro-adolescent years. By the
time these youth enter adolescence, however, their attention tends to shift to other sources
of influence (i.e. peers; Warr, 2002). A good familial environment early in life is crucial,
because it can buffer youth from other risks that will eventually emerge. As explicated in
this section, youth can be positively or negatively persuaded by what they learn within

the family. Poor parenting predicts adolescent drug use, involvement in delinquency, and

16

association with deviant gangs. Conversely, positive parenting protects youth fiom these
and other negative influences. Proficient parenting can protect “. . .even biologically
vulnerable and socioeconomically disadvantaged children” (Westrnan, 1996).

During an impressionable child’s formative years, he or she is inundated with a
host of positive and negative influences. The daunting task for each child is to sort
through the good from the bad and make decisions based on experience, instruction, and
intuition. Parents are instrumental in either directly or indirectly shaping these choices.
As noted throughout this section, youth who fail to receive proper guidance or.
socialization are at an elevated risk to participate in delinquent behaviors. A small, yet
significant proportion of these youth may engage in serious and/or violent behaviors.

The following section will discuss the characteristics of these offenders.

Serious, Violent, and Early Juvenile Offending

Serious and violent juvenile offenders pose the greatest challenges to the juvenile
and criminal justice systems. These offenders are responsible for a disproportionate
amount of deviance and, as a result, a significant amount of fiscal resources are exhausted
due the behavior of a small group of individuals (Farrington, 1987; Farrington & Loeber,
1998; Loeber, Farrington, & Waschbuch, 1998; Wolfgang, Figlio, & Selling, 1972).
Most often, involvement in deviant activities peaks in post adolescence, between about
ages 16 and 18 years (Farrington, 1986; Hirschi & Gottfredson, 1983). Others have
noted, however, that criminal behavior can begin early in one’s deve10pment and

continue throughout the life-course (Sampson & Laub, 1993). Moreover, those youth

17

who initiate offending at an early age (prior to age 14 years) are at an increased risk to
become serious, violent, and/or chronic offenders (Mof’fitt, 1993).

Generally, juveniles who engage in Type I index offenses (homicide, rape,
robbery, aggravated assault, burglary, theft over $100, motor vehicle theft, arson) are
characterized as serious and/or violent offenders. Additionally, this group of offenders is
often involved in a number of other, less serious, offenses (Loeber & F arrington, 1998).
Early starters are those youth who initiate offending prior to age 13 years (Loeber &

F arrington, 2001). Serious offenders who begin at this early age are among the most at-
risk to participate in illegal activities throughout their life course. This section will
discuss the characteristics of serious, violent, and early offenders, focusing on the causes
and correlates of their behavior and what the future holds for such high-risk youth.

Many researchers have argued that age of onset is an important factor in
determining the nature of juvenile offending (Glueck & Glueck, 1950; Sampson & Laub,
1993). Specifically, youth who initiate offending early are at an increased risk to have a
prolonged career of offending that may continue into adulthood (Mazerolle, Brame,
Paternoster, Piquero, & Dean, 2000). Sampson and Laub (1993:135), for example,
presented evidence that “. . .delinquent behavior in childhood has significant and
substantial relationships with a wide range of adult criminal and deviant behaviors,
including charges initiated by military personnel, interview-based reports of involvement
in deviance and excessive drinking, and arrest by the police up to 30 years later.”

Others, however, have argued that the age-crime curve is invariant across race and
sex and has not changed in over one hundred years of research (Hirschi & Gottfredson,

1983; Gottfredson & Hirschi, 1990). As such, consideration of early adolescent behavior

18

 

is unnecessary. Many tests of this assertion have revealed, however, that age is important
(Greenberg, 1985). For example, Steffensmeier and his colleagues (1989) reviewed the
FBI’s Uniform Crime Reports for the years 1940, 1960, and 1980 and discovered that the
age-crime curve is in fact changing (the peak has steadily moved younger over the years).
In short, while some traditional theorists have largely ignored the early years of child
development, and others have argued that age is an unnecessary component of theoretical
analysis, recent research questions these assumptions.

As a result of this debate, many researchers have developed and tested
developmental theories that account for stability and change of deviant tendencies
throughout the life course (Loeber et a1, 1993; Patterson et al, 1989; 1992; Sampson &
Laub, 1993). Two influential studies provided empirical support for this need (Blumstein
et al., 1986; White, Moffitt, Earls, Robins, & Silva, 1990). In 1986, Blumstein and
colleagues began investigating what they termed “career criminals.” These individuals,
they argued, represent a small minority in the population (5%), but are responsible for the
majority of crimes that occurred (see also Wolfgang, F iglio, & Sellin, 1972). Ifsociety
could successfully intervene with this small group, they reasoned, there would be a
dramatic decrease in crime. Similarly, White and colleagues (1990) found that the best
predictor of later adolescent offending (and consequently adult offending) was offending
early in the life course (preschool).

Delinquency committed by children is becoming increasingly common. For
example, in the United States, juvenile courts saw 33% more child delinquents in 1997
compared to 1988 (Loeber & F arrington, 2001). In Denver and Pittsburgh,

approximately 10% of youth with a police record are either 11 or 12 years old (Espiritu,

19

Huizinga, Crawford, & Loeber, 2001). Even though many children are not officially
charged with a crime and referred to the juvenile court (Loeber & Farrington, 2001), over
250,000 arrests were made of individuals under the age of 13 in 1997 (Snyder, 2001).
While some of these arrests were of the same individual, Snyder (2001) estimates that in
the United States about 1% of the youth aged 7 -12 were contacted by the juvenile justice
system in 1997.

In general, causes of crime for early offenders typically occur within the
individual (traits, biological or psychological — impulsivity, negative emotionality,
constraint; Caspi et al., 1994) or in the home (poor parenting), while correlates for later
crime come from the environment (sociological explanations — peers, strain, etc). More
specific examples will be detailed below.

Because crime tends to peak in late adolescence (property offenses earlier than
personal offenses), many criminological theories focus on the adolescent years (or at least
employ adolescent samples in testing their propositions). For example, differential
association theory (Sutherland, Cressey, & Luckenbill, 1992) asserts that youth who
associate with fiiends who have definitions favorable to the violation of the law will be
more likely engage in delinquent activities than youth who associate with friends who
have definitions unfavorable to the violation of the law (see also Matsueda, 1988). In
addition, whether or not one becomes delinquent depends on the frequency, duration,
intensity, and priority of these relationships (Sutherland & Cressey, 1947). Because
youth generally do not develop firm fiiendships until early adolescence — or later, and

most of their influence at this time comes from the home (Glueck & Glueck, 1950,

20

Hirschi, 1969, Nye, 195 8), differential association does not consider preadolescence as a
time that influences firture deviant activity (see also Thomberry, 1987).

Using structural equation modeling to test the applicability of the social
development model, Huang and colleagues (2001 :98) found that “. . .youth who
manifested violent behavior at age 13 were likely to persist in violent behavior at age 18,
although antisocial socialization experiences continued to be an important mechanism for
a significant amount of that persistence.” Based on their findings, Huang et al.

(2001 : 100) concluded that early intervention is best way to interrupt “antisocial
socialization influences” from parents or peers that may be leading youth toward a
delinquent path. Moreover, since prior violent behavior is one of the best predictors of
firture violence and aggression, early interventions that effectively control the behavior of
children may result in long-term positive effects for years to come.

Using a nationally representative sample of youth (the National Youth Survey),
Elliott (1994) found that 45% of respondents who began offending prior to age 11 years
continued committing violent acts into their early 20’s. Similarly, Moffitt (1993:678)
pointed out that “About 6% of boys are first arrested by police as preteens. . .such early
arrests are important because it is the best predictor of long-term recidivistic offending.”
Despite these concerns regarding the future outcomes of early offenders, many
jurisdictions do not arrest or adjudicate such young children (Loeber & F arrington, 2001).

Many developmental researchers argued that antisocial behavior is readily
observable in very young children. For example, Moffitt (1993:675) maintained that
“. . .it becomes obvious that manifestations of antisocial behavior emerge very early in the

life course and remain present thereafter.” Similarly, Patterson (l982:6) suggested that

21

 

“. . .by the age of five most children in this society have learned most of the garden-
variety aggressive behaviors employed by children.” These behaviors may not be
criminal or delinquent as defined by the law, but they are behaviors that may forecast
future destructive behavior and therefore should result in some noninvasive preventative
intervention efforts.

To summarize, early onset offenders persist longer and often engage in more
serious offending for a number of reasons. First, early offending may be a sign that

biological or psychological traits exist that put an individual at a disadvantage in the same

 

environment where others may remain prosocial. Moreover, a disruptive family life (i.e.,
disruption, parental discord, etc.) further disadvantages these youth early in life. Second,
early offending can result in cumulative disadvantages; that is, decisions made and
actions taken early in life can affect opportunities later in life (Sampson & Laub, 1993).
For example, youth who are labeled deviant may be treated differently than others by
teachers, who may relegate them to nonacademic tracks. This decision makes it difficult
for the youth to get accepted into college. Finally, antisocial behaviors can result in
decreased bond to conventional institutions such as the family, school, and work which
can result in further deviance (Thomberry, 1987). For these reasons it is important to
address the deviant behavior of pre-adolescent youth. If not corrected, these youth are at-
risk to become serious and violent offenders.

Increasing evidence is being accumulated which outlines the causes and correlates
of serious and violent juvenile offending. In a review of the development of (and paths
leading toward) serious offending, Tolan and Gorman-Smith (1998:83) argued that

“. . .violent and serious offenders can be differentiated from other offenders based on age

22

of onset, presence of childhood behavior problems, and relative aggression level.”
Indeed, while many studies indicated that the peak period of juvenile offending generally
occurs between the ages of 14 and 17 years, serious offenders often report initiating
offending between 7 and 14 years old (Loeber, Farrington, & Waschbusch, 1998). More
specifically, in a comprehensive meta-analysis of predictors of serious and violent
juvenile offending, Lipsey and Derzon (1998) found that after prior antisocial behavior
between the ages of 6 and 11 years, male gender, low family socioeconomic status, and
minority race were among the strongest predictors of serious and violent offending at age
15 to 25 years. Between 12 and 14 years old, previous antisocial behavior continues to
be a good predictor of later serious/violent offending; additionally, associating with
deviant peers and weak social ties also are strong predictors (Lipsey & Derzon, 1998).
Interestingly, early substance use (6-11 years old) is a better predictor of future
serious/violent offending than later substance use (12-14 years old).

Hawkins, Laub, and Lauritsen (1998:109-42) identified a detailed list of malleable
risk factors for youth violence that could handily be addressed with an appropriate
intervention (see also Herrenkohl, Huang, et al., 2001:217). Notably, several such factors
are embedded within the familial environment, including parental criminality, child
maltreatment, poor parent-child relations and involvement, weak family bonding,
parental attitudes toward violence, and separation from parents and early home leaving.

The more risk factors to which a youth is exposed, the greater the risk of future
serious and/or violent offending. For example, aggressive children Who live in a
disorganized neighborhood with parents who are drug users are undoubtedly at a very

high risk for firture offending. Interventions ought to target specific risk factors in an

23

effort to decrease this likelihood (Wasserman & Miller, 1998). It is also evident that
different risk factors can operate differently at different developmental periods. For
example, family-related risk factors may be more salient for young offenders (Tolan et
al., 1995) while community and peer variables may become more important for older
adolescents (Herrenkohl, Hawkins, et al., 2001). Finally, Guerra (1998:394) reminds
researchers that “. . .it is important to distinguish risk factors useful for identification of
populations (e.g., urban, economically disadvantaged), or subgroups of individuals (e.g.,
males), from risk factors to be targeted by an intervention (e.g., farrrily management
practices)”.

Currently, three longitudinal studies that are underway in Denver, Pittsburgh, and
Rochester are designed to learn more about what causes (and is correlated with) serious
delinquency (Browning, Huizinga, Loeber, & Thomberry, 1999). Results of this research
indicate that family protective factors (adequate supervision, strong attachment,
consistent discipline) are among the most important in promoting resilience.

As seriousness of offending increases, juveniles become involved in a variety of
other antisocial activities (Loeber et al., 1998). Serious offenders use drugs and alcohol
more frequently than nonoffenders; indeed, many excessive substance abusers are
involved in serious crime (Huizinga & J akob-Chein, 1998). Moreover, compared to
nonoffenders, serious offenders are more aggressive, have more mental health and school
problems, and are victimized more often (Huizinga & Jakob-Chein, 1998).

Just because an adolescent has one (or a combination) of the above-described risk
factors does not necessarily predict that he or she will become a serious offender.

Similarly, just because an individual does not exhibit serious behavioral problems should

24

not mean that he or she is unable to receive services that address identifiable risks.
Teachers, parents, and other community members need to be aware of these risk factors
so that preventative interventions can be employed when possible, prior to initiation of
oflending. Once serious offending is exhibited, however, immediate intervention is
essential.

Serious and violent offenders are at a substantial risk for future offending.
Moreover, youth who initiate offending early are at risk for future delinquency and
criminality. As such, serious and violent offenders who begin between the ages of 10 and
13 years are among the most at-risk of all offenders (Loeber, 1982; Loeber & Farrington,
2001). Developing interventions that address their individual risk factors within multiple
domains must be a priority if significant long-term reductions in crime are going to be

realized.

Discussion

This preliminary chapter reviewed the extant literature concerning the main
focuses of the current study. The reviewed body of work frames the current study by
suggesting considerable utility in targeting the family relationships of serious, violent,
and early juvenile offenders. As discussed above, there are several correlates of serious,
violent, and early offending that originate from within the familial environment.
Additionally, intervention efforts aimed at improving family relationships have proven
successful among serious and violent offenders (Loeber & Farrington, 1998). These

conclusions point to the utility of developing more widespread intervention approaches

25

 

for serious and violent juvenile offenders that target familial relationships. The current
study seeks to test this assertion.

The following chapter will refrarne this discussion into contemporary theoretical
perspectives. Several prominent criminological theories identify the importance of
family relations and delinquency desistance. These ideas .will help to shape the way in
which interventions are developed and implemented. Indeed, the lack of evidence
available regarding effective approaches to combat serious and violent juvenile
delinquency is a social problem of significant import. This study seeks to fill this gap by
theoretically and empirically examining the merit of an intervention approach that

involves focusing on the family relationships of serious youthfirl offenders.

26

CHAPTER TWO: THEORETICAL FRAMEWORK

Introduction

Many criminological theories recognize the important socializing role of parents
within the familial domain. For example, social leaming theorists acknowledge the
importance of parents in conditioning, positive and negative reinforcement, modeling,
and imitation (Akers, 1985, 1998; Bandura, 1973; Burgess & Akers, 1966). Social
disorganization theorists also identify the significance of the informal social control
powers of families in neighborhoods (Bursik, 1988; Park & Burgess, 1924; Sampson,
1985; Sampson & Groves, 1989). In addition to this tradition of familial influence
incorporated into traditional criminological theory, the family has taken a more
prominent position in recent developmental theories of criminal behavior. For example,
familial dysfunction can be a primary mechanism by which latent traits become
manifested into deviant behavior. Moreover, inattentive parents, who do not take to time
to positively socialize their children, may actually be causing them to act out on impulses
or negative feelings, thereby leading them toward a ‘persistent’ criminal career.

Despite the role parents and families play in the above-referenced theories, as a
group, control theories most clearly represent the parent-child relationship. Control
theories emphasize “the prevention of crime through consequences painful to the
individ ” (Gottfredson & Hirschi, 1990:85). Historically, classical theorists focused on
the legal sanctions that result from breaches of governmental law (e.g., imprisonment).
More recently, however, researchers have examined the effects of informal social control

agents such as parents, neighbors, and others not affiliated with the justice system. The

27

 

following sections describe this evolution, focusing on how several social control theories
seek to explain crime and delinquency in terms of characteristics within the familial
environment.

Several family attributes have been identified in theoretical models employed to
explain deviance. Hirschi’s (1969) social bond argues that youth who are strongly
attached to their parents are less likely to engage in delinquent behavior. Patterson’s
(1980; 1982) coercion theory stresses the importance of direct supervision and discipline
by parents in controlling the unwanted behaviors of their children. Catalano & Hawkins’
social development model (1996) and Gottfredson and Hirschi’s General Theory of
Crime (1990) discuss the role of parents in socializing youth such that prosocial behavior
is reinforced and proscribed behavior is disciplined. Finally, Sampson & Laub’s (1993)
age-graded theory of informal social control integrates elements of each of these
perspectives. The following sections will review the theoretical-underpinnings of, and

empirical support for, each of these perspectives in terms of their focus on the family.

Social Bond

Social control theories focus on the ability of institutions to control thebehavior
of members of a particular society. Police act as a formal social control agent operating
on behalf of the greater community. Schools and churches also act to influence the
behavior of their members. Similarly, families are recognized as a primary mechanism of
informal social control. Early control theorists such as Reiss (1951) and Nye (1958)
stressed the importance of the family. Moreover, many positivist researchers have also

identified the role of the family in controlling latent traits and ultimately deviance

28

 

(Glueck & Glueck, 1950). In 1969, Travis Hirschi explicated a version of social control
theory that has been one of the most popular perspectives in research involving
delinquency and the family (Rankin & Kern, 1994). This section will review the
arguments of, and evidence for, Hirschi’s theory.

As a control theorist, Hirschi (1969) assumed that individuals are inherently
pleasure-seeking and therefore need to be ‘controlled.’ As such, it is conformity rather
than deviance that needs to be explained in criminological theorizing and empirical
examination. Hirschi (1969) argued that conformity results when an individual is
strongly bonded to conventional institutions such as the family, school, community, and
church. His primary focus, however, concerned relations within the family. The social
bond is comprised of four primary elements: attachment to significant others (notably
parents), commitment to future goals (i.e. through education), involvement in
conventional activities (i.e. extracurricular activities at school), and belief in the moral
values of society (the internalization of conventional values).

In general, social control theories argue that parents buffer their children from
delinquency by controlling their behavior via strong social and emotional ties that bind
youth to their parents, and, by extension, conventional order. Hirschi (1969:86) -
described the relationship between parents and child in terms of their bond
unequivocally:

As is well known, the emotional bond between the parent and the child

presumably provides the bridge across which pass parental ideals and

expectations. If the child is alienated fiom the parent, he will not learn or will

have no feeling for moral rules, be will not develop an adequate conscience or
superego.

29

Parents act not only as moral models, but coercive controllers if exhibited
behavior does not follow the mores, norms, and values of the family, or society at large.
In short, parents are responsible for inculcating their children with positive morals and
strengthening these morals through positive reinforcement and caring correction such that
internal controls are developed which govern their behavior when parents are absent. “If
the bond to the parent is weakened, the probability of delinquent behavior increases; if
this bond is strengthened, the probability of delinquent behavior declines” (Hirschi,
1969:88). By and large, support exists for the hypothesis that increased parental
attachment results in decreased delinquency (Cantor, 1982; Wiatrowski et al., 1982;
Wells & Rankin, 1988).

There are several examples which help to illustrate the logic that an adolescent
who is strongly bonded to his or her parents will engage in fewer delinquent acts than a
weakly bonded individual. First, a child who is well bonded to a parent will likely spend
more time with that parent which may result in closer direct supervision by that parent
and, as a result, fewer opportunities for deviant behavior. Because parents cannot
supervise their children all of the time, the emotional bond between parent and child can
act to control the child’s actions even when the parent is not present, also known as -
“virtual supervision” (Cemkovich & Giordano, 1987; Hirschi, 1969). Children who
perceive that their parents do not know where they are or what they are doing are more
likely to report involvement in deviance (Hirschi, 1969:89). J unger & Marshall (1997),
for example, found that low virtual supervision was significantly related to self-reported

delinquency in a sample of ethnically diverse lower-class youth.

30

Because virtual supervision is concerned with indirect control, Hirschi argued that
its effectiveness is more a function of intimacy of communication than supervision. That
is, youth who share their experiences with their parents on a regular basis are more likely
to consider the response of their parents before engaging in proscribed activities.
Combined with other measures of communication (i.e., “Do you share your thoughts and
feelings with your mother/father”), virtual supervision was again highly predictive of
self-reported delinquency (youth with more intimacy of communication were less likely
to report involvement in delinquency; Hirschi, 1969).

Strong ties to parents inhibit delinquency because children fear unfavorable
responses from parents if antisocial behaviors ever become known. More specifically,
youth who are psychologically attached will fear the emotional damage caused by the
disobedience (as opposed to actual physical punishment). These ideas have also been
termed “stakes in conformity” (Toby, 1957); attached individuals have a “stake” in
conforming to societal expectations - namely, parental approval.

While Hirschi (1969) devoted most of his attention to discussing the effects of a
strong bond to conventional parents (that is, prosocial and law-abiding), he
acknowledged the difficulty in this theory when children are born to deviant parents. He
concludes, nonetheless, that youth who are firmly attached, even to deviant parents, will
be less likely to engage in deviance than those weakly attached to the same type of
parents. Hirschi suggests that even if parents are involved in criminal activities or hold
favorable definitions to deviance, they would do their best to avoid passing those ideals
on to their children. In the eyes of the child, then, the parent acts as a proxy for a

conventional law-abiding citizen.

31

While social learning theory suggests that youth who associate with deviant
others are more likely to engage in deviance than those who do not (Akers, 1985, 1998;
also referred to as cultural deviance theories), Hirschi (1969) contended that delinquent
peers will become salient only if the attachment to parents is weak. Hirschi’s (1969: 100)
data supports this claim; regardless of the number of delinquent friends with which a
youth associates, the more attached he is to his father, the less delinquent he will be.

Using data from the Richmond Youth Survey—a random sample of
approximately 4,000 youth from the San F rancisco-Oakland, California metropolitan
area—Hirschi presented initial empirical support for many components of his theory.
Others, too, have provided empirical support for Hirschi’s social control theory
(Cemkovich, 1978; Cemkovich & Giordano, 1987, 1992; Fagan & Wexler, 1987; Sokol-
Katz, Dunharn, & Zimmerman, 1997; Wiatrowski, Griswold, & Roberts, 1981). For
example, Alarid, Burton J r., and Cullen (2000) tested the efficacy of social control theory
on a sample of approximately 1,100 incarcerated felons and concluded that Hirschi’s
soCial bonding theory is a “general” theory in the sense that it explained variation in a
variety of deviance (general crime scale, drug, property, and violent subscales) and had
similar effects among males and females.

Similarly, Erickson, Crosnoe, and Dombusch (2000) empirically tested the
hypothesis that strong social bonds to parents would reduce the likelihood that youth
would associate with deviant peers. They found that youth who reported strong ties to
parents were significantly less likely to have close friends who participated in delinquent
activities often (Erickson et. a1, 2000). This finding is especially useful because the

researchers actually asked the friends about their involvement in deviance instead of

32

 

relying on the primary individual as a proxy, thus avoiding the problem of the youth
reporting his or her own behavior as that of his or her peers. Parental attachment
(measured as a scale representing questions such as: “I try to have the same opinion as
my parents” and “My parents spend time just talking with me”) and supervision (a scale
including: “How much do your parents really lqrow your fiiends?” and “How much do
your parents know about what you do with your free time?”) and other social control
variables accounted for 18% of the variation in delinquent peer associations (Erickson et.
a1, 2000). Erickson and colleagues (2000:420) concluded that “Adolescents who
experience strong social attachments, commitments, and involvements are less likely to
jeOpardize personal relationships and accomplishments by associating with peers who
support and engage in deviance.” In short, strong attachment to parents can insulate
adolescents from other criminogenic factors and may therefore be considered in any
comprehensive delinquency prevention program.

Dombusch, Erickson, Laird, and Wong (2001) empirically tested the power of
familial attachment to reduce the frequency, prevalence, and intensity of cigarette
smoking, alcohol use, marijuana use, delinquency, and violent behavior among a
nationally representative sample of approximately 13,000 7th through 12th grade students.
They found that familial attachment acted as a protective factor in that youth who were
more strongly attached were significantly less likely to report all forms of deviance
(Dombusch et al., 2001). Moreover, these relationships held across diverse communities
(different levels of economic deprivation), and ethnicities, as well as across gender. A
unique attribute of this study was the utilization of parental reports of attachment,

supplementing that of the youthful respondent. Indeed, the parental report of attachment

33

 

was “the measure most often significantly associated with reductions in deviance...”
(Dombusch et al., 2001: 417).

Interestingly, Anderson, Holmes, and Ostresh (1999) found gender differences in
the relationship between attachment and delinquency (see also Cemkovich & Giordano,
1987). In a sample of incarcerated delinquents in Wyoming, they found that attachment
to parents reduced the severity of delinquent behavior among boys while attachment to
school and peers had the same effect for girls (Anderson et. al, 1999). That is, girls
appeared in this study to be more effected by school and peers than parents. This
research replicated other studies (for example, Cantor, 1982) which found that even
though girls tended to be more strongly attached to their parents in general, intensity of
attachment had a greater delinquency-reducing effect among boys.

May, Vartanian, and Virgo (2002) investigated the effect of parental attachment
and supervision on fear of crime among incarcerated adolescents and found mixed
results. First, parental supervision was significantly related to fear of victimization
(highly supervised youth were more likely to fear victimization), yet respondents were
less likely to acknowledge that they could be at risk to be a victim of a crime (May et al.,
2002). Second, parental attachment had no significant effect on an adolescent’s fear of
crime. May and colleagues (2002) conclude, therefore, that youth who are highly
supervised may be “shelter ” from crime yet not empowered with strategies to deal with
potential situations that might arise.

Others, however, have questioned Hirschi’s assertions (Matsueda, 1982). For
example, Greenberg (1999) reanalyzed the Richmond Youth Survey and found only weak

support for social control theory. Another problem with Hirschi’s analysis, Greenberg

34

(1999) argues, is that Hirschi was assessing the relationship between the level of social
bond at the time of the interview and previous delinquency. Using this interpretation,
delinquent behavior may actually predict the level of social bond because the temporal
ordering is not correctly specified (Thomberry, 1987).

As evidenced by this review, much research has been conducted concerning
Hirschi’s (1969) social control theory and the relationship between family relationships
and adolescent deviance. Despite this wealth of knowledge, however, few concrete
conclusions have emerged. A few generalizations can be offered, based on the
preponderance of the evidence. First, Hirschi’s social bonding theory is an important
contribution to the understanding of deviant behavior. Weaknesses that have been cited
are reconcilable and researchers continue to test and refine the perspective so as to
produce a more comprehensive approach. Second, the family is one of the most
important institutions in an adolescent’s life and therefore must be considered in any
model attempting to explain antisocial behavior. Hirschi provides a starting point from
which other researchers can continue exploring the complex relationships that occur

within the family and how they affect delinquent involvement.

Social Development Model

The social development model integrates multiple theories (control, learning, and
differential association) and stresses the importance of risk and protective factors in
multiple social domains (Catalano & Hawkins, 1996). Familial risk factors included:
“. . .family history of drug abuse or crime, poor family management practices, family

conflict, low family bonding, and parental permissiveness. . .” (Catalano & Hawkins,

35

1996: 152). F arnilial protective factors include: “. . .farnily cohesion and warmth or

 

bonding during childhood...” (Catalano & Hawkins, 1996:153).

Drawing from differential association theory, the social development model
acknowledges that an individual’s behavior is largely determined by the behavior of those
with whom he or she associates. As argued in control theory, individuals who have
strong social bonds to prosocial others are inhibited from engaging in deviant behavior
for fear of damaging that relationship. The social development model views the bond to

the family as critical in the development of bonds to other social institutions (Hawkins,

 

Catalano, Jones, & Fine, 1987). Finally, like learning theory, the social development
model argues that behaviors are learned from others within intimate settings and involve
both imitation and reinforcement. In short, youth are socialized within primary groups
(e.g., family, peers, school) to engage in either prosocial or antisocial ways.

The social development model seeks to explain a wide variety of antisocial
behaviors across the life course, but focuses on the influence of risk and protective
factors during four developmental periods: preschool, elementary school, middle school,
and high school. Transitions between each of these periods are also important stages of
development. The model “hypothesizes that children learn patterns of behavior, whether
prosocial or antisocial, from the socializing agents of family, school, religious and other
community institutions and their peers” (Huang et al., 2001 :77). Moreover, it specifies
multiple pathways to delinquency within multiple developmental periods.

Using structural equation modeling, Huang and colleagues (2001) found that the
social development model was an acceptable fit to the Social Development Project

sample. They found that youth who developed rewarding relationships with antisocial

36

others in rrriddle school were more likely to engage in substance use and violent behavior
later. Interestingly, they also found that antisocial socialization (opportunities,
involvement, and rewards) was more salient than simple bonds (attachment and belief) to
antisocial others. This finding is consistent with Hirschi’s (1969) argument that an
individual who has a strong bond to antisocial others will not necessarily be negatively
influenced by their behavior.

In another examination of the social development model, Herrenkohl, Huang,
Kosterman, Hawkins, Catalano, and Smith (2001) found that the socialization processes
leading to violence at age 18 were the same for youth who initiated offending in
childhood (10-1 1) or early adolescence (12-16). Herrenkohl and colleagues (2001)
suggest that this finding calls into question the assertions of many who argue that
childhood offenders are categorically different from adolescent initiators (Moffitt, 1993).
This conclusion is itself questionable, considering their operationalization of childhood
offending. It could be argued that a 12 or even 13 year old offender is still a childhood
offender (or early starter). Nevertheless, Herrenkohl and associates (2001) find support
for the social development model’s hypothesis which argues that socialization within
antisocial circles will lead to antisocial behaviors, especially when that behavior is
reinforced. These researchers conclude, then, that parenting skills training coupled with
adolescent social skills training appears to be a theoretically promising approach to alter
the socialization mechanisms and the child behaviors (Herrenkohl et al., 2001).
Specifically, parent training should focus on “age-appropriate discipline strategies and

monitoring” (Herrenkohl et al., 2001 :60).

37

 

Ayers and colleagues (1999) found that several social development model
constructs were significantly related to desistance (e.g., higher social skills, higher
grades, more conventional peer associations, more commitment to school, more belief in
moral order). Important for the purposes of the current study, these researchers also
found that females were more bonded to their families at age 12-13 and that this bond
was related to desistance for females but not males. They conclude that interventions
aiming to improve parent-child interactions may be more successful among females at
this age and that such approaches may need to begin earlier to be successful with male
offenders (Ayers et al., 1999).

One limitation to the above-referenced tests of the social development model is
that they were all conducted using the same data (data from the Seattle Social
Development Project). To ensure that the hypothesized relationships occur in a variety of
settings, researchers must begin to test the social development model on other samples.

The social development model is an ideal theoretical framework fiom which to
develop intervention approaches because it identifies specific risk and protective factors
that can be targeted. The social development model acknowledges the inseparability of
multiple domains, and the need to address juvenile delinquency from within multiple
spheres. Additionally, this perspective also proposes that interventions should increase
opportunities for prosocial involvement, enhance the child’s social skills, and ensure
consistent and reasonable rewards and punishments for prosocial and proscribed
behaviors (Hawkins et al., 1987).

The social development model integrates much of the previous theoretical work

that focuses on the process of socialization, particularly the influence of the family in this

38

 

 

process. The model differs from traditional social control theory by arguing that youth
who are strongly bonded to antisocial others are at risk for engaging in delinquent
behavior. Moreover, it includes elements of social learning and differential association
theories which identify the power of others to influence one’s behavior. Finally, similar
to other control theories, the social development model assumes that individuals are
rational actors who are willing to engage in a variety of behaviors (prosocial and
antisocial) to the extent that their own pleasure is maximized and pain minimized.

For the purposes of the current study, it is clear that the family remains firmly as a

 

central tenet in the socialization process. Improper socialization by the family places the
adolescent at an elevated risk for engaging in antisocial activities. Moreover, parents
who are involved in criminal activities may be reinforcing the acceptability of these
activities to their children. In short, if the process by which an early adolescent
delinquent becomes an adult criminal is to be effectively interrupted, program
administrators must remain cognizant of the risk and protective factors associated with

the family environment.

Coercion Theory

Like other variations of control theory, coercion theory is concerned with the
mechanisms by which adolescents refrain from engaging in delinquency (Patterson, 1980,
1982). Coercion theory maintains that parents are responsible for supervising and
otherwise controlling the behavior of their children. Patterson (1980), however, argued
that coercion theory is a reformulation of social learning theory in that children view

deviance as normative because parents either tolerate or are otherwise incapable of

39

addressing the child’s deviant behavior. As such, unskilled parenting can result in
increased risk for children becoming involved in continued delinquency behavior.
Patterson (1980:81) presented an explicit set of skills parents can employ to encourage
prosocial development:

(a) notice what the child is doing; (b) monitor it over long periods; (c) model

social skill behavior; (d) clearly state house rules; (e) consistently provide sane

punishments for transgressions; (f) provide reinforcement for conformity; and (g)

negotiate disagreements so that conflicts and crises do not escalate.

Patterson’s model differs from Hirschi’s (1969) conception of social control in
that Patterson is concerned largely with the direct inhibiting properties of parental
supervision whereas Hirschi proposed that the affective relationship between parent and
child (social bond) indirectly controlled behavior (Sampson & Laub, 1993). That is,
according to Patterson (1980), a child will not engage in delinquent activities because his
or her parent is watching or will find out about such behaviors, resulting in some form of
negative stimuli (punishment). According to Hirschi (1969), children will refrain from
engaging in anti-social activities for fear of damaging their relationship with prosocial
institutions such as the family if caught (see also Thomberry, 1987). Sampson and Laub
(1993) point out, however, that these two processes are likely correlated; that is, parents
who effectively supervise their children often also have well-developed affective ties
with them.

Patterson (1980283) further maintained that “. . .parents reinforce and punish in a
manner isomorphic with their own established hierarchy of social behavior and values.
That is, they support what they themselves actually do.” Additionally, Patterson noted

that while parents themselves may not be criminals or engage in many criminal activities,

they do, however, often deliberately ignore antisocial behavior when it occurs in their

40

children. Parents either do not believe others when they tell them that their children
behaved in certain ways, or are unprepared to prosocially discipline their child
effectively.

Coercion theory argues that the two most important aspects of effective parenting
are monitoring and discipline. According to coercion theory, children do not grow out of
antisocial behavior; they will continue engaging in proscribed activities unless adequately
punished. “Antisocial acts that are not punished tend to persist. Parents of stealers do not
track; they do not punish; and they do not care” (Patterson, 1980:89). Parents who are
incapable or unwilling to nonviolently discipline their children for engaging in deviant
activities are implicitly endorsing the behavior. What is needed, then, is comprehensive
training to teach parents how to authoritatively manage antisocial behavior by their
children. Patterson (1980:77) argues that “. . .what is necessary is that the family directly
punish, nonphysically, the deviant acts and concentrate on teaching prosocial competing
responses.” Patterson (1980) found that parents who participated in skills training were
able to significantly decrease the frequency of stealing among their children.

Coercion theory is “. . .a set of statements about pain-control techniques employed
by one or both members of a dya ” (Patterson, 1982:6). Any member of a group can
apply coercion strategies. For example, children behave in certain ways to achieve an
intended reSult (e.g., toddlers cry to attain the attention of a parent). Also, parents are
impelled to respond to the behavior of their children (e.g., punish or ignore proscribed
behaviors). Because coercion theory predicts a reciprocal relationship between members
of the dyad, it does not necessarily provide insight as to causality. That is, did ineffective

parenting cause the child to tum toward deviance, or did the deviant child cause the

41

 

 

parent to question his or her ability to parent effectively? Patterson suggests that this
dialectic relationship likely occurs as a result of both processes (see also Thomberry,
1987). “In families of antisocial children some parental ‘neuroses’ can be thought of as
the outcome or a process rather than its cause” (Patterson, 1982211).

Coercion theory assumes that aggression in families is not merely the aggregation
of several “fortuitous events” (Patterson, 1982:12). Moreover, while many researchers
focus only on extreme forms of aggression, Patterson (1982) suggests that minor,
relatively harmless acts (such as teasing and disciplining) can escalate into more serious
forms of aggression (hitting).

Patterson proffers a coercion theory that attempts to elucidate the didactic
relationship between parent and child. In short, like other control theories, the theory
predicts that ineffective child rearing (specifically, poor monitoring and/or discipline)
will result in children at risk to engage in delinquent behavior. Youth who perceive that
they are not being treated fairly will foreclose on childhood by engaging in “adult-like”
deviant behaviors (e.g., smoking, drinking, stealing) to demonstrate to others that they are
“grown-up” (see also Moffitt, 1993). While the theoretical analysis has merit, it has not
been empirically scrutinized. Patterson’s insistence upon the significance of parenting
skills in delinquency desistance, however, is important and therefore relevant to the

present study.
General Theory of Crime

Gottfredson and Hirschi (1990) detailed a general theory of crime that holds low

self-control as the primary variable which predicts crime, delinquency, and other

42

analogous undesirable outcomes. They also assert that the propensity to engage in
deviant behavior remains stable over time; as a result, differences in actual involvement
in crime can be attributed to population heterogeneity and variation in opportunity to
engage in illicit activities. In short, Gottfredson and Hirschi (1990:89) “. . .suggest that
high self-control effectively reduces the possibility of crime—that is, those possessing it
will be substantially less likely at all periods of life to engage in criminal acts.” In other
words, individuals with low self-control will always be at risk for engaging in some
behavior that may be harmful to others (whether legally proscribed or not). Gottfredson
and Hirschi (1990:90) argue that “. . .people who lack self-control will tend to be
impulsive, insensitive, physical (as opposed to mental), risk-taking, short-sighted, and
nonverbal, and they will tend therefore to engage in criminal and analogous acts.”

While many criminologists have criticized Gottfredson and Hirschi’s hypothesis
(Akers, 1991; Geis, 2000), their analysis can be informative as to how best to address
deviant behavior. Indeed, despite the criticisms, impressive empirical support has already
emerged for their theory (Burton, Cullen, Evans, Alarid, & Dunaway, 1998; Grasmick,
Tittle, Bursik, & Ameklev, 1993; J unger & Tremblay, 1999; LaGrange & Silverrnan,
1999; Paternoster & Brame, 1998; Pratt & Cullen, 2000). Their ideas regarding the
relationship between family socialization and involvement in delinquency are more
instructive, but less often empirically tested (Hay, 2001; Lynskey, Winfree, Jr., Esbensen,
& Clason, 2000). Gottfredson and Hirschi (1990) suggested that the relationship between
parenting and delinquency is mediated by self-control. “In our view, the origins of

criminality of low self-control are to be found in the first six or eight years of life, during

43

which time the child remains under the control and supervision of the family or a familial
institution” (Gottfredson & Hirschi, 1990:272).

Gottfredson and Hirschi’s (1990) general theory of crime acknowledges the
primary role parents play in socializing their children to develop strong self-control.

“. . .self-control differences seem primarily attributable to family socialization practices”
(Gottfredson & Hirschi, 1990: 107). Indeed, others have found that parental monitoring
and disciplining techniques significantly predict levels of self-control which in turn is
related to delinquency (Hay, 2001). The mediating role of self-control in Hay’s (2001)
study, however, was not as clear. Parents play a crucial role in that self-control is
developed (almost exclusively in the family) in the first few years and remains relatively
stable across the life course (Gottfredson & Hirschi, 1990; Nagin & Paternoster, 1991).
If parents fail to adequately socialize their children such that self-control is not fully
developed, the youth may be at a greater risk for delinquency.

Gottfredson and Hirschi (1990:97) maintain that “the major ‘cause’ of low self-
control thus appears to be ineffective child-rearing.” Further, similar to coercion theory,
they argue that three conditions represent proper child-rearing: “someone must (1)
monitor the child’s behavior; (2) recognize deviant behavior when it occurs; and (3)
punish such behavior” (p. 97). Gottfredson and Hirschi point out that these conditions
are rarely intentionally violated by parents, but that parents often implicitly fail to
adequately socialize their children in one of four ways, which can result in
underdeveloped self-control. First, some parents, whether deliberately of not, do not care
for their children. Second, even caring parents may fail to adequately supervise their

children. As a result, deviant behavior goes unnoticed and unpunished. Third, caring and

attentive parents may not see anything wrong with the types of behaviors in which the
child is engaging. Finally, even if parents are caring, attentive, and cognizant of the
prosocial and antisocial activities in which their child is involved, they may be unable or
unwilling to punish their children for the deviant behaviors. Any of these conditions may
result in the failure of children to develop strong self-control.

Generally, parents cannot deliberately train their children to have low self-control,
thereby increasing their delinquency. It is the lack of involvement such as that described
above that can result in underdeveloped self-control. Gottfredson and Hirschi (1990:94-
5) argue this point: “One thing is, however, clear: low self-control is not produced by
training, tutelage, or socialization. As a matter of fact, all of the characteristics associated
with low self—control tend to show themselves in the absence of nurturance, discipline, or
training.” Neglect in parental responsibilities may result in a child who has not firmly
developed self-control and as a result is at a risk for engaging in deviant activities.
Gottfredson and Hirschi’s (1990:101) general theory “. . .assumes criminality is not
something the parents have to work to produce; on the contrary, it assumes that
criminality is something they have to work to avoid.”

The idea that self-control is an important mechanism by which criminal and
delinquent tendencies are restrained is not new. Approximately fifty years ago, Reiss
(1951) and Nye (1958) also recognized the power of the individual to control his or her
own behavior. Reiss’s concept of personal controls and Nye’s conception of internalized
controls represent the same force that Gottfredson and Hirschi are extolling in their
general theory. Moreover, even Reiss (1951 :198) recognized the role of families and

other “primary groups” in the development of “personal controls” among adolescents:

45

Primary groups are the basic institutions for the development of personal controls
and the exercise of social control over the child. . ..Delinquency and delinquent
recidivism may be viewed as a consequence of the failure of primary groups to
provide the child with appropriate non-delinquent social roles and to exercise
social control over the child so these roles are accepted or submitted to in accord
with needs.

In contrast to Hirschi’s (1969) social bond theory which argues that a strong bond
to any institution should reduce the likelihood of offending, Reiss (1951) suggests that
parents must model prosocial attitudes and behaviors for the child to internalize and
imitate. This earlier conceptualization of control theory is also evident in the social
development model discussed above (Catalano & Hawkins, 1996).

Even though Gottfredson and Hirschi’s (1990) general theory of crime ostensibly
indicates that the development of self-control is the only way to reduce criminality, they
firmly reject the theory’s utility in intervention development: “. . .all indications are that
such teaching is highly unlikely to be effective unless it comes very early in
development” (1990:269). These theorists maintain that socialization must occur prior to
that age at which a child is able to manifest low self-control in delinquent ways (e. g.,
prior to age 8 years). Accordingly, Gottfiedson and Hirschi promote early prevention
efforts that train children to delay gratification. But what is to be done about older
children and adolescents who were unable to grow up in a family with parents capable of
providing these lessons? The implications of their perspective are less clear for older
populations of offenders.

While Gottfredson and Hirschi (1990) did not suggest that their theory can aid in
the development of effective interventions aimed at an offending population, it is possible

that their analysis can inform intervention efforts. If they are correct in asserting that

self-control is the trait responsible for criminality (and many tests of the general theory

46

have provided some supportive evidence), attempts to develop self-control in an
intervention context may be worth pursuing. Also following their analysis, the most
effective way to develop self-control is from within the family context. As such,
interventions ought to focus on the family; more specifically, interventions may be more
effective if they train parents in child-rearing strategies, positive reinforcement, as well as

efficient supervisions and monitoring.

Informal Social Control

Control theorists have long recognized that disapproval by loved-ones is as
powerful a sanction as any formal legal threat (Gottfredson & Hirschi, 1990). As such,
many have explored the relationship between informal social control agents such as the
family and peer group in the management of crime and deviance. One such theoretical
perspective is Sampson and Laub’s (1993) age-graded theory of social control. Sampson
and Laub’s (1993) approach began from a traditional social control perspective in that a -
weakened bond to society can result in deviant behavior (Durkheim, 1951; Hirschi, 1969;
Kornhauser, 1978). Sampson and Laub (1993268) “. . .combined the central ideas of
social control and coercion theory along with the notion of reintegrative shaming into a
unified model of informal family social control that focuses on three dimensions—
discipline, supervision, and attachment.” These scholars began with the view that social
control is “. . .the capacity of a social group to regulate itself according to desired
principles and values, and hence to make norms and rules effective” (Sampson & Laub,

1993:18).

47

 

According to Sampson & Laub (1993), there are several ways in which the family
promotes or reduces delinquency. If turning points do not occur at developmentally
normal stages (e.g., graduation in late teens, followed by college, military or job
experiences in the early 20’s, and marriage and children shortly thereafter), youth (or
adults) could be ‘turned’ toward delinquent paths. For example, adolescents who are
kicked out of the home for inappropriate behavior may sever familial ties too early. This
may result in the over-reliance on peers for emotional and financial support. If prosocial
friends are unavailable to the youth, deviant peers will supplant the family as the
important reference group.

Sampson & Laub (1993, 1995) acknowledged the effects of the family across the
life course. Their theory of informal social control is probably the best example of a
perspective that considers the role of the family over the entire life course. For example,
the family is an important buffer from crime both as a child and adult; parents supervise
their children while spouses and children are incentives for adults to remain crime-free.
Additionally, Sampson and Laub (1993) provided an age-graded theory that accounts for
stability and change throughout the life course; and the family is the primary mechanism
of informal social control at several stages. As an age-graded theory, Sampson and Laub
recognized the different ways in which the family can be involved in delinquency
prevention at several stages of development. First, as youth deveIOp, parents socialize
their children to ingrain in them whatever norms, values, and beliefs the parent’s hold.
Parents reinforce positive behavior while condemning negative behaviors. Moreover,
parents supervise their children so that they are able teach life lessons at appropriate

times. During adolescence, however, parental influence begins to drift. Peers become

48

more important during this stage (Elliott, Huizinga, & Ageton, 1985; Warr, 2002), and
parents need to rely on the values they have instilled to that point. Sampson and Laub
(1993:97) also maintain, however, that “. . .delinquency in adolescence is explained
largely by exogenous family factors that also occur in adolescence,” such as supervision
and effective socialization.

In the early 20’s, the family once again returns as the prominent protective factor
in keeping individuals from engaging in criminal activities. As individuals marry and
have children of their own, they develop stakes in conformity (Toby 1957) or social
capital (Coleman, 1988) that will continue to keep them from engaging in deviant
activities. In brief, Sampson and Laub (19932141) argue, “. . .social ties are important
insofar as they create interdependent systems of obligation and restraint that impose
significant costs for translating criminal propensities into action.” Similarly, Farrington
(19862189) suggests that “the age crime curve probably reflects decreasing parental
controls, a peaking of peer influence in the teenage years, and then increasing family and v
community controls with age.”

Sampson and Laub (199327) argued three main points regarding the relationship
between family and delinquency/crime across the life course:

(1) structural context mediated by informal family and school social controls

explains delinquency in childhood and adolescence; (2) in turn, there is continuity

in antisocial behavior from childhood through adulthood in a variety of life
domains; and (3) informal social bonds in adulthood to family and employment
explain changes in criminality over the life span despite early childhood
propensrtres.

Sampson & Laub (1993) also considered the family in the continuity and change

of criminality. They recognize specific ‘turning points’ and how they may result in an

individual changing his or her trajectory over time. For example, removal of one parent

49

by death or divorce may negatively affect a youth. He or she may attempt to escape
emotional pain by turning to deviant peers or illicit substances. Conversely, if the parent
who was removed was abusing or in some other way disrupting the familial environment,
strain may be lifted and the youth may return to a prosocial trajectory.

Transitions that occur (marriage, first job, an arrest) can positively or negatively
deflect the trajectory on which an individual appears to be headed. There is a
developmentally appropriate progression of transitions that most individuals experience
which often results in normative development. Summarizing the life-course perspective,
Sampson and Laub (199328) note that “. . .life course analyses are often characterized by a
focus on the duration, timing, and ordering of major life events and their consequences
for later social development.” As discussed above, individuals transition into high-
school, then graduate from high school, enter the workforce, get married, then have
children. Individuals who experience these events out of developmental order (i.e.,
having children before graduating high school) are at increased risk for engaging in
criminal behaviors. To be sure, some delinquency, if minor and engaged in during
adolescence (i.e., truancy, fighting, minor theft), is considered developmentally
appropriate and common (Moffitt, 1993). Early or prolonged involvement in these
behaviors, however, is a potent risk factor for future delinquency (see discussion of early
offenders in chapter one).

Sampson and Laub’s approach is unique in that it is a deve10pmental perspective
that extends beyond childhood and adolescence and considers the importance of informal
social control mechanisms into adulthood. More specifically, these theorists maintained

that different formal and informal social control institutions can influence individuals

50

differently at different ages. Additionally, Sampson and Laub (1993219) argued that
“. . .informal social controls derived fi'om the family (for example, consistent use of
discipline, monitoring, and attachment) and school (for instance, attachment to school)
mediate the effects of both individual and structural background variables.”
Sampson and Laub (1993:246) further contend that structural context
(disorganization, disadvantage, etc) is mediated by informal family social controls:
It is our view that family and school processes of informal social control provide
the key causal explanation of delinquency in childhood and adolescence.
Structural background characteristics are important in terms of their effects on
informal family and school processes, but these same characteristics have little
direct influence on delinquency.

That is, youth residing in a poor neighborhood, who regularly witness crime and deviance
may be buffered from the deleterious effects of that environment if effectively controlled
by their parents (Patchin, Bynum, Huebner, McCluskey, & Varano, 2003). Moreover,
underlying traits that may predispose an individual to deviance (i.e., low self-control) can
be adequately handled through a strong bond to the family. As such, living in a
disadvantaged neighborhood may be a risk factor only for those youth who also do not

have competent parents (in contrast see Brooks-Gunn, Duncan, Klebanov, & Sealand).

Theory to Practice: Family-Focused Interventions

Based on the theoretical perspectives reviewed above, it can de deduced that
family relationships are a malleable factor that can be adequately addressed in an
intervention context. Moreover, if a child does not have a strong, positive relationship
with his or her parent(s), other intervention efforts may prove ineffective. The

intervention may be a turning point for the family (both for parents and children), as

51

children learn that there are consequences for their behavior and parents learn that their
children are “out of control.” As such, simply being ordered to participate in an
intervention may improve family relations and therefore reduce offending. Alternatively,
however, interventions may need to specifically target familial processes early in order to
prime the family for other programming (i.e., drug treatment, life skills).

Reviews have suggested that family interventions are more effective than non-
family interventions (Latirner, 2001; Perkins-Dock, 2001). Moreover, Kazdin (1987)
argued that family interventions hold the most promise for the treatnrent of Conduct
Disorder in children and adolescents. Finally, parent-focused interventions implemented
at home or in other community locations also tend to be very successful (Burns et al.,
2003).

Multisystemic therapy (MST) has proven especially effective among samples of
seriously delinquent youth. MST moves beyond traditional family therapy in that it
considers the ecological nature of adolescent development (Bronfrenbrenner, 1979; see
. also Tolan, Guerra, & Kendall, 1995). Specifically, it is argued that dysfunctions in any
system, family, peer group, community, or school, can result in antisocial behavior by the
individual. “MST interventions are child-focused, family-centered, and directed toward
solving multiple problems across the numerous contexts in which the youth is embedded:
family, peers, school, and neighborhood” (Tate, Reppucci, & Mulvey, 1995:779).
Similarly, changes in one domain may or may not result in system-wide changes.
Mihalic and colleagues (2001210) observe that “the overriding purpose of MST is to help

parents deal effectively with their youth’s behavioral problems; help youth cope with

52

family, peer, school, and neighborhood problems; and reduce or eliminate the need for
out-of-home placements.”

A notable attribute of MST and other effective family-centered interventions is
their focus on multiple domains. Indeed, the wealth of knowledge suggests that
individually focused programs, by themselves, do little to effectively prevent future
crime. As discussed at length, juvenile delinquency is a multiply-deterrnined phenomena
resulting from the convergence of risk factors in the absence of effective protective
factors. As such, several agencies (e.g., police, mental health, school, court) must
coordinate efforts in order to provide a “wraparoun ” model whereby all individual and
familial needs are addressed. The challenge, then, is to develop a program that is
comprehensive enough to address a myriad of risk and protective factors, but that is
easily implemented in diverse settings (ecologically valid). In short, serious juvenile
offenders require theoretically grounded comprehensive multirnodal interventions that
address specific risk and protective factors within a flexible framework (Black, Howard,

Kim, & Ricardo, 1998).

Discussion

Social control theories are among the most empirically supported theories in
criminology (Akers, 2000). Even self-control (fostered by parents) has been identified
and empirically supported as a leading correlate of crime (Gottfredson & Hirschi, 1990).
Specific tests of developmental theories such as Sampson & Laub’s (1993) age-graded
theory of informal social control and Moffitt’s (1993) taxonomy are steadily emerging.

In many ways, Sampson and Laub’s age-graded theory of informal social control

53

attempts to reconcile the inherent differences between traditional control theory (Hirschi,
19,69) and contemporary self-control theory (Gottfredson & Hirschi, 1990). That is,
individual differences in early childhood predispositions (low self-control) are mediated
by social structural influences (family). Moreover, these relationships, and by extension
criminality, can at the same time remain relatively stable over time (homotypic
continuity) or change dramatically (see also Loeber, 1982; Moffitt, 1993; Olweus, 1979).
Indeed, Gottfredson and Hirschi (1990) and Sampson and Laub (1993217) are in
agreement in that “. . .the tendency of individuals to remain relatively stable over time on
the dimension of deviance points to the early life course—especially family socialization
and child rearing—as a key causal explanation of early delinquency and a stable self-
control.”

This chapter provided the foundation upon which effective interventions can be
built. Only theoretically meaningful and empirically supported interventions will be able
to effectively intervene in the life course of serious and violent early offenders. The next
chapter will discuss the data, methods and analytic strategy employed in the current study
which is designed to examine the effectiveness of a comprehensive intervention aimed at

serious young offenders.

54

CHAPTER THREE: DATA, METHODS, AND ANALYTIC STRATEGY

Introduction

This chapter will describe in detail the context of the current study, the specific
research questions being posed, and the methods by which each question will be
empirically examined. First, the study context will be presented in terms of a
collaborative intervention being implemented in moderately-sized Midwestern city.
Second, the sample being studied will be discussed in detail. Third, formal hypotheses
will be enumerated which will guide the study. Finally, specific variables and scales to
be employed in testing the hypotheses will be described and an analytic strategy will be

presented.

Study Context

With technical assistance from the Michigan Justice Statistics Center in the
School of Criminal Justice at Michigan State University, police departments in four cities
in the Midwest developed and implemented community-based intervention initiatives
aimed at serious juvenile offenders between the ages of 10 and 13 years old. The current
study will focus on the data obtained from one of the programs. The particular program
at this city was chosen for a number of reasons. First, program administrators adopted
and followed clear selection criteria. Specifically, youth between the ages of 10 and 13
who were arrested for the first or second time for a serious offense who lived in a
specified area of the city were court ordered to participate in the program. Second,

program administrators at this site worked very closely with the evaluation team to

55

provide comprehensive process and outcome data for the evaluation. Third, the city was
the largest among the four sites, and therefore allowed for an adequate sample size for
statistical analysis.

The city in which the program was implemented has a population of
approximately 200,000 and is the second largest city in the state. Based on the 2000
Census, the city was 49% male, 62% White, 20% Afiican American, and 13% Hispanic
or Latino. The city police department employs approximately 387 police officers and
about 100 civilian employees. The target area (Westside of the city) was selected by the
city police department because it had an increasing crime problem and deteriorating
neighborhood conditions. Additionally, residents of this area are generally of lower
socioeconomic status than their counterparts in other areas of the city. For example,
within the target area, the percent of families with annual income greater than $25,000 is
65.91 compared to 68.45 for the city as a whole. Moreover, the unemployment rate in the
target area is 7.6%, compared to 6.1% for the city. Additional demographic
characteristics detailing the city and the target area can be found in Table C1 of Appendix
C.

Program administrators at the site detailed the date and nature (e.g., in person or
by phone) of contacts by police and probation officers and other program officials.
Additional data, including date and duration in minutes, were gathered regarding specific
programs in which individuals participated (e.g., counseling, recreation, and community
service). Discussions with court administrators and police officers revealed that many of
the services incorporated into the intervention have not been readily available to this

population of offenders. As such, this intervention represents a comprehensive approach

56

 

to address serious delinquent behavior that might otherwise be overlooked. According to
program administrators, prior to this intervention, young offenders were commonly
adjusted at intake and released with no formal probation requirements. As a result of this
program, therefore, young offenders receive services not otherwise available to them.

Program administrators collaborated with several stakeholders to develop and
implement an intense programming model to address the needs of youth who had become
involved in delinquent behavior at an early age. The program consisted of three main
components. First, all youth who met the eligibility requirements were identified at arrest
by the Serious Habitual Offender Team (SHOT) of the city police department. The
SHOT team included a number of police officers who had extensive experience working
with serious, violent juvenile offenders. Second, all juveniles were fast-tracked by the
family court. That is, youth who were identified as fitting the criteria were quickly
placed on the caseload of a probation officer participating in the program. The family
court was also responsible for completing risk assessments for all youth in the target age
group. After being placed on formal probation, the youth was enrolled in the program,
which involved intensive supervision probation that includes both concentrated
surveillance and involvement in pro-social services programming. All probation orders
also included mandatory attendance at one of several recreational centers located in the
surrounding neighborhood (or other suitable pro grams as determined by the court).
Throughout the program, members of the SHOT and probation teams maintained regular
contact with participating youth.

The program model was a mixture of intensive supervision, involvement in pro-

social recreation activities, and the reinforcement of positive decision-making within the

57

target population. A summary of the actions that were taken to change how the youth
behave included continuous surveillance, involvement in recreational activities and life-
skills training, and the introduction of positive role models (through mentoring and
tutoring programs).

The intervention was founded on a community policing and probation model that
assigned probation officers to supervise a group of offenders who resided in the same
general neighborhood. This approach allowed community policing officers and
probation officers to coordinate efforts in supervising offenders in the community.
Moreover, the juvenile court employed surveillance officers who intensively monitored
youth above and beyond traditional probation contacts. Surveillance officers commonly
contacted juveniles several times each day (both in person and by phone).

Program administrators noted a number of benefits of the community probation
model. First, because probation officers are physically located in the neighborhoods,
they have better knowledge of the culture of the neighborhood and concerns that
residents may have. Probation officers often attend neighborhood association meetings
and residents occasionally stopped by the probation office to discuss issues relating to
probation clients. Second, probation officers were able to better understand the needs of
the youth and their families because they were able to contact both more regularly.
Historically, probation officers would typically see a client once a month, during the
current intervention, contacts increased to at least once a week (in addition to the
surveillance contacts that occurred daily). Third, probation officers became more aware
of services that were available within individual neighborhoods that were of interest to

youth and their families. For example, a community organization that is located in one of

58

the target neighborhoods houses a Weed and Seed program and provided youth with
recreational programming opportunities.

As described above, many youth involved in the program were ordered to attend
one of several recreational facilities in the target neighborhoods. Funding was directed to
these centers so that they could provide a safe and secure environment where youth could
participate in prosocial activities with peers under the supervision of trained mentors.
The largest and most popular of the recreation facilities had a community policing officer
assigned to it while open to the youth. Facilities offered various activities including pool,
foosball, video games, and organized recreational activities such as basketball.
Additionally, all facilities were equipped with computers that were available for youth to
work on homework or play various games. Finally, each recreation center organized
regular programs such as cooking classes, nature hikes, runs, and other group activities.

All youth involved in the intervention program were subject to intensive
supervision and were required to participate in recreational activities. In addition to these
standard activities, probation officers ordered specific programming to individuals who
had exhibited special programmatic needs. For example, youth who exhibited substance
abuse problems were regularly tested for drug or alcohol use and repeat offenders were
ordered to attend drug counseling. Others attended individual psychological counseling,
family counseling, or anger management classes. Additionally, the court developed an
after-school tutoring program, organized summer recreation camps, and coordinated
community service activities for youth to complete within their own neighborhoods. In a
minority of cases youth were ordered to short-terrn residential treatment or boot camp for

violations of the conditions of probation. As can be seen, the program incorporated

59

common elements yet individualized the program to account for specific needs.
According to program administrators, the intervention aimed to instill in youth positive
attitudes and to create a positive environment for youth to participate in prosocial
activities.

While the family domain was not a primary target of the intervention, some youth
participated in family counseling. More specifically, the family counseling employed as
an element of this program involved regular sessions (usually weekly or twice a month)
with parent(s) and their child or just the youth’s parent(s). Therapists all had Masters
Degree’s in Social Work, and were usually employed through the statewide department
responsible for youth and farnily-related issues.

A key aspect of the program was that it addressed the behavior of youth within
their familial and neighborhood contexts. While extreme family dysfunction or
continued antisocial behavior resulted in an out-of-home placement (detention or foster
care), the intervention was based on a community model that kept youth in their homes if

at all possible.

Sample

Data for this study were collected from program participants and a comparison
group of officially nondelinquent offenders using in-person structured interviews (see
Appendix A for the questionnaire) and fi'om official sources. Program participants self-
reported attitudes, beliefs, behaviors, and relationships with others at three time periods:

upon beginning the intervention, approximately 6 months following the initial interview,

60

and approximately 12 months following the initial interview. Because data are still being
collected, this analysis will be restricted to the first two interviews.

Data were collected from comparison youth at two time points, also with
approximately 6 months between collection points. Comparison youth were
. approximately the same age (10-13 years old) and were recruited from the same
neighborhoods as the program youth (youth were interviewed at a middle school and two
recreation centers in the program target area). This officially nondelinquent group
allowed for the comparison of the two groups on all of the salient family characteristics
(bond and structure) and self-reported participation in delinquency over time.

While exceptions exist (e.g., Hirschi & Gottfredson, 1990), the criminological
community largely agrees that longitudinal designs are preferable to cross-sectional
designs. For example, Farrington (19862212) reports that one “. . .advantage of a
longitudinal study is its superiority over cross-sectional research in establishing cause and
effect, by showing that changes in one factor are followed by changes in another.” For
this reason, outcome variables (delinquency) were measured at interview 2, while
predictor variables were measured at interview 1. In this way, temporal ordering was
specified such that cause preceded effect.

Serious juvenile offenders between the ages of 10 and 13 years were targeted for
this intervention. Initially, serious nonviolent delinquent offenders were the population
of interest. It quickly became apparent (and will be demonstrated in the next chapter)
that program administrators also included many violent offenders (youth arrested for
criminal sexual conduct or assault). As such, the sample represents youth who were

between the ages of 10 and 13 years old when arrested for a serious offense. As

61

discussed in Chapter One, serious offenses include type I index offenses: homicide, rape,
robbery, aggravated assault, burglary, theft over $100, motor vehicle theft, and arson.
From a research standpoint, this strategy selects individuals who are most likely to be
involved in a number of delinquent behaviors. While this approach will yield valuable
information about the frequency, intensity, and variety of incidents in which known
offenders are involved, and is likely the best way to explain variations in offending, it too
has some limitations. The primary limitation concerns external validity. Adolescents
who have been adjudicated delinquent are likely to be very different fi'om the broader
population of active offenders. These youth may engage in more serious offenses, may
be involved in a wider variety of offenses, or may have different social histories than
those who are not formally adjudicated delinquent (Loeber & Farrington, 1998). Due to
the filtering function of the juvenile justice process, many less-serious offenders are often
diverted from the official system and therefore would not be included in a sample of
adjudicated delinquents. Focusing on arrestees addresses this latter issue, but some who
are arrested may not have actually broken the law. As a result, findings from this study
are generalizable only to juveniles between the ages of 10 and 13 who are arrested for a
serious (violent or nonviolent) offense.

Table 1 presents the offenses for which youth were arrested who were involved in
the intervention. As noted, many of the treatment youth had committed a serious
personal offense which resulted in their placement into the program. Specifically,
approximately 22% committed assault, almost 20% engaged in criminal sexual conduct,
and 3% robbed another individual. While the majority of youth were referred to the

program for having committed a property offense (see Table 2), most were of a serious

62

nature (e.g., home invasion or burglary). Also, even though a few drug and status
offenders participated in programming (less than 5%), they did not represent a sizeable
component to the degree that resources were unduly used up by them. These findings
suggest that, in general, integrity was maintained in terms of the seriousness of the

behaviors committed by program youth.

Table 1. Qualifying Offense Information (83 youth)

 

 

 

 

 

 

Percent
B & E/Burglary/Home Invasion 28.9
Assault 21.7
Criminal Sexual Conduct 19.3
Larceny 6.0
R & C 4.8
Retail Fraud 4.8
Auto Theft 3.6
Status (curfew, truancy, runaway) 3.6
MDOP 2.4
Robbery 2.4
Arson 1.2
Drug and Alcohol 1.2
Total 100.0

 

Note: Qualifying offense data were not available for 3 youth.

Table 2. Type of Qualifying Offense (83 youth)

 

 

 

 

 

 

Percent
Property 51.8
Personal 43.4
Status 3.6
Drug 1.2
Total 100.0

 

Sampson and Laub (1993) note that most large-scale longitudinal studies of
adolescent development fail to sample sufficient numbers of serious and/or persistent
offenders (see also Blumstein, Cohen, Roth, & Visher, 1986). While the sample

employed here was not large, it is comprised entirely of youth who have exhibited serious

63

behavioral problems. Similarly, Moffitt (1993) argues that criminological research using
adolescent samples may be inherently flawed because it commonly involves comparing
“apples and oranges;” that is, the grouping of serious and nonserious offenders. This
study avoids this criticism by focusing only on pre-adolescent offenders; those Moffitt
(1993) would characterize as at an elevated risk to become life-course persistent
offenders. Because the subjects initiated offending in childhood and engaged in serious
delinquent behaviors, they represent some of the most at-risk for continuing to engage in

deviance throughout the life course.

S_ample Attrition

Many of the youth who participated in the program moved several times or were
otherwise difficult to locate for follow-up interviews. For example, there is a sizeable
population of migrant workers in the target area, and, as a result, several families moved
considerable distance away from the city, including even leaving the country. With
assistance from the program administrators, the research team made numerous attempts
to contact missing individuals, but was often unsuccessful. Attrition is a common
problem in longitudinal research designs. Specific biases may taint results if youth who
were re-interviewed differ significantly from those who were not re-interviewed. For
example, follow-up interviews that missed the most delinquent of youth may
misrepresent differences that emerge at the second interview. To determine the extent to
which attrition may negatively influence the results of this study, using information

gathered during the first interview, descriptive statistics were calculated for youth for

64

which two interviews were obtained and compared to youth who were only interviewed
at one time point.

In total, 68 out of the 86 treatment youth (79.1%) and 83 out of the original 135
comparison youth (61.5%) participated in a follow-up interview. Table 3 notes some
differences between those who were successfully re-interviewed for this study and those
who were not. For example, among the treatment group more males and younger youth
were more likely to receive a follow-up interview. Among the comparison group, female
and white youth were more likely to participate in the second interview. Moreover, in the
comparison group youth who received a follow-up interview reported a stronger bond to
their primary caretaker and were generally less delinquent at the first interview. This
finding is problematic, but not unexpected. Because participation was completely
voluntary for the comparison group (and only slightly more encouraged for the treatment
group) it is not surprising that youth involved in delinquent behavior were less likely to
submit to be interviewed a second time. Even though participants were reassured that
study staff were not associated with school, police, or other formal social control
agencies, and confidentiality was repeatedly stressed, some youth may have not felt
comfortable continuing to participate. This problem is not unique to the current study,
and likely permeates many repeated-measure designs. Nevertheless, based on these

attrition diagnostics, results should be interpreted accordingly.

65

Table 3. Attrition Diagpostics

 

 

Treatment Group Comparison Group

i1 i2 missing i1 i2 missing

Number 86 68 18 135 83 52
(20.9%) (38.5%)

% Male 80.0 94.0 76.0* 57.0 49.0 69.0*
% White 50.0 45.6 66.7 34.1 41.0 23.1*
Mean Age 12.4 12.2 13.0* 11.9 11.9 11.9
Mean Family Bond 3.30 3.29 3.36 3.53 3.59 3.43“
% Single Parent 74.4 75.0 72.2 39.3 37.3 42.3
i1 SR Violent 75.6 77.9 66.7 45.9 38.6 57.7“
i1 SR Nonviolent 72.1 69.1 83.3 35.6 31.3 42.3
i1 SR Drug 66.2 67.7 61.1 16.3 10.8 25.0*

 

Significant difference between missing and i2 - p“ < .05 (two-tailed test)

Missing values are not otherwise problematic with these data. If values were
missing, however, the cases were excluded listwise for all analyses. The final sample
consisted of 68 treatment youth and 83 comparison youth who were interviewed at least

two times over the course of six to nine months.

Hypotheses

All of the following hypotheses sought to assess the importance of family
relationships on intervention success for young offenders. As reviewed above, the family
is especially important among younger individuals (preadolescent or younger). As such,
this study explicitly tested the importance of family relationships on intervention success

with this special population of offenders. The first two hypotheses focused on conditions

66

at the beginning of the intervention, while the latter two hypotheses tested the effect of
intervention-influenced changes in family relationships. The primary research question
for this study was: are early serious offenders are more amenable to treatment efforts if
they begin the intervention having strong relationships with their parent(s)? Specifically,

the first hypothesis asserts:

Hypothesis 1. Youth who have a strong relationship with their primary caretaker upon
beginning an intervention are more likely to desist offending.

 

 

 

Strength of Initial Family Bond (-) Delinquency

 

 

 

 

 

 

Based on the theoretical and empirical evidence reviewed above, it was expected that
youth who reported a strong family bond at the start of the intervention were more likely
to desist offending. Strong relationships with parents provided youth with a supportive
environment within which delinquency desistance could occur. A youth who is strongly
attached to his or her parent is more likely to trust in his or her advice and follow
directives given by that parent. Strong communication indicates that the child is willing
to discuss problems with his or her parent. Sufficient supervision suggests that parents
know what their children are doing, and therefore are able to control deviant behavior and
reinforce prosocial decision-making. Finally, parents who employ creative and positive
parenting skills are better able to manage stressful situations. As detailed below, the
measure of family bond employed in the current analysis included elements of all of these

family dynamics (attachment, communication, supervision, and parenting skills).

67

The second hypothesis will examine the effects of family structure at intake on

intervention success:

Hypothesis 2. Youth living with two parents upon beginning the intervention will be
more likely to desist offending.

 

 

Two Parent Family (-) Delinquency

 

 

 

 

 

 

As reviewed in Chapter One, a growing body of research suggests that youth living in a
single-parent family are at an elevated risk for delinquency, drug use, and other
undesirable outcomes. Two parents can more effectively supervise their children which
may buffer the youth from other negative influences, such as deviant peers. On the other
hand, others have noted that it is not the structure of the family that is important, but the
quality of the relationship between parent and child. While the first hypothesis addressed
the importance of “quality,” hypothesis two considers structure.

The third hypothesis assessed the importance of including family programming

(such as family counseling) in interventions that target young offenders.

Hypothesis 3. Youth who receive family-focused programming will be more likely to
desist offending.

 

 

Family-based Programming (-) Delinquency

 

 

 

 

 

 

While many of the youth did not receive family-based programming, it is hypothesized
that those who did will be more likely to desist offending. F amily-based programs have

proven more effective at rehabilitating juvenile offenders than programs that do not target

68

 

 

the family domain (Latimer, 2001; Perkins-Dock, 2001). Moreover, when included as
one of many programmatic activities the youth participates in, family-focused
programming can prove especially effective.

The final hypothesis examined the extent to which youth who improved their
relationship with their primary caretaker were more successful at abstaining from

delinquency during the intervention.

Hypothesis 4. Youth participating in the intervention who report improved family
relations from interview 1 to interview 2 will be more likely to desist
offending.

 

 

(-)

 

 

 

 

Improvement of Family Bond Delinquency

 

 

 

Even though many individuals began the intervention reporting poor relationships with
their primary caretaker, if relations improved—either as a direct result of the intervention
or not—it is hypothesized that behavior will also improve. Some youth participated in
family counseling, while others received a variety of programs that may have indirectly
resulted in improvement in the relationship between the child and his or her parent(s).
Moreover, all treatment youth were placed on intensive supervision which required them
to be at specified places throughout the day, including being home by a specified time.
These restrictions may have resulted in the youth spending more time at home or with

parents thereby improving their relationship with them.

69

Analytic Strategy
Dependent Lariajrles

Both official and self-reported measures of delinquent involvement were
employed as outcome variables in this study. It is important to note that official and self-
reported measures of delinquent behavior are independent approximations of one’s actual
involvement in delinquent behavior (Thomberry & Krohn, 2000). Further, Maxfield,
Weiler, and Widom (2000:108) note that “Multiple measures are especially valued for
such elusive constructs as criminal offending.” Moreover, researchers recognize that
both self-report and official measures of delinquency have notable sources of error. “The
types of errors associated with police arrest records and self-report surveys are
increasingly known, if not always acknowledged by criminologists” (Maxfield et al.,
20002108). For example, due to inconsistencies in measurement and possible screening
biases throughout the justice process, the reliability of official statistics has been
questioned for years (Hindelang, Hirschi, & Weis, 1981). Undoubtedly, a significant
number of individuals who break the law are not apprehended and therefore are not
included in official reports of delinquency. Finally, a wide range of deviant activities that
are of interest to juvenile delinquency researchers are often not captured by official
reports, such as truancy, parental defiance, or sexual behaviors.

Official recidivism was measured using arrest records of the individual for the six
months following the initial interview. This approximates the time that the youth was
still involved in the intervention, though some youth may have already been discharged

from it. Individuals who were not rearrested following the first interview were coded a

70

success using a binary variable (0). Any arrest, excluding technical violations of
probation, was coded failure (1).

Self-reported involvement in delinquent behavior was measured at the second
interview using three prevalence scores: violent offending, nonviolent offending, and
drug or alcohol use. This strategy more clearly specified the types of behavior in which
the juvenile was involved. In contrast, studies which operationalize delinquency as a
frequency of a wide variety of offenses may blur the extent to which the individual is
involved in serious offending (Loeber, Fanington, & Washbusch, 1998). In addition,
because a different reference period was used for the first interview (previous 12 months)
and the second interview (previous 6 months) in this study, frequency measures of
involvement in delinquent behavior that were collected may misrepresent participation in
those activities. For example, a youth who reported smoking everyday would be coded
as having smoked 365 times at interview 1 and 180 times at interview 2. Even though the
youth was still involved in the activity at the same frequency, statistically it would appear
that the youth decreased involvement by half. While prevalence scores are a conservative
estimate of the frequency and intensity of participation in specific activities, the strategy
allows for an understanding of what factors may contribute to desistance. Future
analyses may revisit frequency measures of delinquency to determine if any differences
exist between the two approaches.

The three prevalence scores each include multiple offenses in an effort to measure
the extent to which youth were involved in a specified type of offending. Violent
offenses included: assault a peer, assault an adult, and thrown rocks or bottles at people.

Nonviolent offenses included: MDOP, theft fiom parents, trespassing, theft from car,

71

shoplifting, graffiti, arson, and auto theft. Drug and alcohol use included: consumed any
liquor, smoked marijuana, smoked cigarettes, and used cocaine. Youth who reported at
the second interview that they abstained from engaging in violent behaviors since the
previous interview were coded as desistors (0) for that offense type. This group of
offenders also included youth who did not participate in the activity during the 12 months
preceding the initial interview. Individuals who report any involvement will be coded as
persistors (1). This strategy was also employed for nonviolent offending and drug and

alcohol use.

_Ir_r_dependent Varjabfi

As discussed above, the primary independent variable for this study is the strength
of the relationship a youth has with his or her primary caretaker. Respondents were
asked questions about their mother and father (or, if applicable, a different guardian) if
they resided with him or her (or both) for the majority of the previous six months. A
primary caretaker variable was created which represents the parent with whom the child
spent the most time during the assessment period. After interviewing began, evaluators
added a question which asked respondents this question. For some individuals, this
question was not asked, but primary caretaker was inferred based on the following
criteria: primary caretaker was determined by selecting the only parent (if it was a single-
parent home), the biological parent (if one parent was biological and the other was not),
or the mother if neither of the other conditions were present.

The family bond scale consists of items representing a variety of questions

pertaining to the relationship one has with his or her primary caretaker. Confinnatory

72

factor analysis was employed to ensure that each of the variables measures the same
underlying construct (family bond). After confirmation, an mean score was computed for
family bond that includes the following 10 questions: (1) “I feel I can really trust my
primary caretaker”; (2) “I really enjoy my primary caretaker”; (3) “I think my primary
caretaker is tenific”; (4) “I feel proud of my primary caretaker”; (5) “I feel I can trust my
primary caretaker with just about anything I tell him/her”; (6) “I talk to my primary
caretaker about trouble I am having at work or school”; (7) “I go places or do things with
my primary caretaker”; (8) “I think highly of my primary caretaker”; (9) “my primary
caretaker is a person I want to be like”; and (10) “I really enjoy spending time with my
primary caretaker”.

The response set for items 1-4 is “never”, “once in a while”, “sometimes”, and
“always”. For items 5-7, the response set is “very likely’, “likely”, “unlikely”, and “very
unlikely”. Finally, the response set for items 8-10 is “strongly agree”, “agree”,
“disagree”, and “strongly disagree”. Items loaded on a single factor with loadings greater
than 0.5. Some items were reverse coded such that higher values on the mean scale
indicate a stronger relationship with one’s parents. While some have suggested that
problems arise when the number of variables used to comprise a scale approaches N,

Marsh & Ban (1999) demonstrate that in most cases, the higher the number of indicators

per construct, the more valid and reliable the measure.

73

Table 4. Family Bond Component Variables and Factor Loadings

 

 

Factor
Loading

1. I feel I can really trust my primary caretaker 0.616
2. I really enjoy my primary caretaker 0.776
3. I think my primary caretaker is terrific 0.786
4. I feel proud of my primary caretaker 0.623
5. I feel I can trust my primary caretaker with just about anything 0.752
6. I talk to my primary caretaker about trouble I am having at work or school 0.539
7. I go places or do things with my primary caretaker 0.759
8. I think highly of my primary caretaker 0.616
9. My primary caretaker rs a person I want to be like 0.708
10. I really enjoy spending time with my primary caretaker 0.730

 

Eigenvalue: 4. 83
Chronbach’s Alpha: 0.875

 

In addition to using a mean score as continuous variables representing the level of
bonding a youth has to his or her parent at the first interview, a change score was created
representing the change in family bond from interview 1 to interview 2. It is
hypothesized that youth who report improved relations with their parents will be more
likely to desist offending.

The number of minutes of family—based programming was also used to predict
intervention success. As noted, it was hypothesized that youth who received family-
based programming such as family counseling will be more likely to desist offending.
These data were obtained from program administrators who documented the dates and
duration (in minutes) of all programs in which the intervention youth participated.
Finally, family structure was operationalized as single-parent (coded “1”) and two-parent

(coded “0”) households.

74

Control Variables

Because a small sample was used for this study (86 treatment and 135 comparison
youth), few additional variables can be added to the models while maintaining sufficient
statistical power. This is particularly true for the longitudinal analyses in which the
sample size was even smaller (68 treatment and 83 comparison youth). Thus, only four
control variables were added to the multivariate models. Age (age in years), ethnicity
(white/nonwhite), and gender (male/female) will be used to control for the individual
effects of each of these variables. Additionally, for the treatment group, the number of
days between the intake date and the initial interview will be included as a control for any
treatment received prior to the initial interview. Even though the research team attempted
to interview treatment youth immediately upon beginning the program, delays inevitably

occurred.

Quantitative Method_s and Modefi

In the first stage of this analysis, data will be presented to describe the sample
being studied. Relatively few existing studies have focused on young serious offenders;
therefore, information gleaned from these initial analyses will-provide researchers and
practitioners with a better understanding of the characteristics unique to this population.
Specifically, descriptive statistics were conducted for all of the variables being employed
in later analyses. Group means were compared using independent samples t-tests to
determine if there were any statistically significant differences between the comparison
and delinquent groups on important characteristics (including self-reported delinquency

and family bond).

75

The next stage of this analysis will involve statistical analyses to test each of the
hypotheses enumerated above. Where appropriate, models were estimated for both the
treatment and comparison groups, so that any differences between the groups can be
ascertained. Conducting a combined analysis with both groups would not allow for the
comparison of changes over time (the comparison of slopes). It is already known that the
two groups differed in terms of their family relationships at the time of the first interview.
This study seeks to determine .whether there are significant differences between the two
groups in terms of how participants change over time. If delinquent youth improve their
relationship with their parent(s) more dramatically than the comparison youth, some of
those changes may be attributable to the intervention (although lack of random
assignment to each group precludes any degree of certainty).

Bivariate and multivariate logistic regression analysis will be performed to assess
the impact of continuous predictor variables (strength of family bond, change in family
bond, and family programnring received) on each of the binary outcome variables (self
reported violent, nonviolent, and drug offending, and official delinquency). The final
multivariate models sought to determine the extent to which familial relationships and
family programming were important while controlling for the effects of age, ethnicity,
gender, and previous treatment. Contingency table analysis was used to determine the
relationship between family structure at interview 1 and future delinquency.

The final stage of the analysis will employ Hierarchical Linear Modeling (HLM)
to further test the association between the strength of the family bond at interview 1 and
future delinquency (Raudenbush, Bryk, Cheong, & Congdon, 2000). There are a number

of advantages to using HLM. First, modeling change within the individual using HLM

76

does not require that the spacing be the same between observations (Raudenbush & Bryk,
2002). Additionally, HLM also considers the effect of individuals for whom a
subsequent interview was not obtained (Raudenbush & Bryk, 2000). Finally, HLM
allows for the use of parallel scores to represent the measurement of key concepts which
estimates statistical models for a (albeit artificially) larger sample. Specifically, as
discussed above, the family bond concept in the current study was measured using ten
variables that estimate the relationship youth have with his or her primary caretaker.
Parallel scores were created by computing two mean scores that each used five of the
variables from the original operationalization of family bond. Variables can be assigned
randomly (see, Lyons, Zarit, Sayer, & Whitlatch, 2002) or deliberately to ensure similar
variances (Coley & Morris, 2002). The latter method was chosen for the current study.
Theoretically, the two scores still represent a latent approximation of the relationship
each youth has with his or her primary caretaker.
Assumptions and Diagnostics

The primary quantitative method employed in the current study was logistic
regression. Logistic regression is the variation of regression most appropriate for models
which include a dichotomous outcome variable. Using a maximum-likelihood estimation
method, logistic regression predicts the odds (or likelihood) of an event occurring given
specified conditions (or independent variables). While logistic regression is not
constrained by all of the assumptions of ordinary least squares (OLS) regression
(Menard, 1995), there are some assumptions that need to be kept in mind. For example,
both OLS and logistic regression assume that the model is correctly specified. Including

irrelevant or excluding relevant variables can result in a misspecified model. This is less

77

a concern for the proposed study because the purpose of this research is not to test the
merits of a particular theory per se, but to assess the predictive power of one construct
common in many theories (family bond) on an outcome variable. Statistically, the odds
ratio will be of more interest in this study than the explained variance (r2). Nevertheless,
there could be additional variables not considered that may affect the way a person’s
family bond influences delinquency.

Collinearity is another potential problem that can negatively bias both OLS and
logistic regression models (Menard, 1995). Independent variables that are highly
correlated may result in inflated standard errors, thereby increasing the likelihood of
Type H error. Collinearity was initially assessed by reviewing the bivariate correlations
between all variables (see Tables B1 and B2, in Appendix B). No predictor variables
included in any of the models together were significantly related to each other.
Collinearity statistics were also computed in OLS to determine the influence of
relationships between independent variables. Specifically, tolerance statistics below 0.20
(Menard, 1995) and variance inflation factors greater than 1.5 (Fox, 1991) may indicate a
problem. With regard to multicollinearity, variance inflation factors (VIF) ranged
between 1.012 and 1.087 and tolerance statistics ranged between 0.920 and 0.988 which
suggests little cause for concern (Fox, 1991).

Unlike OLS, logistic regression does not assume that errors are normally
distributed or homoscedastic (Menard, 1995). As such, influential outliers are not as
problematic using logistic regression. As a final diagnostic, however, Standardized
residuals were computed and Cook’s Distance (D) statistic was reviewed with an eye

toward values inconsistent with others. Cases with extreme residuals do not fit the model

78

and therefore may indicate a typology that is inconsistent with the current hypotheses.
Standardized residuals ranged from -1.808 to 1.296 and Cook’s D ranged from 0.000 to

0.075. These numbers are not indicative of problematic cases in the sample.

Summary

For a number of reasons, this research represents an important contribution to the
criminological literature. First, few studies have focused on serious juvenile delinquents
between the ages of 10 and 13. As discussed above, youth who initiate offending at this
early age are at a greater risk to become career criminals. Successfully interrupting their
paths at an early age can result in substantial fiscal as well as social benefits. Second,
this study attempts to demonstrate that the influence of the family is an important
dimension often missing from juvenile interventions. Results from this study will help
program developers in creating interventions that are responsive to the needs of serious
juvenile offenders. It will also seek to demonstrate how a common element fiom several
criminological explanations (family influences) can be used to ground intervention
efforts. The following chapter presents the results of several statistical tests conducted to

test the four hypotheses presented.

79

CHAPTER FOUR: FINDINGS

As should be clear by now, the purpose of this study was to assess the importance
of targeting family relationships in interventions aimed at young offenders. Another
notable characteristic of this research is that hypotheses are being tested using a unique
sample, namely, serious young offenders. In testing the hypotheses enumerated in
Chapter Three, this chapter will proceed in the following stages. First, sample
characteristics will be presented, followed by a comparative look at the differences
between the delinquent group and a comparison group. Second, each of the hypotheses
will be tested, in turn, using both bivariate and multivariate statistical models estimated
for both the treatment and comparison groups in an effort to ascertain any differences.
between the groups in terms of the influence of the family on future delinquency.
Because very few youth in the comparison group were arrested in the 6 months following
the initial interview as gleaned from official reports (less than 4%), statistical models for
these youth will rely on self-reported outcome measures. Finally, Hierarchical Linear
Modeling will be employed to conduct additional analyses aimed at better understanding

the relationship between familial relationships and intervention amenability.

Descriptive Statistics

As reported in Table 5, 80% of the treatment youth were male, 50% were white,
24% were Afiican American, and 24% were Hispanic or Mexican. As specified by the
program, the vast majority of youth (87.2%) were between 10 and 13 years old when first
interviewed (mean=12.4). Finally, most of the treatment youth (75.6%) were in junior

high school (grades 6-8) when participating in the program. There are some notable

80

differences in the demographic makeup of the comparison group. For example, 43% of
comparison youth were female, 34% were white, 31% were Hispanic or Mexican, and
27% were Afiican American. Moreover, the treatment group was older and more
advanced in schooling than the comparison group. Due to these differences, results based

on comparisons of the two groups must be interpreted with caution.

Table 5. Descriptive Characteristics of the Treatment and Comparison Groups

 

 

 

 

Treatment Comparison Total
=86) (N=135) (N = 221)
Number Percent Number Percent Number Percent
Male 69 80.2 77 57.0 146 66.1
Female 17 19.8 58 43.0 75 33.9
m
African American 21 24.4 37 27.4 58 26.2
Hispanic/Mexican 21 24.4 42 31.1 63 28.5
White 43 50.0 46 34.1 89 40.3
Native American 1 1.2 0 0.0 l 0.5
Biracial O 0.0 6 4.4 6 2.7
Missing 0 0.0 4 3.0 4 1.8
433
9 3 3.5 4 3.0 7 3.2
10 4 4.7 16 l 1.9 20 9.0
1 l 9 10.5 37 20.0 46 , 20.8
12 21 24.4 38 28.1 59 26.7
13 41 47.7 43 31.9 84 38.0
14 8 9.3 4 3.0 12 5.4
15 0 0.0 2 1.5 2 0.9
Mean (Std. Dev.) 12.36 (1.16) 11.90 (1.23)*
Grade
3 2 2.3 1 0.7 3 1.4
4 3 3.5 16 1 1.9 19 8.6
5 5 5.8 12 8.9 17 7.7
6 17 19.8 30 22.2 47 21.3
7 26 30.2 39 28.9 65 29.4
8 22 25.6 17 12.6 39 17.6
9 5 5.8 0 0.0 5 2.3
Missing 6 7.0 20 14.8 26 l 1.8
Mean (Std. Dev.) 6.85 (1.31) 6.23 (1.28)*

 

‘Mean difference p < .05 (two-tailed test).

81

Table 6 presents statistics describing the nature of the family enviromnent for
both the treatment and comparison groups. For example, 74% of treatment youth and
39% of the comparison youth lived with one parent at the time of the first interview.
Moreover, as predicted by the theories reviewed in Chapter Two, treatment (officially
delinquent) youth reported a weaker bond with their primary caretaker than comparison
(officially nondelinquent) youth. Over time, the average family bond decreased for both
groups at about the same rate. This trend is consistent with developmental research
which indicates that during adolescence youth move away fi'om the family and toward
the peer group as a primary influence (W arr, 2002). This finding also suggests that the
intervention did little to improve family relationships among the treatment youth as a
whole. Because the current study lacks a control group of delinquent youth who did not
receive intervention programming, it is uncertain whether the family relationships of

delinquent youth would have deteriorated fiuther without the programming.

Table 6. Family Relationships by Group

 

Treatment (N=68) Comparison (N=83)
Mean Std. Dev. Mean Std. Dev.
Single Parent F amin 0.74 0.44 ' 0.37* 0.49
Family Band
Interview 1 3.29 0.56 3.59* 0.39
Interview 2 3.27 0.54 3.57* 0.42
Change over time -0.016 0.54 -0.018 0.33
Family Programming‘ 379.14 257.31 -— --

 

1Mean number of minutes of family programming for youth who received it.
"Mean difference statistically significant (p < .05; two-tailed tests).

Table 7 displays the self-reported and official delinquency of all youth. Almost

78% of treatment youth and 39% of comparison youth reported participation in violent

82

behaviors in the 12 months preceding the initial interview. Significantly fewer youth in
both groups reported participation in these behaviors in the 6 months leading up to the
second interview; about 63% of treatment youth and less than 22% of comparison youth
reported violent behaviors at interview 2. There was also a substantial (though not
statistically significant) decrease in the percent of youth reporting involvement in
nonviolent delinquency. The percentage of treatment youth involved in nonviolent
offending dropped from 69 to 56, while comparison youth involved in the same behaviors
decreased from 31% to less than 22%. Moreover, fewer youth in both groups reported
using drugs or alcohol (67.6% to 51.5% for treatment youth and 10.8% to 8.4% for
comparison youth). Finally, about 22% of treatment youth were arrested in the six

months following the initial interview compared to less than 4% of the comparison youth.

Table 7. Self-Reported and Official OffendinLCharacteristics

 

 

 

Treatment (N=68) Comparison (N=83)
Number Percent Number Percent
Interview 1
Self-reported violent 53 77.9 32 38.6
Self-reported nonviolent 47 69.1 26 31.3
Self-reported drug use 46 67.6 9 10.8
Interview 2
Self-reported violent 43 63.2*- 18 21.7*
Self-reported nonviolent 38 55.9 18 21.7
Self-reported drug use 35 51 .5“ 7 8.4
6 month official failure 15 22.1 5 3.7

 

*Represents a statistically significant difference from interview 1 (p < .05; two-tailed tests)

Findings from this first stage of the analysis indicate that there are some important
differences between the comparison and treatment groups in terms of demographic

characteristics, family structure and bond, and self-reported delinquency. Notably, the

83

treatment youth start out with a weaker bond to their primary caretaker than their
comparison counterparts, but this relationship changed over time at about the same rate
for both groups. This finding suggests that the intervention was not successful at
improving family relationships among treatment youth. Additionally, fewer youth from
both groups reported participating in delinquent behavior and drug use at the second
interview. Decreases were more dramatic for the treatment youth, suggesting the
improvement may have been associated with intervention programming. The following
sections will evaluate the empirical support for the four hypotheses outlined in Chapter

Three.

Hypothesis 1: Family Bond at Intake and Delinquency Desistance

The first question this study addressed was upon beginning an intervention what
effect do family relationships have intervention success? Simply put, do youth who
begin; an intervention with a strong bond to their parent(s) succeed at desisting offending
at a disprOportionate rate compared to youth who are not strongly bonded to their
parent(s)? Using logistic regression analysis, this stage of the study modeled the
association between the strength of the family bond at intake (interview 1) and future
delinquency (interview 2).

As Specified in Hypothesis 1, it is expected that youth with a strong bond to their
primary caretaker will be more likely to desist offending. Tables 8 through 11 display
results from the logistic regression analysis. While largely statistically insignificant, a
negative relationship between family bond and delinquency emerged in almost all of the

bivariate and multivariate models. For the treatment group, the stronger the family bond

84

at intake into the program, the less likely the youth was to self-report future participation
in violent activities and drug use. A small and insignificant positive relationship emerged
between strength of family bond and future self-reported nonviolent behaviors. In terms
of control variables, age was positively related to delinquency; that is, older youth were
more likely to report involvement in delinquency and to be arrested. Among members of
the treatment group, number of days of programming prior to the first interview was
inconsistently related to delinquency. Youth who were in the program longer prior to
being interviewed for the first time were more likely to self-report drug use and to be
rearrested (statistically significant); however, these youth were less likely to self-report
violent or nonviolent offending (statistically insignificant). These findings are difficult to
explain, and require additional analysis to better understand.

Similar findings emerged for the comparison youth. That is, as theoretically
expected, youth who reported a strong bond with their primary caretaker were less likely
to self-report firture delinquent behavior (though statistically insignificant). Among both
groups, males and older youth were generally more likely to self-report involvement in
delinquency and drug use.

In short, while there appears to be a negative relationship between strength of
family bond at interview 1 and later delinquency, results are mostly statistically
insignificant and largely inconsistent. Moreover, because results were relatively similar
for both the treatment and comparison groups, it is does not appear that the intervention
improved the delinquency-prevention capabilities of a strong bond with a parent, as

hypothesized.

85

A88“ “co—5625 3. V a: “no. v a... Aewnz Joeofi Decadence 3050 E unooxov wonz $805589 5 Echo 3353 ”802

 

 

 

 

 

 

 

as. m8. m2. 8N. Ne geomomaz

m _ N. one. coo. m2. NM =oem am 30
A83 83; am: 33

m _ a. ewe: o2: c8. Rh. ohm- mwe. 3m: 83:: “a 28m £88.."
9.25 A83 38¢ 92:

_ S.— ..L S. gee. So... wee. woo; 08; .80. Samoa 3 939
some A83 £5 £3

emeg anew. 2: 2:. SN; EN. owed sums. ow<
:03 ans 68¢ ca:

own; «mm. «mm; ewe. 34.; own. emu. 3 mt £32
53 some 53 $3.

vow.“ emm. :b. one: 2%. «no: Nmn. «8.. BE?
$3.3 $8.3 32.0 ants

mid- out"- v2..- $28. “5800
Exam .mooo Seam ..eooo Exam .eooo Exam coco

83 an wotomom..=om E33262 ootoqomfiom 3035 votoQoM$ow moaosunflon 3050

49.20 «5826...: 3.3:— «a one: man—ah I 53.81..me 83g.“ Saw—55...: .e «5:.

Gmuz 4268 mono—avenue 308° 3 «nooxov weuz 683883 5 adobe Eden—Sm 6qu

So. so. so. So. am oxsaomaz

So. So. so. So. «a :06 a. U50
:3; $43 $5 . as:

com. mNNr EQ— NS. Se. wear on». 57 £85 “a teem 38am
as... c 81 can: 82.:

men. :5. now. own; «.5930
Evan. ..euoo Seam £80 Seam e08 Exam ..eooo

9.5 man entomomfiom 322an Z entomomfiom =83; “coughing—om mono—29509 3050

 

aseO 2.253an 8.55 “a teem—N=83— I =efi8fie¢ flames— Satazm— .a «Emu.

86

A88“ 3:863: 3. v 9. "885539 8 98.5 9353 No.02

 

 

 

men. an. E _. R 33:82
.5. 9:. as. "m :08 a. use
as. a some 653
a3. «2.. ma. 33$ 8m. «3.. £85 a 28m Seam
$5 $3 53
:3 L8. 8: e8. 8: E. owe.
22.: 3.3 A83
5.2 .33 So. as..- 83 o8. 032
2a.: $3 53
$5. 2.2 new 83 :2 3a. 223
82.3 See 93$
33.: 3.4.4 8a.- 3388
@me ..eooo @mxm ego Exam Eco
83mg venomomfiom 303332 mahogannom «no—o; potency—.aom

 

38a 3:86.53 3. v a. ”momueeoeaa 3 adobe 3%qu ”8oz .

Amwuz 3:30 courage—e8 a 33:8:— «a BBQ bun—am I eefimoflmm otawex— 8.3.3.532 .2 03am.

 

 

 

:o. 28. «3. Ne oasaomaz

so. So. as. Ne :05. a. .80
Age: as: $3

New. 5.. SN. .3. 7 Sn. 3a.- 395 a 28m amuse
a: .c awmd ENS

an- as... $2 .888
Exam e08 @axm .eooo scam e80
omDMaQ votomomubom ado—ogaoz votoamoymrfiom ado—0C, UntOfiMIu—ow

 

Amwuz E520 eaten—neg ~ .5328:— 3 teen bias..— I 5.80.83— oufiwed 825$:— 6— 93:.

87

Hypothesis 2: Family Structure at Intake and Delinquency Desistance

Table 12 presents results from contingency table analysis testing the relationship
between family structure at intake and delinquency for the treatment youth. The chi-
square coefficient was employed to test the null hypothesis that there is no relationship
between family structure and delinquency. Because the chi-square statistic was not
statistically significant in any of the models, there does not appear to be a relationship
between family structure and delinquency — contrary to hypothesis 2. Moreover, in
general, a higher proportion of youth who lived with two parents persisted in delinquent
activity than those who lived with only one parent (except for the drug use model). In
short, results indicate no relationship between family structure and delinquency
desistance in the treatment sample.

Table 12. Contingency Table Analysis: Family Structure at Intake and Future
Delinquency (Treatment Group; N=68)

 

 

6 Month Self- Self- Self-

Official Reported Reported Reported

Delinquency Violentl Nonviolentl Drug Use1
Single- NO (N=17) 29.4 70.6 64.7 47.1
Parent YES (N=51) 19.6 60.8 52.9 52.9
Chi-square .399 .468 .398 .674
Phi -.102 -.088 -.103 .051

 

Values represent percent delinquent in each category
lReported delinquency at interview 2.

Table 13 presents the findings from the same contingency table analysis (family
structure and delinquency) for the comparison group. Again, there does not appear to be
a significant relationship between family structure at interview 1 and self-reported

delinquency at interview 2. In two of the models (violent and nonviolent offending),

88

youth living with only one parent were more likely to be involved in offending, however,

this relationship did not hold for the drug and alcohol use model.

Table 13. Contingency Table Analysis: Family Structure at Intake and Future
Delinquency (Comparison Group; N=83)

 

 

Self- Self- Self-
Reported Reported Reported
Violentl Nonviolentl Drug Usel
Single- NO (N=52) 21.2 19.2 9.6
Parent YES (N=31) 22.6 25.8 6.5
Chi-square .879 .482 .616
Phi .017 .077 -.055

 

Values represent percentage delinquent in each category
1Reported delinquency at interview 2.

Taken together, results are largely equivocal, and do not allow for a better
understanding of the relationship between family structure and delinquency, regardless of
whether or not a child was involved in intervention programming. Better specification of
the type of family structure may be necessary. For example, youth who live in a two-
parent household in which one of the parents is a non-biological guardian (step-parent)
may be more at risk for participation in delinquency (Demuth & Brown, 2004).
Moreover, youth who reside with two deviant parents would be more at risk than youth
who live with one prosocial parent. These issues suggest that a simple dichotomy
between one- and two-parent households may be inadequate in assessing risk. Type of
relationship (or strength of affect), however, may be a more important characteristic of

family dynamics.

89

Hypothesis 3: Family Based Programming and Delinquency Desistance

Hypothesis 3 seeks to better understand the effect of family programming on
delinquency among members of the treatment group. Because many effective
interventions focus on improving family functioning (e. g., Functional Family Therapy,
Parent Management Training), it was hypothesized that youth who received family based
programming would be more likely to desist offending. Again, using logistic regression,
this hypothesis was tested by determining the relationship between total number of
minutes of family programming received and future delinquency. Tables 14 and 15
report the results of this analysis. Both at the bivariate and multivariate level, odds ratios
were very close to 1.0 (0998-1002), which indicates that the family programming in the
current intervention had little effect on either self-reported or official delinquency.
Consistent with previous analyses, age and number of days in the program were
positively related to drug use and official delinquency.

Table 16 presents the distribution of persistent offenders by treatment received
(comparison group who received no treatment, basic treatment group who received
intensive supervision and participated in recreational and other general programming, and
a family treatment group who received the same programming as the basic treatment
group but who also participated in family counseling). Matched pair t-tests were used to
assess the significance of changes in the percent of youth self-reporting delinquency over
time. As noted in Table 16, the proportion of comparison youth reporting all forms of
deviant behavior decreased over time (the only significant difference was in the percent
reporting violent offending). Specifically, about 39% of comparison youth reported

violent offending at interview 1, compared to less than 22% reporting the same behaviors

90

at interview 2. Similar decreases were noted for nonviolent offending (from 31.3% to
21.7%) and drug use (from 10.8% to 8.4%). It is unclear what mechanism caused these
changes; it could be that more delinquent youth in the comparison group were
unavailable for reinterview (see attrition analysis in Chapter 3) or that the shorter follow-
up period (6 months compared to 12 months) did not allow ample time for participation
in the activities.

Encouraging findings emerged for the basic treatment group in that significantly
fewer youth reported all forms of offending. For example, almost 87% of the basic
treatment youth reported violent offending at interview 1 compared to about 62% at
interview 2. Equally as dramatic was the decrease in the percent of youth in the basic
treatment group who reported involvement in nonviolent offending (from 78.4% to
48.7%) and drug use (from 75.7% to 54.1%). These findings are promising for the
intervention as a whole in that they suggest at least modest short-term improvement in the
behavior of the basic treatment group.

The crux of the current study and hypothesis 3, however, is to assess the
importance of family treatment among serious young offenders. The third group (family
treatment group) received the same general services as the basic treatment group, but also
participated in some form of family counseling. Data presented in the last three colmnns
of Table 16 suggest that these youth did not respond as well to the intervention as the
basic treatment group did. No statistically significant decreases were reported for any
form of offending and, in fact the percent of youth involved in nonviolent offending

increased.

91

While it is difficult to explain the failure of family programming to emerge as an
important component in this intervention program, some speculation is warranted.
Notably, the intensity and duration of family programming in this intervention was
relatively weak. Some youth received family counseling while the parents of others
participated in counseling. Youth who received family counseling demonstrated familial
dysfunction sufficient to warrant increased services. Because more comprehensive
services were not available in this community, traditional family counseling was
employed. While family counseling has proven generally effective in other contexts
(Lipsey & Wilson, 1998), more comprehensive family programming such as Functional
Family Therapy or Parent Management Training may have been more effective than just
counseling. Future research should seek to determine the effectiveness of various forms
of family programming in an effort to identify which elements are successful for which
groups.

Another potential explanation for the apparent lack of effectiveness of the family
counseling may be related to the relatively short follow-up period associated with the
current analysis. Family problems are often longstanding and difficult to ameliorate.
Counseling over a longer period of time may increase the likelihood of obtaining positive
results from its programming. Future analyses of these data with a longer follow-up

period may shed more light on this issue.

92

oma coma-93v 3. v a... ”335589 5 82.8 3356 ”302

 

 

 

 

 

 

gm. m8. “2. E. "a £35032

3.. a8. 2 _. E. "a :06 a. Uhco
:2: 28c :23 A83

83 8o. 83 N8. 83 «8. as. so.- mafifiuoa Esau
A33 32: 283 A33

:3 a; 8. 83 25. 8a. 8a.. 83 :8. 5&5 a 35
A83 82¢ $3 $3

32 .8». 33 a. $2 an. 3.2 :3. ow<
53 A83 $3 $5

22 ca. 83 80. n8. :3. ”8. ms..- 032
$3 $3 :43 $3

32 mm. 8.3. 2:.- oom. NB: a... E.- 233
$09 83.3 38.9 50.3

.83- SE. 8?... .§.:- 2888
Exam H280 “Exam £80 Exam £80 Exam H280

odeED weakened-bum 320352 vetoed-bow “no—o; untamed-bow 5:39:30 3080

acuz 3:20 .5832: afifiauweum £38..— I nemmmohwom 83w?“ Saw—«>252 .3 finch.

momofiqouaa 5 33.8 @555 ”802

3o. 03. 8o. :5. am oxsaowaz

:o. 98. NS. So. ”a :25 a. use
:2: A83 :83 A83

8a. 89. 83 8o. 33 So. ”8. 8c.- wafiaauea Esau
A33 A83 A83 53

SN. m8.- mmm. 8a.- 6588
439m £80 8x5 ..coou 8E6 680 Sufi .eooo

OmD aun— Uotomomubom «co—01502 votOQoymtbom «Go—OT) vetoemufiom .aoaoawamoa imam-«O

 

Awwuz 35qu 2.253un uni—Echo; bun—ah I nova-Ewe“ 39%-wed Sam-:3:— .: 03.;

93

Eogunon Eozquov 955%: 0388 .3 E083 =88an moEa>
92 835g so. v as: :o. v a:

 

 

v.5“ fiwm :- 11% Numb hm Yw ma: mm 3.2 MEG
mic fiwm Mm 33.5w? fwn mm 5.5 mém mm ”EU—3&0 «no—£232
mic 5.3 “m 33.3 Wow hm SIR-AN odm mm ”Etna-GO «no—or»
N» B o\e 2 B X. 3952 Q 3 X. E E “\e guesaz N» 3 ..\e E E o\o x3532

 

3.0-.0 Eon—32H hubs-m

 

9.30 2.253.;- 223—

 

mac-.0 near—«Ebony

 

98.5 .3 2::- -_o>o zone—35.0: “Stamp—flow .3 «Sub

94

Hypothesis 4: Change in Family Bond and Delinquency Desistance

The final hypothesis tested in the current study examined the extent to which
changes in strength of family bond are associated with delinquency desistance. As
detailed in Chapter Three, it is hypothesized that youth who improve their relationship
with their primary caretaker will be more likely to desist offending during the
intervention. This is consistent with several theoretical perspectives that posit a negative
relationship between strength of the family bond and participation in delinquent activities
(see Chapter 2).

Tables 17-20 present the results of the bivariate and multivariate logistic
regression models for both the treatment and comparison groups. In almost all of the
models, improvement in family bond was associated with improvement in behavior. That
is, youth who reported an improved relationship with their primary caretaker were less
likely to self-report all forms of delinquency and drug use. The only exception to this
finding emerged in the bivariate official delinquency model with the treatment group
where a very small, statistically insignificant positive relationship emerged. Notably,
members of the treatment group who improved their bond to their primary caretaker were
significantly less likely to report involvement in nonviolent delinquency. Additional
analyses (not presented) employed ordinary least squares regression using a variety
measure for each of the self-reported delinquency outcomes (violent offending ranged O-
3, nonviolent offending ranged 0-8, and drug use ranged 0-4). Results were substantively
similar: improvement in family bond over time translates into involvement in fewer types

of delinquent behaviors over time.

95

While findings generally support Hypothesis 4, it is unclear what role the
intervention played in improving relationships between child and parent among members
of the treatment group. Indeed, as noted in Table 6 above, treatment and comparison
youth improved at about the same rate over time. Moreover, it is unclear whether the
improvement in family bond led to improvement in behavior or whether improvement
behavior was associated with less hassles and more positive interactions with parents. It
is not surprising that youth who report no involvement in delinquency are getting along
better with their parents compared to when they were involved in delinquency. Despite
these potential alternative interpretations, results suggest that efforts aimed at improving

family relationships may also result in the improvement of behavior over time.

96

c8. “co—“8.033 no. v n... ”885539 5 Echo Egg.“ ”8oz

 

 

 

 

 

 

 

mom. 2:. we. «2. ”a oxeaomaz

SN. 3. 2:. 3:. am :25 a. gco

$3 an: :2 3 $3 Samoa $96

36. So. SN. .3%. _- can. an- vs. «a .- 20m 55m a $56
:53 :2: 28¢ :53

:3 L S. 83 8o. 08. 3a.- 83 $8. semen a $5
A83 82¢ A83 $3

3.3 as. a: 3 _. SS 2”. 83 reg. ow<
53 $3 53 $5

82 SH. 82 EN. 83 an. a3. Sn- on:
$3 53 $3 . 88¢

w :2 own. 23 me. a. 30.. m3. Gm: 223
2.3.3 83d 23$ A83

and- 8m. 7 35- .837 3588
8x5 ago 25.5 .35 Ewan ago 8x5 €80

33 ac votomomfiom E32282 vetoemfiom 5205 entomomfiom 5559509 3050

Aweuz $59.0 «nofiuaoubunom wan—ah 5 03:50 I gang—wow— uwmmwea Saw—«>332 .3 03:9

8855.89 5 fiche 23:5» ”802

o8. m3. :6. 25. "m oxeaomaz

8o. 2:. So. . 8a. ”m 3% e u80

$3 653 53 A83 apnea 3.56

.3. may as. 39.7 2 m. N8; 23 «8. 28m base a amuse
5.3 A33 A33 A83

m3. SN. Sn. SS- 6888
Amvaxm ago @fim ..aooo @qu £80 8E6 £80

omDJEQ automomfiom E22282 categomfiom 8065 votoqomfiom mucosa—Eon 3050

 

Aweuz 3:20 3253»: 9:5 man—uh 5 magi—U I gig—wow— otawea 83.8.5— .5 035—.

97

 

Ono“ vo=8-ogv no. v a: ”$855.89 5 fiche Bags.“ ”802

 

94. OS. 2:. am excaomaz

m2. ”8. :8. am :05 a Uso

22.: 88¢ 93¢ 05:30

:8. Sam- 8N. on. _- ME. 5. 7 28m .235 a 0mg
so: 83¢ 82¢

82 :8. :2 SN. 82 an. um...
32.: $83 A23

32: $2 3. 8n- :3 2». on:
$3.: 33 63¢

82 a. 33 MS. 3: man. as?
893 836 85.3

“3:- 831 .85- 3388
85m $80 @qu 580 €qu 580

33:5 8:25.:an 253382 359.53 32> 35953

 

Amwuz 3:qu newt-"mace unem 58am 5 mug—6 I :emmmouMom 54%.: 85.52232 .3 03:.

momofinouaq 5 59:0 05:8.“ 6qu

 

 

£8. 98. So. am £55032

m3. Re. 35. NM :05 a. x00

82.: 5.3 33 25 55

£0. Sta. 5 m. on 3.. m3. mm}- 58m 358m 5 0936
82¢ 83¢ 85¢

Red. meA. 092. 58800
$96 580 @mxm ..coou €qu .580
33mg automomfiom “no—5282 entomomfiom E505 venomoyzfim

 

Ammuz 3:20 nemtaafieoudaem bun—am 5 wand—.0 I nemmmouwom 955N9— ouatazm .3 03:.

98

Additional Analyses: Hierarchical Linear Modeling

This final stage of analysis further examines the association between family
relationships and intervention amenability by employing Hierarchical Linear Modeling
(HLM). As discussed in Chapter Three, HLM has many advantages compared to
common regression approaches. In short, the method is particularly apprOpriate for the
current study because it allows the researcher to analyze individual change over time
within and across individuals — instead of simply changes in group means. Moreover,
HLM controls for the time-dependent nature of the data. For example, time 2
delinquency is dependent on time 1 delinquency - someone who is involved in a lot of
delinquent activities at time 1 is more likely to be involved in a lot of delinquentactivities
at time 2 than someone who has never engaged in delinquent activities. Finally, through a
Bayesian estimation process, HLM can estimate difference scores for respondents who
are missing time 2 data.

A two-level HLM model was estimated with the first level representing two time
points within individuals (pre and post; approximately 6 months between points) and the
second level representing individuals (see Figure 1). Between-individual measures (level
2) in the models include: group status (delinquent or comparison), race, age, and gender.
Within-individual measures (level 1) include strength of family bond and delinquency
over time (measured at time 1 and time 2). As discussed at length, it is hypothesized that,
as the strength of family bond increases over time, involvement in delinquent behavior
will decrease. In these analyses, a variety measure of delinquency (ranging from 0 to 9)
was employed representing a wide array of behaviors. Using this as the outcome variable

allows for a more precise look at changes in behavior over time. Additionally, changes in

99

strength of family bond over time will be assessed by analyzing the family bond measure
as a level-one covariate. While delinquency is the outcome variable of interest
throughout this study, it is also important to determine how strength of family
relationships changed over time —— either as a result of the intervention or not.

Difference scores were calculated for both outcome measures (delinquency and
strength of family bond) by entering them as level one (within-individual) variables. In
essence, this stage of the analysis estimates agreement and difference in measures at two
time points (pre and post) and attempts to predict agreement and discrepancy using the
salient independent variables outlined above. For example, it is hypothesized that
delinquent youth will report significantly different (fewer) time 2 delinquent activities if
they have a strong bond with their parents at time 1. Additionally, it is hypothesized that
differences (improvement) in family bond from time 1 to time 2 are associated with

differences in self-reported delinquency (improvement).

Figure l. HLM Two-Level Model

 

 

 

 

 

 

 

 

 

 

 

 

3

'3 Individual 1 Individual 2

8

6‘3

E

.4 Time 1 Time 2 Time 1 Time 2
E (pre) (Post) (pre) (post)
:2

 

 

 

 

 

 

 

 

 

 

 

 

100

the whole sample decreased the number of delinquent activities in which they were

involved. In addition, delinquent youth reported a more dramatic decrease. Specifically,
at time 2 the sample as a whole reported engaging in about 0.71 types of offenses (1.11 —
0.40). Delinquent youth reported participating in about 3.00 (3.52 - .52) offenses at time

2.

Table 22. HLM: Final Estimation of Fixed Effects — Family Bond Over Time
=217)

 

 

 

 

Coefficient Std. Error T
Intercept (BO)
Intercept (G00) 1.1 1 0.110 1008*"
Delinquent Group (G01) 2.41 0.260 926*"
Slope (time -— BI)
Intercept (G10) -0.40 0.101 -3.95***
Delinquent Group (G11) -0.52 0.270 -1.92

 

‘p < .05; “p < .01; mp < .001 (two-tailed test)

Table 23 presents the results of a basic HLM model testing the influence of
demographic variables in predicting delinquent behaviors (a variety score) over time.
Specifically, time and delinquency were entered at level 1 (within individual) and group
status (treatment or comparison), race (white), gender (male), and age were entered at
level 2 (between individual). At the first interview, delinquent youth reported more
delinquency than comparison youth, white youth reported more delinquency than
nonwhite youth, males reported more delinquency than females, and older youth reported
participation in more delinquent behaviors than younger respondents. Between the first
and second interview, the group as a whole decreased offending, while delinquent youth
reported an even more significant decrease (-0.58). White youth increased offending

over time (0.87 increase in the number of offenses compared to -0.59 for the group as a

102

whole). While statistically insignificant, males did not decrease offending as much as

females and older youth reported fewer delinquent activities than younger youth.

Table 23. HLM: Final Estimation of Fixed Effects — Demographic Controls and

Variety of Delinquent Activities (N=21D

 

 

 

 

Coefficient Std. Error T
Intercept (BO)
Intercept (G00) ' -2.96 1.217 -2.43*
Delinquent Group (G01) 2.22 0.263 845*"
White (G02) -0.62 0.230 -2.68**
Male (G03) 0.49 0.224 2.17*
Age (G04) 0.34 0.010 339*”
Slope (time - Bfi ,
Intercept (610) -0.59 1.115 -0.53
Delinquent Group (G11) -0.58 0.260 -2.23*
White (G12) 0.87 0.270 3.01 **
Male (G13) 0.09 0.268 0.33
Age (614) -0.02 0.090 -0.18

 

*p < .05; "p < .01; mp < .001 (two-tailed test)

Table 24 presents the results of the Hierarchical Linear Modeling analysis

predicting change in delinquency with group status (treatment or comparison) and

strength of family bond at interview I entered at level 2. As a whole, the sample engaged

in an average of 3.81 different delinquent acts. The delinquent youth engaged in an

average of 2.23 additional types of offenses than the group as a whole. Additionally, for

every unit increase in strength of family bond at interview 1, youth engaged in 0.76 fewer

types of offenses. Over time, youth engaged in 1.36 fewer types of offenses. Delinquent

youth engaged in even fewer types of offenses (-0.46). Finally, for each unit increase in

family bond at interview 1, youth engaged in fewer types of offenses, but this decrease

was not as dramatic as the sample as a whole (1.09 decrease over time [-1.36-.27]).

103

Because respondents who have a strong family bond start out committing fewer types of

offenses, it appears that they do not decrease as much as the others.

Table 24. HLM: Final Estimation of Fixed Effects — Family Bond at Time 1 and

Variety of Delinquent Activities (N=217)

 

 

 

 

Coefficient Std. Error T

IntercgpflBO)

Intercept (G00) 3.81 0.852 4.46***

Delinquent Group (G01) 2.23 0.236 9.45***

Mean Family Bond at Interview 1 (G02) -0.76 0.238 -3.21**
Slope (time — BI)

Intercept (G10) -1.36 0.958 -1.42

Delinquent Group (G11) -0.46 0.262 -1.76

Mean Family Bond at Interview 1 (G12) 0.27 0.264 1.04

 

"3p < 01; **‘p < 001 (two-tailed tCSt)

Figure 2 graphically depicts the findings from Table 24. Notably, delinquent

youth start out participating in a wider variety of offenses than the group as a whole, and

decrease more dramatically over time. Because they start out committing significantly

more types of delinquent activities, they still reported more types of offending at time 2

than the group as a whole. Youth who report a strong bond to their parent at time 1

(graphically presented as a one unit increase) start out committing fewer types of

offenses, and decrease in variety over time, but not as considerably as the sample as a

whole.

104

Figure 2. HLM: Discrepancy within Individuals over Time (Family Bond at Time 1)

7

 

 

 

 

 

 

 

 

 

 

fime1 fimez

 

[+Total Sample +Delinquent Group +Mean Family Bond Strong

 

Table 25 presents the result of an HLM model estimating the within-individual
relationship between change in family bond and delinquency over time by group status.
Specifically, time and mean family bond were entered as level-one predictors and group
status (delinquent or comparison) was entered as a level-two predictor of self-reported
delinquency (a level-one outcome variable). As expected, delinquent youth reported
participation in a greater variety of delinquent offenses at time 1 (2.88 more than the
group as a whole). In addition, over time delinquent youth reported a more dramatic
decrease in the number of delinquent activities participated in during the follow-up
period, however they still reported a wider variety of offenses than youth in the
comparison group. Finally, youth who reported increases in the strength of the family

bond over time were more likely to report fewer types of delinquent offenses over time.

105

Again, decreases were more dramatic (though not significantly different) for delinquent
youth. While these results are similar to what was found using logistic regression
models, the HLM more clearly demonstrates the differences between the delinquent and

comparison groups.

Table 25. HLM: Final Estimation of Fixed Effects — Family Bond Change over Time
and Variety of Delinquent Activities (N=Zlu

 

 

 

 

Coefficient Std. Error T

Intercept (BO)

Intercept (G00) 2.54 0.610 4.167***

Delinquent Group (G01) 2.88 0.993 2.897“
Slope {time — BZ)

Intercept (G20) -0.39 0.096 -4.1 1*"

Delinquent Group (G21) -0.56 0.264 -2. 14*
Slope (meanfamilv bond — BI)

Intercept (G10) -0.41 0.160 -2.57*

Delinquent Group (G11) _ -0.17 0.271 -0.62

 

“p < .01; *"p < .001 (two-tailed test)

Results from the HLM analysis largely replicate earlier findings but also suggest
that delinquent youth improved their behavior more dramatically than comparison youth.
While this may simply be a function of regression to the mean, it also may be indicative
of some success on the part of the intervention. That is, over time a greater proportion of
youth who were involved in the intervention desisted offending. Moreover, strength of
family bond was important in determining delinquent involvement at time 1 and time 2.
Specifically, youth who reported a stronger bond to their parent at time 1 reported
participation in fewer types of delinquent behavior at that time than youth who were not

strongly bonded to their parent. Also, improvement in family bond over time was

106

associated with improvement in behavior; that is, youth who improved their relationship

over time reported participation in fewer types of delinquent behaviors.

107

CHAPTER FIVE: CONCLUSION

Introduction

While findings from the current study are equivocal, they do provide some insight
into the effect of a strong family bond on successful delinquency desistance in a specific
intervention setting. First, a general inverse relationship emerged between strength of
family bond upon beginning the intervention and subsequent delinquency. That is, while
largely statistically insignificant, youth who began the intervention with a strong bond to
their primary caretaker were more likely to desist offending during the intervention. This
relationship also emerged, however, for the comparison group of youth who were not
involved in the intervention, suggesting the program may have added little to the dynamic
between family bond and delinquency.

Additionally, youth who reported improvement in their family bond over time
were also less likely to report involvement in delinquent behavior over time. Because
improvement in family relationships occurred concomitantly with improvement in
behavior, targeting family dynamics appears to be an effective strategy to reduce
involvement in delinquent behavior. Whether this relationship is direct or indirect,
findings from the current study indicate that focusing on strengthening family
relationships may result in decreased involvement in delinquency. In the current study,
however, the effect appeared to be more indirect; that is, the family programming
component did not appear to directly impact involvement in delinquency.

Results from the current study add to the sizeable literature that explores the

relationship between family structure and delinquency. Taken together, results are

108

largely unclear. There were no statistically significant results indicating two parents were
better than one; indeed, a higher proportion of treatment youth who lived in a two-parent
household were arrested within 6 months and reported involvement in violent and
nonviolent delinquency. Among the comparison group, results were mixed and also
statistically insignificant. Recent research has suggested that measuring family structure
as simply a one- or two-parent household may mask important differences that may exist
within different households (Demuth & Brown, 2004). For example, a male guardian
may be different than a female guardian depending on the gender of the child. Moreover,
households with a nonbiological second parent may not necessarily be better than single-
parent households. Again, the quality of the parent—child relationship appeared to be
more salient.

The current study also failed to uncover promising results for youth who received
family counseling. As discussed above, the ineffectiveness of the family counseling may
be attributed to several processes. First, the family counseling may not have been
comprehensive enough to improve family relationships among youth in the sample.
Lipsey and Wilson’s (1998) meta-analysis of effective interventions for serious offenders
revealed that family counseling was among the programs that were moderately effective.
While we know that youth and/or parents were counseled by trained facilitators with
MSW degrees, it is unclear the specific nature of the family counseling employed in the
current intervention. Future research evaluating the effectiveness of family programs
ought to determine exactly what methods are being used by counselors to better identify

the type of family counseling that is most effective.

109

Second, the counseling may take a long period of time to produce the desired
effects. Family problems are ofien long-standing and difficult to fix quickly. Since the
current study employed a short follow-up period (approximately 6 months), changes may
not yet be measurable. Additional follow-ups with this sample may better assess the
long-term effect of the family counseling, and future studies should employ multiple
follow-up periods (both short- and long-term if possible). There are several theoretical
and policy implications resulting fiom the current study. These as well its limitations

will be discussed in the subsequent sections.

Theoretical Implications

In general, results of the current study are supportive of the theoretical
perspectives that place importance on the relationship between parent and child. As
reviewed in Chapter Two, several theoretical formulations argue for the delinquency-
inhibiting properties of the parent-child bond. Youth in both the comparison and
treatment groups who reported stronger bonds to their parent were less likely to report
involvement in delinquent behaviors. This finding remained consistent both
contemporaneously and longitudinally. That is, the stronger the strength of family bond -
at interview 1, the less likely the youth reported involvement in delinquent behavior at
interview 1 and interview 2.

Results from the current study also provide evidence that emotional attachment
may be more salient than family structure. While the emotional bond measures were
related to delinquency in the theoretically expected direction in all of the models, family

structure was inconsistently and in some cases opposite of what was expected. Previous

110

research has suggested, however, that attachment and structure may be related. Sokol-
Katz and associates (1997) found that family structure was significantly related to family
attachment which had a significant and direct effect on minor and serious delinquency
and drug, alcohol, and tobacco use (Sokol-Katz et. a1, 1997). These researchers,
therefore, concluded that individuals who are firmly tied to a “broken” or reconstituted
(remarried) family are less likely to engage in deviant activities than individuals
unattached to intact families; level of attachment proved more salient than structure of
family (Sokol-Katz et. a1, 1997). On the other hand, Thomberry and colleagues (1999)
argue that transitions that occur as a result of changes in family structure are associated
with a variety of negative outcomes.

These findings are especially important because little empirical attention has been
devoted to testing theoretical tenets among samples of young children in general and
serious young offenders in particular. As expected, the family appears especially
important among this specialized population. Thomberry (1987:873) stresses the
importance of family during the preadolescent years: “During the early adolescent years,
the family is the most salient arena for social interaction and involvement and, because of
this, attachment to parents has a stronger influence on other aspects of the youth’s life at
this stage than it does at later stages of development.” Future research ought to continue
this trend by testing other criminological perspectives using samples of offenders at

various developmental stages.

111

Policy Implications

In the early 1970’s, many researchers concluded that “nothing works” to
rehabilitate offenders, particularly serious, violent, and/or chronic offenders. Gottfredson
and Hirschi (1990:268) summarize this position aptly: “Among academicians, the
conclusion that rehabilitation programs have been extensively tried and have been found
wanting is generally accepted.” As a result, many advocated for a more deterrence-based
approach focusing on incarceration programs to keep communities safe (Martinson,
1974; Whitehead & Lab, 1989; see also Gibbons, 1999 or Palmer, 1991 for a historical
review). Since then, however, a wealth of knowledge has been accumulated that suggests
many programs do work for certain offenders under certain circumstances and in certain
situations (Andrews et al., 1990; Palmer, 1991, 1995; Sherman et al., 1997; Wasserman,
Miller, & Cothern, 2000). Howell (1995: 158) summarizes this sentiment: “It can be said
with confidence that some programs do work when they are carefirlly conceived, properly
implemented, and provided with enough resources to do the job they set out to do.”

Moreover, after reviewing over 500 intervention evaluations, Lipsey (1999:564)
concluded that “. . .well designed rehabilitative strategies do reduce recidivism for [the
most serious offenders] and cannot be dismissed on the grounds that they are ineffective.”
Indeed, Martinson has even tempered his earlier argument that nothing works, suggesting
that some programs have demonstrated desired results (Martinson, 1979). Despite the

mounting body of supportive research and overwhelming public support for early

intervention (Cullen et al., 1998; Moon, Sundt, Cullen, & Wright, 2000), it has only been

112

relatively recently that interventions have been developed, implemented, and evaluated
which target early offenders.

Early offending is a paradox for justice professionals. Police officers do not know
how do handle young delinquents, resulting in many needy youth slipping through the
proverbial cracks. Minor deviance such as truancy, substance use, and fighting is _
ubiquitous during adolescence; however, these same behaviors exhibited earlier (at 6-11
years of age) are potent risk factors for continued involvement in delinquency. Police
officers and juvenile justice professionals must be prepared to address the causes of early
offending. In its current state, the juvenile justice system can offer little in the way of
services to this small but difficult population. Mental health officials are in a better
position to provide for these families, yet are often kept “out of the loop.” A partnership
must be forged between state and local mental health and juvenile justice officials in a
cooperative attack on child delinquency. Approaches that avoid the formal stigma of the
juvenile court are preferred, yet officials must ensure treatment plans are followed. The
evidence is convincing that failure to intervene with children who initiate offending at an
early age may result in a population of serious, violent, and chronic juvenile and adult
offenders.

In general, family interventions have proven more effective than non-family
interventions (Latimer, 2001; Perkins-Dock, 2001). For example, Kazdin (1987) argued
that family interventions hold the most promise for the treatment of Conduct Disorder in
children and adolescents. Moreover, parent-focused interventions implemented at home
or in other community locations tend to be very successful (Burns et al., 2003). The

specific program employed, however, will vary depending on the problems exhibited by

113

the youth (as evidenced in a comprehensive risk and needs assessment). For instance, a
youth with few family problems will require fewer resources devoted to the family (e.g.,
occasional family counseling) whereas youth experiencing crises require more intensive
family programming such as family preservation or even temporary out-of-home
placement (therapeutic foster care). In general, however, most families will benefit from
the basic services provided through Functional Family Therapy or multisystemic therapy
(Mihalic, Irwin, Elliott, Fagan, & Hanson, 2001).

The current study reveals that interventions that simply add-on family counseling
may not be as effective as more comprehensive family based programming. At least in
the short-term, youth who received family counseling were no less likely to report
involvement in delinquency than youth who did not participate in family counseling.
Program administrators acknowledged that rehabilitating family relationships was not the
primary focus of the intervention. As a result, findings presented here should not be
misinterpreted as evidence against the use of family-focused programming. Indeed, they
may be interpreted as a call for more comprehensive family oriented approaches.

Another strength of the current intervention was the commitment to keep youth in
their familial environments as much as possible. This was more easily facilitated by
employing a unique intensive supervision program that organized probation officers
around specific neighborhoods. This approach allowed probation officers to coordinate
supervision activities with community police officers and other neighborhood institutions
(e.g., school officials, neighborhood groups, churches). In the current program,
community police officers often joined probation officers on their rounds and pointed out

characteristics of the environment (such as potential resources) or of the particular family

114

(such as relatives known to police) that the probation officer may not have known. Other
jurisdictions ought to consider this approach as it appeared to aid in the overall
continuum of care offered to participants.

There are also several recommendations for firture studies stemming from lessons
learned during this project. First, it would be interesting to collect data from caretakers
regarding their perceptions of the relationship they have with their child. Furthermore,
future studies ought to consider questioning youth about any other potential influences in
their lives (secondary caretakers). A strong bond to a teacher or mentor may have the
same power as that with a parent. Second, it is important that program evaluators collect
specific data regarding programming. While this study was able to analyze‘the number
of minutes of family programming received, it was unable to clearly discern differences
that may have existed in terms of the quality of the family counseling in the current
program compared to others. Finally, it is imperative that future studies randomly assign
participants to a treatment or control condition such that any differences between the two

groups can be more clearly attributed to the treatment program.

Limitations

A primary concern with the current analyses is the size of the sample. While the
small number of serious juvenile offenders is good for society, it is problematic for
statistical analysis. Even though some statistically significant findings emerged, the
results must be interpreted with caution and, overall, generalizability is questioned due to

the small number of individuals in the study.

115

Another notable limitation of the current study is the lack of a control group of
offenders who did not participate in the intervention. Indeed, the ideal method would
have involved the random assignment of delinquent youth to either the program or a
control condition. Due to the relative scarcity of youth under the age of 14 who
participate in serious delinquency, there were not enough youth to assign to different
conditions. As a result, it is difficult to accurately ascertain the influence of the
intervention on the treatment youth with regard to the strength of their family bond.

Moreover, the comparison group employed was one of convenience, and may not
represent the attitudes, beliefs, and behaviors of “normally” developing preadolescents.
While efforts were made to recruit a representative sample of youth (by supplementing
youth from a middle school with youth from two recreation centers in the target area),
sample selection bias is a potential problem. Because active consent was required for
participation in the study, youth interested in participating were required to obtain
permission via a signed form from a parent. Biases may have been introduced as a
function of those parents willing to consent and return the signed form to researchers on
time.

As noted, there are substantial differences between the comparison and treatment
groups that limit the ability of the findings to definitively answer the research question
posed. Specifically, the two groups were not equal in terms of their family relationships
at the first interview. It is interesting to note, however, that even though the comparison
youth started out with significantly stronger bonds to their primary caretaker, they

changed at about the same rate as the delinquent youth. ,

116

Sample attrition is another limitation of the current study. As presented in
Chapter Three, there were significant differences between youth who were successfully
re-interviewed and those who were not. This was especially striking for the comparison
group. While every effort was made to interview all treatment and comparison youth at
two time periods, many families moved out of the area or were otherwise unreachable.
As a result, selection-mortality may threaten internal validity of the data analyzed in the
current study.

Another potential limitation of the study is the reliance on self-reported data.
While official delinquency was used as an outcome measure for the treatment youth, self-
reported measures of family bond and delinquency were the primary variables employed
in the study. With regard to strength of the family bond, youth may have different
perceptions about this relationship than parents. It would be interesting and important to
ask parents the same questions about their children to determine the extent of agreement.

Social scientists recognize that self-report measures of delinquency have notable
sources of error (Maxfield et al., 2000). For example, some respondents may be more or
less likely to report offending to researchers in certain circumstances (experimenter
effect). In addition, respondents who are surveyed multiple times may be influenced by
previous interviews (panel effect). It has also been argued that self-reports do not
measure serious delinquency, are not standardized, and are invalid (Hindelang, Hirschi,
& Weis, 1981). While these latter concerns are arguably less applicable to recent
surveys, questions still exist as to the extent to which youth will honestly report deviant

(especially serious) offenses.

117

Self-report data seem to better measure minor deviances such as smoking,
truancy, and marijuana use because these offenses are often not captured in official
records (Elliott & Ageton, 1980). Similarly, a respondent may be less likely to report a
very serious offense (sexual assault, robbery) to a survey researcher but there is more
likelihood these offenses will emerge in official data — especially if the youth is a repeat
or chronic offender (Elliott & Ageton, 1980). Because of the strengths and limitations
associated with self-report and official sources of data, both were used independently as
outcome measures in an effort to broaden generalizability.

The face-to- face interviewing approach employed in the current study has several
strengths. Indeed, this approach is among the most highly regarded in social science
research (Singleton & Straits, 1999). First, interviewers were able to clarify questions
that respondents were unclear about. This proved especially important considering the
target population — youth between the ages of 10 and 13. Even though the research staff
attempted to develop an interview schedule that was easily understandable and based on
previously used questions, some of the younger respondents had difficulty understanding
concepts such as “criticize” or even “unlikely” and still others did not know what
“ecstasy” or “a rave” was. Trained interviewers were able to explain these concepts to -
youth in an effort to elicit the most accurate response to any given question.

Second, sensitive or uncomfortable questions cannot be overlooked by
respondents as easily when interviewed in person as compared to self-administered
questionnaires. Even though respondents were not pressured into answering questions

they were uncomfortable about, very few missing values were present in these data.

118

One problem associated with face-to-face interviews, however, is the potential for
bias to be introduced (either explicitly or implicitly) by the individual interviewer. For
example, female youth may respond differently to male interviewers than female
interviewers. Respondents may answer questions differently depending on the race of the
interviewer compared to that of their own. Finally, different interviewers may phrase
questions a bit differently or inadvertently look intently at respondents when asking them
about delinquent activities. All of these issues are potentially damaging to any social
scientific effort using face-to-face interviews, however, good training of interviewers can
limit their negative effect. The current study attempted to limit interviewer bias by
training each interviewer, systematizing instructions to by read by interviewers to
respondents, and, when possible, having female researchers interview female youth and
male researchers interview male youth.

Another concern that has been leveled by researchers regarding self-report studies
is the extent to which youth may exaggerate, forget, or lie about their activities. Of
particular concern with face-to-face interviews is the extent to which respondents will be
completely honest with their responses. This concern is especially appropriate given the
types of questions asked in surveys of delinquent behavior. As with all research using
this methodology, respondents were reassured several times throughout the interview
session that their individual responses would be protected to the maximum extent
allowable by law.

As a final measure to assess the truthfirlness of respondents, at the end of the
study they were asked “While you were answering the interviewer’s questions, how

truthful were your responses?” After the first interview, 81% of youth answered “very

119

truthful” and an additional 18% responded “somewhat truthful”. Concluding the second
interview, 85% of youth responded “very truthful” while an additional 14% responded
“somewhat truthful”. While this is a rudimentary assessment of the truthfulness of

respondents, it does provide an additional level of confidence in the answers obtained.

Conclusion

Results of this study are important for a number of reasons. First, findings
presented support theoretical models that argue for the delinquency inhibiting properties
of a strong bond to parents. In almost all models for both the delinquent and comparison
youth, respondents who reported a stronger relationship with their parent were less likely
to report involvement in delinquency. Second, family structure does not appear to be an
important predictor of delinquency desistance for most youth. Program administrators
cannot simply assume that youth who reside with two parents are going to demand fewer
services than youth residing with only one parent. Third, the family programming
employed in the current intervention appeared, at least in the immediate short-term,
ineffective at reducing involvement in delinquent behaviors. Future interventions must
incorporate more rigorous forms of family-based programming such as Functional
Family Therapy or Parent Management training and employ a longer follow-up period to
determine long-term outcomes of the treatment.

Program administrators must continue to employ innovative treatment approaches
that are theoretically grounded and empirically supported. Moreover, comprehensive
process and outcome evaluations must be associated with these efforts so that results of

promising and proven programs can be propagated and programs replicated in various

120

settings with diverse populations of offenders. While results from the current study were
mixed, they help to advance the field incrementally by identifying what aspects of
programming are effective (and ineffective) at improving the behavior of serious, violent,
juvenile offenders.

Despite the wealth of promising evidence in favor of conununity-based programs,
little has changed in the juvenile justice system. Most jurisdictions still largely rely on
incapacitation and surveillance approaches to manage serious, violent, and chronic
offenders. In many cases, the problem is not that multisystemic therapy or similar
strategies are too costly. Indeed, most community-based interventions are much less
expensive than detention or other out-of-home placement options. What is needed, in
short, is a paradigm shift. Legislatures must fund treatment-based alternatives to
incarceration; practitioners must implement empirically-supported and theoretically
meaningful programs; and researchers must rigorously evaluate existing and innovative

interventions. The current study represents a step in this direction.

121

APPENDICES

122

APPENDIX A: Survey Instrument

MICHIGAN JUVENILE INTERVENTION INITIATIVE

 

 

 

 

ASSESSMENT

REVISED VERSION (06-06-02)
FIRST INTERVIEW
Youth Last Name MSU ID#
Youth First Name
Interview Date
Interview Number 1 2 3
Site FL GR LAN SAG
Interviewer

 

Sample Youth D

Control Youth El

ENTRY

DATE

123

Directions to be read to juvenile. I am from Michigan State University, and am not in any way
working for the police, the schools, or any other agency in the city. We are looking to gather
information about youth to better understand why they sometimes get into trouble. Your
individual responses will be protected to the maximum extent allowable by law

I am going to read you some statements about your attitudes towards certain behaviors, school,
and your beliefs about events in your life. Think about each statement I make and then tell me if
you Strongly Agree with the statement, Agree with the statement, Disagree with the statement, or
Strongly Disagree with the statement. If you are not sure, give the response that is closest to how
you feel about the statement I make.

(Provide response card # I to the juvenile).
(ilwai)

1. I’m not very sure of myself.
Cl Strongly Agree C] Agree 0 Disagree :1 Strongly Disagree

2. I really don’t like myself very much.
0 Strongly Agree 0 Agree 0 Disagree :1 Strongly Disagree

3. I sometimes feel so bad about myself that I wish I were somebody else.
13 Strongly Agree 1:) Agree :1 Disagree 0 Strongly Disagree

4. I usually feel I’m the kind of person I want to be.
[:1 Strongly Agree I: Agree 13 Disagree :1 Strongly Disagree

5. I feel I can do things as well as other people can.
[:1 Strongly Agree [:1 Agree 1:1 Disagree D Strongly Disagree

6. I feel that I am a special or important person.
[:1 Strongly Agree 1:1 Agree :1 Disagree :1 Strongly Disagree

7. I feel that I am really good at things I try to do.
0 Strongly Agree [3 Agree :1 Disagree 1:1 Strongly Disagree

(ilipfi)

1. Other people decide what happens to me.
:1 Strongly Agree [3 Agree [:1 Disagree [:1 Strongly Disagree

2. It is important to think before you act.
0 Strongly Agree :1 Agree [3 Disagree 1:] Strongly Disagree

3. If I study, Iwill get better grades. .
1:1 Strongly Agree :1 Agree 0 Disagree :1 Strongly Disagree

4. When I try to be nice, people notice.
1:1 Strongly Agree D Agree 0 Disagree (:1 Strongly Disagree

124

M

. Ifyou work hard, you will get what you want.
0 Strongly Agree Cl Agree 1:1 Disagree 1:1 Strongly Disagree

Ofl

. To make a good decision, it is important to think.
:1 Strongly Agree :1 Agree 1:1 Disagree 0 Strongly Disagree

7. I am responsible for what happens to me.
1:1 Strongly Agree :1 Agree 1:1 Disagree 1:1 Strongly Disagree

8. Sometimes you have to physically fight to get what you want.
1:1 Strongly Agree :1 Agree :1 Disagree :1 Strongly Disagree

\0

. Iget mad easy.
0 Strongly Agree 1: Agree Cl Disagree 1:1 Strongly Disagree

10. I do whatever I feel like doing.
13 Snongly Agree 1:1 Agree 13 Disagree 1:1 Strongly Disagree

11. When I get mad, Iyell at people.
c1 Strongly Agree c1 Agree 1:1 Disagree D Strongly Disagree

12. Sometimes I break things on purpose.
1: Strongly Agree 1: Agree [:1 Disagree 1:1 Strongly Disagree

13. If I feel like it, Ihit people.
1:1 Strongly Agree 1:1 Agree :1 Disagree 1:1 Strongly Disagree

(ilbsa)

1. It makes you feel big and tough when you push someone around.
:1 Strongly Agree :1 Agree :1 Disagree 1:1 Strongly Disagree

2. If you back down from a fight, everyone will think you are a coward.
c1 Strongly Agree 0 Agree :1 Disagree :1 Strongly Disagree

3. Sometimes you have only two choices — get punched or punch the other kid first.
1:1 Strongly Agree 1:1 Agree 1:1 Disagree 1: Strongly Disagree

4. It’s OK to hit someone if you just go crazy with anger.
c1 Strongly Agree :1 Agree 1:1 Disagree 1:1 Strongly Disagree

5. A guy who doesn’t fight back when other kids push him around will lose respect.
:1 Strongly Agree :1 Agree :1 Disagree 1: Strongly Disagree

6. A guy shows he really loves his girlfriend if he gets in fights with other guys about her.
1:1 Strongly Agree :1 Agree :1 Disagree 1: Strongly Disagree

125

 

(ilatv)

1. IfI walk away from a fight, I’d be a coward (“chicken”).
D Strongly Agree 1:1 Agree D Disagree c1 Strongly Disagree

2. I don’t need to fight because there are other ways to deal with being mad.
1:1 Strongly Agree 1:1 Agree 1:1 Disagree 1:1 Strongly Disagree

3. It’s ok to hit someone who hits you first.
1:1 Strongly Agree 1: Agree 1:1 Disagree 1:1 Strongly Disagree

4. If a kid teases me, I usually cannot get him/her to stop unless I hit him/her.
1:1 Strongly Agree 1:1 Agree :1 Disagree 1:1 Strongly Disagree

5. IfI refuse to fight, my friends will think I’m afraid.
c1 Strongly Agree :1 Agree c1 Disagree 1:1 Strongly Disagree

(ilatg)

1. I think you are safer, and have protection, if you join a gang.
0 Strongly Agree 1:1 Agree 1:1 Disagree 1:1 Strongly Disagree

2. Iwill probably join a gang.
1:1 Strongly Agree :1 Agree 1:1 Disagree c1 Strongly Disagree

3. Some of my friends at school belong to gangs.
1:1 Strongly Agree c1 Agree 1:1 Disagree 1:1 Strongly Disagree

4. My fiiends would think less of me if I joined a gang.
1:1 Strongly Agree 0 Agree 13 Disagree 1:1 Strongly Disagree

5. I believe it is dangerous to join a gang; you will probably end up getting hurt or killed if you
belong to a gang.
1:1 Strongly Agree :1 Agree c1 Disagree 1:1 Strongly Disagree

6. I think being in a gang makes it more likely that you will get into trouble.
1:1 Strongly Agree 1: Agree 1:1 Disagree 1:1 Strongly Disagree

7. Some people in my family belong to a gang, or used to belong to a gang.
1:1 Strongly Agree 1: Agree 1:1 Disagree 1:1 Strongly Disagree

8. Ibelong toagang.
13 Strongly Agree 1:1 Agree 0 Disagree 1: Strongly Disagree

(ilats)

1. Homework is a waste of time.
1:1 Strongly Agree 1:1 Agree 1:1 Disagree 1:1 Strongly Disagree

126

2. Itryhardinschool.
1:1 Strongly Agree C] Agree :1 Disagree C] Strongly Disagree

3. Education is so important that it’s worth it to put up with things about school that I don’t like.
1:1 Strongly Agree 1:1 Agree 1:1 Disagree 1:1 Strongly Disagree

4. In general, I like school.
1:1 Strongly Agree 1:1 Agree 13 Disagree C] Strongly Disagree

5. I don’t care what teachers think of me.
D Strongly Agree c1 Agree 0 Disagree 1:1 Strongly Disagree

6. I expect to graduate from High School.
1:1 Strongly Agree :1 Agree :1 Disagree 1:1 Strongly Disagree

7. Iexpect to go to college.
13 Strongly Agree 1:1 Agree 13 Disagree 1: Strongly Disagree

8. I expect that my education will influence my future success.
1:1 Strongly Agree 1: Agree 1:1 Disagree 1:1 Strongly Disagree

Directions to be read to juvenile. I have a few more questions about school. I am going to ask
you about your involvement in activities that take place in school. Think about this year and tell
me if you Always, Sometimes, Once in a While, or Never are involved in these school activities.

(Provide response card # 2 to the juvenile).

9. During the school year I do my homework.
13 Never 1:1 Once in a While 1:1 Sometimes 1:1 Always

10. Dming the school year I participate in school sports besides gym.
1:1 Never 1: Once in a While 1:1 Sometimes 1:1 Always

11. During the school year I belong to clubs or organizations in the school.
13 Never 13 Once in a While 13 Sometimes 1:1 Always

12. During the school year I skip classes
:1 Never c1 Once in a While [3 Sometimes 1: Always

13. During the school year I get help with my homework from a teacher or tutor
1: Never 1:1 Once in a While 1:1 Sometimes c1 Always

(ilpaid)

Directions to be read to juvenile. I am going to read you some statements about what your
closest friends think about some behaviors. Think about all the people you hung around with in
the past year, including your two or three closest friends. Tell me how yom' friends would
respond to some situations in terms of Strongly Approving, Approving, Disapproving, or
Strongly Disapproving of the behavior.

(Provide response card # 3 to the juvenile).

127

How would your friends respond if they knew you...

1. Used a weapon or force to get money or things from people?
C] Strongly Approve 1:1 Approve 1:1 Disapprove 13 Strongly Disapprove

2. Hit someone with the idea of hurting them, for example, fist fighting?
Cl Strongly Approve 1:1 Approve 1:1 Disapprove 1:1 Strongly Disapprove

3. Stole something worth $50? ‘
13 Strongly Approve El Approve a Disapprove 1:1 Strongly Disapprove

4. Damaged or destroyed someone else’s property on purpose?
1: Strongly Approve 0 Approve 13 Disapprove 1:1 Strongly Disapprove

5. Took a car or motorcycle for a ride without the owner’s permission?
:1 Strongly Approve 0 Approve c1 Disapprove 1:1 Strongly Disapprove

6. Used marijuana, reefer, or pet?
1:1 Strongly Approve :1 Approve 1:1 Disapprove 1:1 Strongly Disapprove

7. Used crack?

1:1 Strongly Approve :1 Approve 1:1 Disapprove 1:1 Strongly Disapprove
8. Used heroin?

1:1 Strongly Approve :1 Approve 13 Disapprove U Strongly Disapprove
9. Used cocaine?

1:1 Strongly Approve :1 Approve 1:1 Disapprove El Strongly Disapprove
10. Used LSD or acid?

c1 Strongly Approve 1:1 Approve c1 Disapprove 13 Snongly Disapprove
11. Carried a gm for protection?

13 Strongly Approve 0 Approve c1 Disapprove c1 Strongly Disapprove

Directions to be read to juvenile. Now, you have thought about people you have hung out with
in the past year including your two or three closest friends. I would like to ask you some
questions about them during the last year. I am going to read you some activities that kids are
sometimes involved in and, think about these friends, then tell me whether None of Them, A Few
of them, Some of Them, or All of Them were involved in these activities.

(Provide response card # 4 to the juvenile).

In the last year how many of your friends...

12. Drank beer, wine or liquor?
13 None of Them 1:1 A Few of Them :1 Some of Them C] All of Them

128

 

13.
14.
15.
16..
17.
18.

19.

Used a weapon or force to get money or things from people?

C1 None of Them C1 A Few of Them C1 Some of Them C1 All of Them

Attacked someone with a weapon or with the idea of seriously hurting them?
C1 None of Them C1 A Few of Them C1 Some of Them 1:1 All of Them

Hit someone with the idea of hurting them, for example, fist fighting?
1:1 None of Them C1 A Few of Them 1:1 Some of Them 1:1 All of Them

Stole something worth more than $100?

13 None of Them C1 A Few of Them C1 Some of Them C1 All of Them

Stole something worth more than $5 but less than $50?

1:1 None of Them 13 A Few of Them C1 Some of Them C1 All of Them

Damaged or destroyed someone else’s property on pmpose?

13 None of Them C1 A Few of Them C1 Some of Them 0 All of Them

Took a car for a ride without the owner’s permission?

1:1 None of Them 13 A Few of Them 13 Some of Them C1 All of Them

(ilecv)

Directions to be read to juvenile. I am going to ask you about things that happened in your
neighborhood during the last year. Think about the place(s) that you have lived in the last year
and tell me if these things happened "Never, Once or Twice, A Few Times, or Many Times.

(Provide response card # 5 to the juvenile).

In my neighborhood...

1.

I have heard guns being shot.
0 Never 1:1 Once or Twice C1 A Few Times

. I have seen somebody arrested.

C1 Never C1 Once or Twice C1 A Few Times

. I have seen drug deals.

1: Never 1:1 Once or Twice 1:1 A Few Times

I have seen someone being beat up.
1:1 Never C1 Once or Twice C1 A Few Times

. My house has been broken into.

1:1 Never 1:1 Once or Twice C1 A Few Times

. I have seen somebody get stabbed or shot.

It] Never C1 Once or Twice 1:1 A Few Times

C1 Many Times

U Many Times

13 Many Times

Many Times

C1

C1 Many Times

U Many Times

129

7. Ihave seen gangs in my neighborhood.
1:1 Never 1:1 Once or Twice 1:1 A Few Times 13 Many Times

8. I have seen somebody pull a gun on another person.
C1 Never C1 Once or Twice C1 A Few Times 0 Many Times

9. I have seen someone in my home get shot or stabbed.
C1 Never 1:1 Once or Twice 1:1 A Few Times 13 Many Times

(ilfas)

Directions to be read to juvenile I am going to ask you some questions about your family. The
first few involve some questions about who you live with, then I will ask you some questions
about the activities of the parent(s) or guardian that you live with.

 

1. Who did you live with for the majority of the last calendar year? (check appropriate box)

C1 Mother only [ask questions in series A and D only, skip series B and C]

1:1 Father only [ask questions in series B and D only, skip series A and C]

C1 Both Mother and Father (includes step-parents) [ask questions in series A, B, and D then
skip C]

C1 Neither Mother or Father [ask questions in series C and D only, A and B do not apply]

2. Is the person most responsible for you, that is the person considered yom' primary caretaker or
guardian during the last 12 months, male or female?
C1 FEMALE C1 MALE

Directions for interviewer afler juvenile responds to who they lived with.

Below are 4 series of questions, I for Mother (SERIES A), I for Father (SERIES B), I for
Guardian, or the person most responsible for the juvenile ’s care in the last year (SERIES C) and
1 for how often the youth is involved in family centered activities (SERIES D). Series D should be
asked of ALL juveniles, but the other series are conditional based on who the juvenile has lived
with for the last year. If the youth identifies himself/herself as living with both mother and
father, two series of questions (A and B) should be asked in addition to series D. If the youth
identifies only mother or only the father as the person that he or she lived with, then the series for
that parent should be filled out as well as series D. If the youth lived with neither parent in any
combination for the majority of the last year then only series C, which asks questions about their
guardian or caretaker, should be filled out along with Series D.

SERIES A [MOTHER SPECIFIC QUESTIONS) (ilfasm)

1. Is your mother, your biological mother?
13 YES C1 NO

Directions to be read to juvenile. I am going to read you some questions about your Mother.
First, I want to ask you about how you relate to your Mother. I am going to read some statements
and you tell me if they are Always, Sometimes, Once in a While, or Never true in terms of how
you relate to your mother.

(Provide response card # 2 to the juvenile).

130

2. You get along well with your mother. . .
C1 Never 1:1 Once in a While C1 Sometimes C1 Always

3. You feel you can really trust your mother...
C1 Never C1 Once in a While C1 Sometimes C1 Always

4. Your mother does not understand you...

C1 Never C1 Once in a While C1 Sometimes 1:1 Always
5. Your mother is too demanding...

C1 Never 1:1 Once in a While C1 Sometimes C1 Always

6. You really enjoy your mother...
13 Never 13 Once in a While 13 Sometimes C1 Always

7. You have a lot of respect for your mother...
C1 Never C1 Once in a While C1 Sometimes 1:1 Always

8. Your mother interferes with your activities...
C1 Never 1:1 Once in a While C1 Sometimes 1:1 Always

\O

. Youthinkyour mother is terrific...
C1 Never C1 Once ina While 1:1 Sometimes 1:1 Always

10. You feel very angry toward your mother...
C1 Never C1 Once in a While 13 Sometimes C1 Always

11. You feel violent towards your mother...
0 Never C1 Once in a While 1:1 Sometimes 1:1 Always

12. You feel proud of yom' mother...
C1 Never C1 Once in a While 1:1 Sometimes 1:1 Always

Directions to be read to the juvenile. Now I am going to ask you some more questions about
how likely you would be to approach your mother about certain things. I will ask you a question
and you tell me if you would be Very Likely, Likely, Unlikely, or Very Unlikely to take the
action that I am talking about.

(Provide response card # 6 to the juvenile).

If you needed to, how likely would you he to...

13. Talk to your mother about personal or private things? Would you say...
13 Very Likely C1 Likely C1 Unlikely 1:1 Very Unlikely

14. Ask your mother for advice when you need to make an important decision?
1:1 Very Likely C1 Likely C1 Unlikely C1 Very Unlikely

15. Borrow money from your mother?
C1 Very Likely C1 Likely 1:1 Unlikely C1 Very Unlikely

131

16.

17.

18.

19.

20.

21.

Talk to your mother about problems you are having with your family?
1:1 Very Likely C1 Likely 13 Unlikely C1 Very Unlikely

Talk to your mother about problems you are having with a friend?
C1 Very Likely C1 Likely C1 Unlikely C1 Very Unlikely

Feel you can trust your mother with just about anything you tell her?
13 Very Likely C1 Likely 1:1 Unlikely 1:1 Very Unlikely

Talk to your mother about trouble you are having at work or school?
C1 Very Likely C1 Likely C1 Unlikely C1 Very Unlikely

Go to your mother for help in an emergency?
C1 Very Likely C1 Likely C1 Unlikely 13 Very Unlikely

Go places or do things with your mother?
C1 Very Likely C1 Likely Cl Unlikely C1 Very Unlikely

Directions to be read to the juvenile. Now I am going to ask you some more questions about
what your mother knows about your activities. I will ask you a question and you tell me if she
Knows Nothing, Knows Just A Little, Knows Some Things, or Knows Everything about the
things that I describe.

(Provide response card # 7 to the juvenile).

22.

23.

24.

25.

26.

27.

How much does your mother know about your close friends, that is who they are?
13 Knows Nothing C1 Knows Very Little C1 Knows Something 13 Knows Everything

How much does your mother know about what you do with your friends?
C1 Knows Nothing 1:1 Knows Very Little 1:1 Knows Something 13 Knows Everything

How much does your mother know about your close friends’ parents, that is, who they are?
13 Knows Nothing 1:1 Knows Very Little 1:1 Knows Something C1 Knows Everything

How much does your mother know about who you are with when you are not at home?
1:1 Knows Nothing 1:1 Knows Very Little 1:1 Knows Something 1:1 Knows Everything

How much does your mother know about who your teachers are?
C1 Knows Nothing 1:1 Knows Very Little 1:1 Knows Something D Knows Everything

How much does your mother know about what you are doing in school?
1:1 Knows Nothing C1 Knows Very Little C1 Knows Something C1 Knows Everything

Directions to be read to juvenile. I am going to read a couple of statements about your mother.
Tell me if you Strongly Disagree, Disagree, Agree, or Strongly Agree with these statements.

(Provide response card # I to the juvenile).

28.

I think highly of my mother.
C1 Strongly Agree 1: Agree C1 Disagree 1:1 Strongly Disagree

132

29. My mother is a person I want to be like.
13 Strongly Agree 13 Agree C1 Disagree C1 Strongly Disagree

30. I really enjoy spending time with my mother.
13 Strongly Agree C1 Agree C1 Disagree 13 Strongly Disagree

31. My mother praises me for doing well.
13 Strongly Agree 13 Agree C1 Disagree 13 Strongly Disagree

32. My mother criticizes me and my ideas.
C1 Strongly Agree 13 Agree 13 Disagree 13 Strongly Disagree

33. My mother helps me do things that are important to me.
C1 Strongly Agree 13 Agree C1 Disagree C1 Strongly Disagree

34. My mother blames me for her problems.
C1 Strongly Agree 13 Agree 13 Disagree 13 Strongly Disagree

35. My mother makes plans with me and cancels them for no good reason.
13 Strongly Agree C1 Agree 13 Disagree C1 Strongly Disagree

SERIES B [FATHER SPECIFIC QUESTIONS) (ilfasd)

1. Is your father, your biological father?
13 YES 13 NO

Directions to be read to juvenile. I am going to read you some questions about your Father.
First, I want to ask you about how you relate to your Father. I am going to read some statements
and you tell me if they are Always, Sometimes, Once in a While, or Never true in terms of how
you relate to your father.

(Provide response card # 2 to the juvenile).

2. You get along well with your father. . .

13 Never 13 Once in a While 13 Sometimes Always

Cl

3. You feel you can really trust your father...
C1 Never 13 Once in a While 13 Sometimes C1 Always

4. Your father does not understand you. . .
C1 Never 13 Once in a While C1 Sometimes C1 Always

5. Your father is too demanding...
13 Never 13 Once in a While C1 Sometimes C1 Always

6. You really enjoy your father. ..
- 13 Never 13 Once in a While 13 Sometimes

Cl

Always

133

 

7. You have a lot of respect for your father...

13 Never 13 Once in a While C1 Sometimes Always

C]

8. Your father interferes with your activities...

13 Never C1 Once in a While 13 Sometimes Always

D

9. You think your father is terrific. ..

C1 Never C1 Once in a While C1 Sometimes Always

D

10. You feel very angry toward your father...
C1 Never C1 Once in a While C1 Sometimes

D

Always

11. You feel violent towards your father...
C1 Never 13 Once in a While C1 Sometimes 13 Always

12. You feel proud of your father...
13 Never C1 Once in a While C1 Sometimes 13 Always

 

Directions to be read to the juvenile. Now I am going to ask you some more questions about
how likely you would be to approach your father about certain things. I will ask you a question
and you tell me if you would be Very Likely, Likely, Unlikely, or Very Unlikely to take the
action that I am talking about.

(Provide response card # 6 to the juvenile).

If you needed to, how likely would you be to...

13. Talk to your father about personal or private things? would you say...
C1 Very Likely 13 Likely C1 Unlikely 13 Very Unlikely

14. Ask your father for advice when you need to make an important decision?
C1 Very Likely C1 Likely C1 Unlikely 13 Very Unlikely

15. Borrow money from your father?
C1 Very Likely C1 Likely C1 Unlikely 13 Very Unlikely

16. Talk to your father about problems you are having with your family?
13 Very Likely 13 Likely C1 Unlikely C1 Very Unlikely

17. Talk to your father about problems you are having with a friend?
13 Very Likely C1 Likely C1 Unlikely 13 Very Unlikely

18. Feel you can trust your father with just about anything you tell him?
[J Very Likely C1 Likely 13 Unlikely C1 Very Unlikely

19. Talk to your father about trouble you are having at work or school?
13 Very Likely C1 Likely 13 Unlikely C1 Very Unlikely

20. Go to your father for help in an emergency?
C1 Very Likely 13 Likely C1 Unlikely C1 Very Unlikely

134

21.

Go places or do things with your father?
13 Very Likely 13 Likely C1 Unlikely C1 Very Unlikely

Directions to be read to the juvenile. Now I am going to ask you some more questions about
what your father knows about your activities. I will ask you a question and you tell me if he
Knows Nothing, Knows Just A Little, Knows Some Things, or Knows Everything about the
things that I describe.

(Provide response card # 7 to the juvenile).

22.

23.

24.

25.

26.

27.

How much does your father know about your close fiiends, that is who they are?
13 Knows Nothing C1 Knows Very Little C1 Knows Something C1 Knows Everything

How much does your father know about what you do with your friends?
13 Knows Nothing C1 Knows Very Little 13 Knows Something C1 Knows Everything

How much does your father know about your close fiiends’ parents, that is, who they are?
13 Knows Nothing C1 Knows Very Little C1 Knows Something 13 Knows Everything

How much does your father know about who you are with when you are not at home?
C1 Knows Nothing C1 Knows Very Little C1 Knows Something C1 Knows Everything

How much does your father know about who your teachers are?
13 Knows Nothing 13 Knows Very Little C1 Knows Something 13 Knows Everything

How much does your father know about what you are doing in school?
C1 Knows Nothing C1 Knows Very Little 13 Knows Something 13 Knows Everything

Directions to be read to juvenile. I am going to read a couple of statements about your father.
Tell me if you Strongly Disagree, Disagree, Agree, or Strongly Agree with these statements.

(Provide response card # 1 to the juvenile).

28.

29.

30.

31

32.

33

I think highly of my father.
13 Strongly Agree C1 Agree 13 Disagree C1 Strongly Disagree

My father is a person I want to be like.
13 Strongly Agree 13 Agree C1 Disagree C1 Strongly Disagree

I really enjoy spending time with my father.
13 Strongly Agree C1 Agree 13 Disagree 13 Strongly Disagree

My father praises me for doing well.
C1 Strongly Agree C1 Agree 13 Disagree C1 Strongly Disagree

My father criticizes me and my ideas.
'13 Strongly Agree 13 Agree C1 Disagree C1 Strongly Disagree

My father helps me do things that are important to me.

135

 

13 Strongly Agree 13 Agree C1 Disagree C1 Strongly Disagree

34. My father blames me for his problems.
C1 Strongly Agree 13 Agree C1 Disagree C3 Strongly Disagree

35. My father makes plans with me and cancels them for no good reason.
13 Strongly Agree 13 Agree C1 Disagree C1 Strongly Disagree

SERIES C [Guardian SPECIFIC QUESTIONS) (ilfasg)

1. Is the person most responsible for you, that is the person considered your primary caretaker or
guardian during the last year, male or female?
13 FEMALE C1 MALE

2. What is that person’s first name?
2a. What is this person’s relationship to you?

 

 

Substitute that person’s name for the blanks in the questions below

Directions to be read to juvenile. I am going to read you some questions about

First, I want to ask you about how you relate to .I am going to read some statements
and you tell me if they are Always, Sometimes, Once in a While, or Never true in terms of how
you relate to

(Provide response card # 2 to the juvenile).

3. You get along well with
13 Never C1 Once in a While 13 Sometimes 13 Always

4. You feel you can really trust
C1 Never C1 Once in a While C1 Sometimes 13 Always

5. does not understand you...

C1 Never 13 Once in a While 13 Sometimes C1 Always
6. is too demanding...

C1 Never 13 Once in a While 13 Sometimes 13 Always
7. You really enjoy .

13 Never 13 Once in a While C1 Sometimes C1 Always

8. You have a lot of respect for
C1 Never 13 Once in a While C1 Sometimes C1 Always

9. interferes with your activities...
C1 Never 13 Once in a While C1 Sometimes 13 Always

10. Youthink is terrific...

C1 Never 13 Once in a While C1 Sometimes Always

D

136

 

11. You feel very angry toward
C1 Never C1 Once in a While 13 Sometimes 13 Always

12. You feel violent toward ..
13 Never 13 Once m a While C1 Sometimes 13 Always

13. You feel proud of

C1 Never 13 Once in a While C1 Sometimes 13 Always

Directions to be read to the juvenile. Now I am going to ask you some more questions about
how likely you would be to approach about certain things. I will ask you a question
and you tell me if you would be Very Likely, Likely, Unlikely, or Very Unlikely to take the
action that I am talking about.

(Provide response card # 6 to the juvenile).

If you needed to, how likely would you be to...

14. Talk to about personal or private things? Would you say...
13 Very Likely C1 Likely 13 Unlikely C1 Very Unlikely

15. Ask for advice when you need to make an important decision?
C1 Very Likely C1 Likely 13 Unlikely C1 Very Unlikely

16. Borrow money from ?
C1 Very Likely C1 Likely 13 Unlikely C1 Very Unlikely

17. Talk to about problems you are having with your family?
13 Very Likely C1 Likely C1 Unlikely 13 Very Unlikely

18. Talk to about problems you are having with a friend?
13 Very Likely 13 Likely 13 Unlikely C1 Very Unlikely

19. Feel you can trust with just about anything you tell him/her?
13 Very Likely 13 Likely C1 Unlikely 13 Very Unlikely

20. Talk to about trouble you are having at work or school?
C1 Very Likely C1 Likely C1 Unlikely 13 Very Unlikely

21. Go to for help in an emergency?
13 Very Likely C1 Likely 13 Unlikely 13 Very Unlikely

22. Go places or do things with ?
13 Very Likely 13 Likely 13 Unlikely 13 Very Unlikely

Directions to be read to the juvenile. Now I am going to ask you some more questions about
what knows about your activities. I will ask you a question and you tell me if she/he

137

 

Knows Nothing, Knows Just A Little, Knows Some Things, or Knows Everything about the
things that I describe.

(Provide response card # 7 to the juvenile).

23

24.

25.

26.

27.

28.

Directions to be read to juvenile. I am going to read a couple of statements about

. How much does know about your close friends, that is who they are?
13 Knows Nothing C1 Knows Very Little 13 Knows Something 13 Knows Everything
How much does know about what you do with your fiiends?

C1 Knows Nothing C1 Knows Very Little C1 Knows Something C1 Knows Everything

How much does know about your close fiiends’ parents, that is, who they are?
C1 Knows Nothing 13 Knows Very Little C1 Knows Something 13 Knows Everything
How much does know about who you are with when you are not at home?

13 Knows Nothing 13 Knows Very Little 13 Knows Something 13 Knows Everything

How much does know about who your teachers are?

13 Knows Nothing C1 Knows Very Little C1 Knows Something C1 Knows Everything

How much does know about what you are doing in school?

C1 Knows Nothing C1 Knows Very Little 13 Knows Something C1 Knows Everything

 

 

Tell me if you Strongly Disagree, Disagree, Agree, or Strongly Agree with these statements.

(Provide response card # I to the juvenile).

29. Ithinkhighly of .
C1 Strongly Agree C1 Agree C1 Disagree C1 Strongly Disagree

30.

31

is a person I want to be like.

C1 Strongly Agree 13 Agree 13 Disagree C1 Strongly Disagree

. I really enjoy spending time with

Cl

32.

Strongly Agree 13 Agree 13 Disagree

Cl

33.

Cl

34.

D

35.

Cl

36.

D

Strongly Agree 13 Agree 13 Disagree 13 Strongly Disagree

praises me for doing well.
Strongly Disagree

El

criticizes me and my ideas.
Strongly Agree 13 Agree 13 Disagree C1 Strongly Disagree

helps me do things that are important to me.
Strongly Agree 13 Agree 13 Disagree 13 Strongly Disagree

blames me for his/her problems.
Strongly Agree 13 Agree 13 Disagree C1 Strongly Disagree

makes plans with me and cancels them for no good reason.
Strongly Agree C1 Agree C1 Disagree 13 Strongly Disagree

138

SERIES D [GENERAL FAMILY QUESTIONS) (i1 gfq)

Directions to be read to juvenile. I want you to think about the family that you spent most of
your time with in the last year. I am going to ask you some questions about how often you do
things with the rest of your family during a normal week. Tell me if you do these things 0, 1, 2,
3, 4, 5, 6, or 7 times a week with your family during a typical week.

Interviewer: Fill in juvenile ’s response in the space provided.

1. In a typical week, how many days, from 0 to 7,
do you eat dinner with your family?

2. In a typical week, how many days, from 0 to 7, does housework get done when it is supposed
to, for example cleaning up after dinner, doing dishes, or taking out the trash?

 

3. In a typical week, how many days, from 0 to 7, do you do something frm as a family such as
play a game, go to a sporting event, go swimming, or some other fun activity?

 

4. In a typical week, how many days, from 0 to 7, do you do something religious as a family such
as go to chmch, pray or read the scriptures together?

 

(ilsuls)

Directions to be read to juvenile. I am going to read you a list of things that kids may or may not
experience. Think back to the last year and tell me YES if you experienced this or were involved
in the activity that I ask about in the last year or NO if you did not.

1. During the last year, did you get poor grades on your report card?
C1 Yes 13 No

2. During the last year, have you gotten into trouble with a teacher or principal at school?
C1 Yes 13 No

3. During the last year, did you get suspended from school?
C1 Yes 13 No

4. During the last year, did your family move to a new home or apartment?
13 Yes C1 No

5. During the last year, has your family had a new baby come into the family?
13 Yes 13 No

6. During the last year, has anyone moved out of your home?
C1 Yes 13 No

7. During the last year, did a family member die?
13 Yes 13 No

139

 

8. During the last year, has a family member become seriously ill, injured badly, and/or had to
stay at the hospital?
C1 Yes C1 No

9. During the last year, has someone else you know, other than a member of your family, gotten
beaten, attacked or really hurt by others?
13 Yes C1 No

10. During the last year, have you seen anyone beaten, shot or really hurt by someone?
13 Yes C1 No

11. In the past year, did you change where you go to school?
C1 Yes 13 No

12. During the last year, have you seen or been around people shooting guns?
13 Yes 13 No

13. During the last year, have you had to hide someplace because of shootings in your
neighborhood?
13 Yes 13 No

Directions to be read to juvenile As I mentioned at the beginning of this interview, I am from
Michigan State University, and am not in any way working for the police, the schools, or any
other agency in the city. We are looking to gather information about youth to better understand
why they sometimes get into trouble. Your individual information will be kept confidential and
will be protected to the maximum extent allowable by law.

(ilabs)

Directions to be read to juvenile. This section asks some questions about situations that you
might have been involved in. Think back on the last year and respond to the following questions
as completely and accurately as possible. If you had no opportunity to do this, please tell me. If
you did, indicate how often you did it in the past year (Never, 1-2 times, 3-4 times, 5 or more
times) even if you never did it, but had the opportunity to do it in the last year.

(Provide response card # 8 to the juvenile).

1. I hit back when someone hit me first.
13 No Opportunity 13 Never 13 1-2 Times C1 3-4 Times C1 5 + Times

2. Iencouraged others to fight.
C1 No Opporttmity C1 Never C1 1-2 Times 13 3-4 Times C1 5 +Times

3. I pushed, shoved, slapped, or kicked others.
C1 No Opportunity C1 Never C1 1-2 Times C1 3-4 Times C1 5 + Times

4. I got into a physical fight because I was angry.
C1 No Opportunity 13 Never C1 1-2 Times C1 3-4 Times 13 5 + Times

140

 

 

5. I walked away from a fight.
C1 No Opportrmity 13 Never 13 1-2 Times 13 3-4 Times C1 5 +Times

6. Iteased other kids.
13 No Opportunity 13 Never C1 1-2 Times 13 3-4 Times 13 5 +Times

7. I said things about other kids to make others laugh (made fun of them).
C1 No Opportunity 11 Never 13 1-2 Times C1 3-4 Times C1 5 + Times

8. I called other kids names.
13 No Opportunity 13 Never 13 1-2 Times 13 3-4 Times C1 5 +Times

9. I threatened to hit or hmt another kid.
C1 No Opportunity 13 Never C1 1-2 Times 13 34 Times C1 5 + Times

(1 l fdb)

Directions to be read to juvenile. The following section asks some questions about whether you
were *EVER" involved in certain activities that kids sometimes do. Answer YES if you have,
and NO if you have not. Then, I will ask you to think back about that activity and the last year
and how many times you were involved in that activity.

1. Have you ever skipped school without an excuse?
a. C1 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

2. Have you ever cheated on school tests?
a. C1 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

3. Have you ever been sent home from school for bad behavior?
a. C1 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

4. Have you ever run away from home?
a. C1 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

5. Have you ever, on pmpose, broken or damaged or destroyed something belonging to someone
else?

a. C1 Yes 13 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?

6. Have you ever taken money at home that did not belong to you like from your mother’s purse
or your parents’ dresser?

141

a. 13 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

7. Have you ever gone into somebody’s garden, backyard, house, or garage when you were not

 

supposed to be there?
a. C1 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

8. Have you ever taken anything from a car that did not belong to you?
a. 13 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

9. Have you ever taken something from a store without paying for it?
a. C1 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

10. Have you ever thrown rocks or bottles at people?
a. C1 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

11. Have you ever taken a car or motorcycle for a ride without the owner’s permission?
a. 13 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

12. Have you ever used a weapon or force to get money or things from people?
a. C1 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

13. Have you ever consumed any liquor?
a. 13 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

14. Have you ever chewed tobacco?
a. C1 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

15. Have you ever smoked marijuana?
a. 13 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

142

 

 

 

 

 

16. Have you ever smoked cigarettes?

a. C1 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
17. Have you ever sniffed glue?

a. 13 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
18. Have you ever used cocaine?

a. 13 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
19. Have you ever abused prescription medications?

a. 13 Yes 13 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
20. Have you ever used ecstasy?

a. C1 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
21. Have you ever attended a rave?

a. C1 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
22. Have you ever used any other types of drugs (besides marijuana)?

a. 13 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?

d. What drugs have you used ? -
23. Have you ever hit, slapped, or shoved a grown-up such as a parent or teacher?

a. C1 Yes 13 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
24. Have you ever hit, slapped, or shoved other kids or got into a physical (fight with them?

a. C1 Yes C1 No

b. At what age did you first do this ?

c. How many times have you done this in the last 12 months ?
25. Have you ever written things or sprayed paint on walls or sidewalks or cars, where you were
not supposed to do that?

a. 13 Yes C1 No
b. At what age did you first do this ?

143

c. How many times have you done this in the last 12 months ?

26. Have you ever been loud or rowdy in a public place so that people complained about it or

 

you got into trouble?
a. 13 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

27. Have you ever purposely set fire to a building, a car, or something else or tried to do so?
a. 13 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

28. Have you ever canied a weapon with you?
a. C1 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?
(1. Type of weapon:

 

29. Have you ever avoided paying for things such as movies, bus or subway rides, or food?
a. 13 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

30. Have you ever snatched someone’s purse or wallet or picked someone’s pocket?
a. 13 Yes 13 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

31. Have you ever been arrested?
a. C1 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

32. Have you ever gone to court?
a. C1 Yes C1 No
b. At what age did you first do this ?
c. How many times have you done this in the last 12 months ?

ASSESSMENT OF THE INTERVIEW (ilaoi)

This section asks you about your responses to questions in this survey.

1. In your opinion, what causes kids to get in trouble? Check all that youth mentions

C1 Family problems C1 Bad Luck .
13 School problems 13 Using alcohol or drugs
C1 Being Angry C1 Friends

13 Being Depressed C1 Other

 

 

 

144

Respond with the answer that is, in your opinion, most accurate.

2. While you were answering the interviewer’s questions, how accurate were your responses?
C1 Very Accurate C1 Somewhat Accurate 13 Somewhat Inaccurate 13 Very Inaccurate

3. While you were answering the interviewer’s questions, how truthful were your responses?
13 Very Truthful C1 Somewhat Truthful C1 Somewhat Untruthful C1 Very Untruthful

4. How helpfirl will these questions be in helping researchers understand why kids get in
trouble?
13 Very Helpful 13 Somewhat Helpful 13 Not Helpful at A11
5. How helpful will these questions be in tmderstanding why you get in trouble?
C1 Very Helpful 13 Somewhat Helpful C1 Not Helpful at All

Interviewer Assessment" Poor Fair Good Excellent

*Interviewers interpretation of how well the interview went. For example, did the youth
understand the questions? Did the youth appear attentive? Did the youth appear to put some
thought into the answers?

145

 

 

 

 

one Baa-pee 8. v a.

 

 

 

 

 

- medal 200m coco—«>95 a?
..NDV. - 95—03521 988 coco—«>05 N?
.Rm. .3. - 353$ - 88 8832a S
SN: 5.- SN.- - 855 econ baa-a .x
ham.- .8? .mmmc .§. - Afioevmeaom base .x
«3.. .m8- ME.- .Rm- .5. - cause :88 rag .x
- - - - - - - 8332 wEEEBwoi 328m ox
m8.- Ro. to. NS. .mmmc .Nar - - 253355 swam .x
- - - .. - - - - - 8305 Emban— J"
.03. NE. :3. So. Soc 82.- - N2. - - 09.. an
.3... :7 m2. so: 9;.- 5: - e8.- - an? - 2.2 an
2:. SN. :8. 3:.- a8. 2:. - 2:.- - :3. .am- 22; .x
9 c» S .x .x .x .x .x an .x .x
Ga u 2 3.5.5 :emtamfieuv x232 zeta—9.30 .Nm 059—.
can Ems-03v 8. v a.
- came 586 e .w
N2 - wEQI Boom 02836.5 a.”
Re. .an - engage/T 28m 3835 or
Mac. 0mm. :3. - 3835-88 88.3.55 S
So. as.- .vmm- m2- - 8E6 23 been .x
Re.- aSc .8? 2:.- .Sv. - A525 308m 55$ .x
:6.- So: 8o. e8.- .:m.- .wum. - A825 :28 base i
9;.- 3? :2. 9;. SN.- So- at. - maze: mega-ace base .x
«2.. Eo. 2:.- es. mun. mt. mac.- Eq- - 2.3.5385 336 .x
:N. .mmm. 35.. :~.- 5. «S. es.- 93: se,- - €355.28 .x
.30.. 38. So. 3:. 3°.- amor 48. .80- to. E.- - om... ”x
amo- 8o. 92. 3o. «3. 42. 2o. woo. .oem. 5. So.- - 2.2 an
5.- §. as.- em? as. 5.. ws.. So. .3? So. So. So.- 323 .x
S 9 9 S .x .x ..x .x .x .x .x .x

 

8e 1 Z E520 «5532-5 5.532 zeta—9.39 Am 03:.

doom-5“: EOuG—Ohhcnv um 597“th

146

APPENDIX C: City and Target Area Characteristics.

Table 2. Descriptive Statistics for the City and Target Area

 

 

City Taget Area
Total population 207,886 51,734
Percent male 48.92 50.54
Percent minority 3 1 .56 29.24
Percent foreign born 10.35 16.29
Percent renter occupied units 37.87 38.22
Percent of families with children split 19.39 22.87
Percent of individuals with college education (over 25 24.66 13.71
years old)
Percent of families with income greater than $25,000 68.45 65.91
Percent of families below the poverty level 11.47 13.06
Percent of households receiving public assistance 4.71 5.60
Percent female headed households 23.73 25.69
Percent unemployed (over 16 years old and in labor 6.11 7.60
force)
Percent youth 26.79 29.48
Percent African-American 19.30 10.34
Percent of buildings that are vacant 6.11 7.09
Concentrated disadvantage’ -0.004 0.011

 

IF actor Score — higher values indicate more disadvantage

147

 

 

 

REFERENCES

Akers, R. (1985). Deviant Behavior: A Social Learning Approach (3rd ed.). Belmont,
CA, Wadsworth.

Akers, R. (1991). Self-control as a general theory of crime. Journal of Quantitative
Criminology, 7(2), 201 -21 1.

 

Akers, R. L. (1998). Social Learning and Social Structure: A General Theory of Crime
and Deviance. Boston: Northeastern University Press.

Akers, R. L. (2000). Criminological Theories. Los Angeles: Roxbury.
Alarid, L. F., Burton, Jr., V. S., Cullen, F. T. (2000). Gender and crime among felony

offenders: Assessing the generality of social control and differential association
theories. Journal of Research in Crime and Delinquency, 37(2), 171-199.

 

Anderson, A. L. (2002). Individual and contextual influences on delinquency: The role
of the single-parent family. Journal of Criminal Justice, 30(6), 575-587.

Anderson, B. J., Holmes, M. D., Ostresh, E. (1999). Male and female delinquents’
attachments and effects of attachments on severity of self-reported delinquency.
Criminal Justice and Behavior, 26(4), 435-452.

Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J ., Gendreau, P., & Cullen, F. T. (1990).
Does correctional treatment work? A clinically-relevant and psychologically-
infonned meta-analysis. Criminology, 28(3), 369-404.

Ayers, C. D., Williams, J. H., Hawkins, J. D., Peterson, P. L., Catalano, R. F., & Abbott,
R. D. (1999). Assessing correlates of onset, escalation, de-escalation, and
desistance of delinquent behavior. Journal of Quantitative Criminology, 15(3),
277-306.

 

Bandura, A. (1973). Aggression: A Social Learning Analysis. Englewood Cliffs, N.J.:
Prentice-Hall.

Bandura, A. (1986). Social Foundations of Thought and Action. New York: Prentice
Hall.

Bank, L. & Burraston, B. (2001). Abusive home environments as predictors of poor
adjustment during adolescence and early adulthood. Journal of Community
Psychology, 29(3), 195-2 1 7.

Baumrind, D. (1991). The influence of parenting style on adolescent competence and
substance use. Journal of Early Adolescence, I I, 56-95.

148

Besharov, D. J. (1987). Giving the juvenile court a preschool education. In J. Q. Wilson
& G. C. Loury (eds.). From Children to Citizens (Vol III): Families, Schools, and
Delinquency Prevention (pp. 207-238). New York: Springer-Verlag.

Black, M. M, Howard, D. E., Kim, N., & Ricardo, I. (1998). Interventions to prevent
violence among Afiican American Adolescents from low-income communities.
Aggression and Violent Behavior, 3(1), 17-33.

Blumstein, A., Cohen, J ., Roth, J. A., & Visher, C. A. (1986). Criminal Careers and
Career Criminals. Washington DC: National Academy Press.

Bowbly, J. (1947). Forty-Four Juvenile Thieves. London: Bailliero, Tindall, and Cox. H

Bowen, M. (1978). Family Therapy in Clinical Practice. Northvale, NJ: Jason Aronson.

 

Bray J. H. & Hetherington, E. M. (1993). Families in transition: Introduction and
overview. Journal of Family Psychology, 7(1), 3-8.

 

Brooks-Gum, J., Duncan, G. J., Klebanov, P. K, & Sealand, N. (1993). Do
neighborhoods influence child and adolescent development? American Journal of
Sociology, 99(2), 353-395.

Bronfrenbrenner, U. (1979). The Ecology of Human Development: Experiments by
Design and Nature. Cambridge, MA: Harvard University Press.

Browning, K., Huizinga, D., Loeber, R., & Thomberry, T. P. (1999). Fact Sheet: Causes
and correlates of delinquency program. Washington, DC: Office of Juvenile
Justice and Delinquency Prevention.

Burgess, R. & Akers, R. (1966). A differential association-reinforcement theory of
criminal behavior. Social Problems, 14, 128-147.

Bums, B. J ., Howell, J. C., Wiig, J. K., Augimeri, L. K., Welsh, B. C., Loeber, R. etal.,
(2003). Treatment, services, and intervention programs for child delinquents.
Washington DC: Office of Juvenile Justice and Delinquency Prevention.

Bursik, R. J. Jr. (1988). Social disorganization and theories of crime and delinquency:
Problems and prospects. Criminology, 26, 519-551.

Burton, V. S. Jr., Cullen, F. T., Evans, D., Alarid, L F., & Dunaway, R. G. (1998).
Gender, self-control, and crime. Journal of Research in Crime and Delinquency,
35, 123-147.

Cantor, R. J. (1982). Family correlates of male and female delinquency. Criminology-
20, 149-166.

149

Caspi, A., Moffitt, T. E., Silva, P. A., Loeber, M. S., Krueger, R. F., & Schmutte, P. S.
(1994). Are some people crime-prone? Replications of the personality-crime
relationship across countries, genders, races, and methods. Criminology, 32, 163-
196.

Catalano, R. F. & Hawkins, J. D. (1996). The social development model: A theory of
antisocial behavior. In J. D. Hawkins (ed.). Delinquency and Crime: Current
Theories (pp. 149-197). Cabridge, UK: Cambridge University Press.

Cemkovich, S. A. (1978). Evaluating two models of delinquency causation. .
Criminology, 16(3), 335-354. A
i

 

Cemkovich, S. A. & Giordano, P. C. (1987). Family Relationships and Delinquency.
Criminology 25(2), 295-321.

Cemkovich, S.A. & Giordano, P. C. (1992). School bonding, race, and delinquency.
Criminology, 30(2), 261-289. ' j

 

Cobb, N. J. (2001). Adolescence. Mountain View, CA: Mayfield Publishing Company.

Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal
of Sociology, 94, S95-120.

Coley, R. L. & Monis, J. E. (2002). Comparing father and mother reports of father
involvement among low-income minority families. Journal of Marriage and
Family, 64(4), 982-995.

 

Connell, A. M. & Goodman, S. H. (2002). The association between psychopathology in
fathers versus mothers and children’s internalizing and externalizing behavior
problems: A meta-analysis. Psychological Bulletin, 128(5), 746-773.

Cullen, F. T ., Wright, J. P., Brown, S., Moon, M. M., Blankenship, M. B., & Applegate,
B. K. (1998). Public support for early intervention programs: Implications for a
progressive policy agenda. Crime & Delinquency, 44(2), 187-204.

Davies, M. & Sinclair, I. (1971). Families, hostels and delinquents: An attempt to assess
cause and effect. British Journal of Criminology, 11(3), 213-229.

Demuth, S. & Brown, S. L. (2004). Family structure, family processes, and adolescent
delinquency: The significance of parental absence versus parental gender.
Journal of Research in Crime and Delinquency, 41(1), 58-81.

Dombusch, S. M., Erickson, K. G., Laird, J ., Wong, C. A. (2001). The relation of family

and school attachment to adolescent deviance in diverse groups and communities.
Journal of Adolescent Research, 16(4), 396-422.

150

Durkheim, E. (1951). Suicide. New York: Free Press.

Elliott, D. S. (1994). Serious violent offenders: Onset, developmental course, and
termination: The American Society of Criminology 1993 presidential address.
Criminology, 32, 1-21.

Elliott, D. S. & Ageton, S. S. (1980). Reconciling race and class differences in self-
reported and official estimates of delinquency. American Sociological Review,
45, 95-110.

Elliot, D. S., Huizinga, D., & Ageton, S. S. (1985). Explaining delinquency and drug
use. Beverly Hills, CA: Sage.

Erickson, K. G., Crosnoe, R., Dombusch, S. M. (2000). A social process model of
adolescent deviance: Combining social control and differential association
perspectives. Journal of Youth and Adolescence, 29(4), 395-425.

Espiritu, R. C., Huizinga, D., Crawford, A. M., & Loeber, R. (2001). Epidemiology of
self-reported delinquency. In R. Loeber 8: D. P. Farrington (Eds). Child
Delinquents: Development, Intervention, and Service Needs (pp. 47-66).
Thousand Oaks, CA: Sage.

Fagan, J .A. and Wexler. S. (1987). Family origins of violent delinquents. Criminology
25(3), 643-669.

F arrington, D. P. (1986). Age and crime. In M. Tonry and N. Morris (Eds). Crime and
Justice: An Annual Review of Research (V 01. 7, pp 189-250). Chicago:
University of Chicago Press.

Farrington, D. P. (1987). Early precursors of frequent offending. In J. Q. Wilson and G.
C. Loury (Eds), From Children to Citizens (Vol [11): Families, Schools, and
Delinquency Prevention (pp. 27-50). New York: Springer-Verlag.

Farrington, D. P. & Loeber, R. (1998). Major aims of this book. In R. Loeber & D. P.

F arrington (Eds), Serious & Violent Juvenile Ofi'enders: Risk Factors and
Successful Interventions (pp. 1-12). Thousand Oaks: Sage.

Fox, J. (1991). Regression Diagnostics. Thousand Oaks, CA: Sage.

Geis, G. (2000). On the absence of self-control as the basis for a general theory of crime.
Theoretical Criminology, 4, 35-53.

Gerard, J. M. & Buehler, C. (1999). Multiple risk factors in the family environment and
youth problem behaviors. Journal of Marriage and the Family, 61, 343-361 .

Gibbons, D. C. (1999). Review essay: Changing lawbreakers—what have we learned

151

 

 

 

since the 1950’s? Crime and Delinquency, 45, 272-293.

Glueck, S. & Glueck, E. T. (1950). Unraveling Juvenile Delinquency. New York:
Commonwealth Fund.

Glueck, S. Glueck, E. T. (1952). Delinquents in the Making: Paths to Prevention. New
York: Harper and Row.

Glueck, S. & Glueck, E. (1962). Family Environment and Delinquency. Boston:
Houghton Mifflin.

German-Smith, D., Tolan, P. H., Zelli, A., & Huesmann, L. R. (1996). The relation of
family functioning to violence among inner-city youth. Journal of Family
Psychology, 10(2), 1 15-129.

Gorman-Smith, D., Tolan, P. H., Loeber, R., & Henry, D. B. (1998). Relation of family
problems to patterns of delinquent involvement among urban youth. Journal of
Abnormal Child Psychology, 26(5), 319-333.

Gottfredson, M. R. & Hirschi, T. (1990). A General Theory of Crime. Stanford, CA:
Stanford University Press.

Grasmick, H. G., Tittle, C. R., Bursik, Jr. R. J ., & Ameklev, B. J. (1993). Testing the
core empirical implications of Gottfredson and Hirschi’s general theory of crime.
Journal of Research in Crime and Delinquency, 30(1), 5-29.

Greenberg, D. F. (1999). The weak strength of social control theory. Crime and
Delinquency, 45, 66-81.

Greenberg, D. F. (1985). Age, crime, and social explanation. American Journal of
Sociology, 91(1), 1-21.

Guerra, N. G. (1998). Serious and violent juvenile offenders: Gaps in knowledge and
research priorities. In R. Loeber & D. P. F anington (Eds), Serious & Violent
Juvenile Ojfenders: Risk Factors and Successful Interventions (pp. 389-404).
Thousand Oaks: Sage.

Hawkins, J. D., Catalano, R. F., Jones, G., & Fine, D. (1987). Delinquency prevention
through parent training: Results and Issues from work in progress. In J. Q.
Wilson and G. C. Loury (Eds), From Children to Citizens (Vol III): Families,
Schools, and Delinquency Prevention (pp. 186-204). New York: Springer-
Verlag.

Hawkins, D. F., Laub, J. H., & Lauritsen, J. L. (1998). Race, ethnicity, and serious

152

 

 

 

juvenile offending. In R. Loeber & D. P. F arrington (Eds), Serious & Violent
Juvenile Oflenders: Risk Factors and Successfirl Interventions (pp. 30-46).
Thousand Oaks: Sage.

Hay, C. (2001). Parenting, self-control, and delinquency: A test of self-control theory.
Criminology, 39(3), 707-736.

Heck, C. & Walsh, A. (2000). The effects of maltreatment and minor and serious
delinquency. International Journal of Oflender Therapy and Comparative
Criminology, 44(2), 178-193.

Herrenkohl, T. I, Hawkins, J. D., Chung, I-J., Hill, K. G., & Battin-Pearson, S. (2001).
School and community risk factors and interventions. In R. Loeber & D. P.
F arrington (Eds), Child Delinquents: Development, Intervention, and Service
Needs (pp. 211-246). Thousand Oaks: Sage.

 

.mr‘r
n
-
-

 

Herrenkohl, T. 1., Huang, B., Kostennan, R., Hawkins, J. D., Catalano, R. F., & Smith, B. J
H. (2001). A comparison of social development processes leading to violent
behavior in late adolescence for childhood initiators and adolescent initiators of
violence. Journal of Research in Crime and Delinquency, 38(1), 45-63.

Hill, K. G., Howell, J. C., Hawkins, J. D., & Battin-Pearson, S. R. (1999). Childhood risk
factors for adolescent gang membership: Results from the Seattle social
development project. Journal of Research in Crime and Delinquency, 36(3), 300-
322.

Hindelang, M. J ., Hirschi, T., & Weis, J. G. (1981). Measuring Delinquency. Beverly
Hills, CA: Sage.

Hirschi, T. (1969). Causes of Delinquency. Berkley: University of California Press.

Hirschi, T., & Gottfi'edson, M. (1983). Age and the explanation of crime.- American
Journal of Sociology, 89, 552-584.

Howell, J. C. (1995). Guide for implementing the comprehensive strategy for serious,
violent, and chronic juvenile oflenders. Washington, DC: Office of Juvenile
Justice and Delinquency Prevention.

Huang, B., Klostennan, R., Catalano, R. F., Hawkins, J .D., & Abbott, R. D. (2001).
Modeling mediation in the etiology of violent behavior in adolescence: A test of
the social development model. Criminology, 39(1), 75-107.

Huizinga, D. & J akob-Chien, C. (1998). The contemporaneous co-occurrence of serious

153

 

and violent juvenile offending and other problem behaviors. In R. Loeber & D. P.

Farrington (Eds), Serious & Violent Juvenile Offenders: Risk Factors and
Successful Interventions (pp. 47-67). Thousand Oaks: Sage.

Hunner, R. & Walker, Y. (Eds). (1981). Exploring the relationship between child abuse
and delinquency. Montclair, NJ: Allanheld & Schrarn.

Ireland, T. O., Smith, C. A., & Thomberry, T. P. (2002). Developmental issues in the
impact of child maltreatment on later delinquency and drug use. Criminology 40,
359-399.

Johnson, R. E. (1986). Family structure and delinquency: General patterns and gender
differences. Criminology, 24, 65-80.

Jonson, M. R. (1998). Youth violent behavior and exposure to violence in childhood: An
ecological review. Aggression and Violent Behavior, 3(2), 159-179.

J uby, H. & Farrington, D. P. (2001). Disentangling the link between disrupted families
and delinquency. British Journal of Criminology, 41, 22-40.

Junger, M. & Marshall, I. H. (1997). The interethnic generalizability of social control
theory: An empirical test. Journal of Research in Crime and Delinquency, 34(1),
79-1 12.

Junger, M. & Tremblay, R. E. (1999). Self-control, accidents, and crime. Criminal
Justice and Behavior. 26, 485-501.

Kazdin, A. E. (1987). Treatment of antisocial behavior in children: Current status and
future directions. Psychological Bulletin, 102, 187-203.

Knutson, J. F. (1995). Psychological characteristics of maltreated children: Putative risk
factors and consequences. Annual Review of Psychology, 46, 401-431.

Komhauser, R. (1978). Social Sources of Delinquency. Chicago: University of Chicago
Press.

Krohn, M. D. & Massey, J. L. (1980). Social control and delinquent behavior: An
examination of the elements of the social bond. The Sociological Quarterly, 21,
529-543.

LaGrange, T. C. & Silverrnan, R. A. (1999). Low self-control and opportunity: Testing
the general theory of crime as an explanation for gender differences in

delinquency. Criminology, 37, 41-72.

Latirner, J. (2001). A meta-analytic examination of youth delinquency, farmly treatment,
and recidivism. Canadian Journal of Criminology, April, 237-253.

154

 

 

 

Laub, J. H. & Sampson, R. J. (199.1). The Sutherland-Glueck debate: On the sociology of
criminological knowledge. American Journal of Sociology, 96, 1402-1440.

Lipsey, M. W. (1999). Can intervention rehabilitate serious delinquents? Annals of the
American Association of Political and Social Science, 564, 142-166.

Lipsey, M. W. & Derzon, J .H. (1998). Predictors of violent or serious delinquency in
adolescence and early adulthood: A synthesis of longitudinal research. In R.
Loeber & D. P. Fanington (Eds), Serious & Violent Juvenile Oflenders: Risk
Factors and Successful Interventions (pp. 106-146). Thousand Oaks: Sage.

Lipsey, M. W. & Wilson, D. B. (1998). Effective intervention for serious juvenile
offenders: A synthesis of research. In R. Loeber 8: D. P. F arrington (Eds),
Serious & Violent Juvenile Ojfenders: Risk Factors and Successful Interventions
(pp. 313-345). Thousand Oaks: Sage.

Loeber, R. (1982). The stability of antisocial and delinquent child behavior: A review.
Child Development, 53, 1431-1446.

Loeber, R. (1987). What policy makers and practitioners can learn from family studies of
juvenile conduct problems and delinquency. In J. Q. Wilson and G. C. Loury
(Eds), From Children to Citizens (Vol III): Families, Schools, and Delinquency
Prevention (pp. 78-111). New York: Springer-Verlag.

Loeber, R. 8:. F arrington, D. P. (Eds). (1998). Serious & Violent Juvenile Offenders:
Risk Factors and Successful Interventions. Thousand Oaks: Sage. Thousand
Oaks, CA: Sage.

Loeber, R. & Fanington, D. P. (Eds). (2001). Child Delinquents: Development,
Intervention, and Service Needs. Thousand Oaks, CA: Sage.

Loeber, R, Farrington, D. P., & Waschbuch, D. A. (1998). Serious and violent juvenile
offenders. In R. Loeber & D. P. F anington (Eds), Serious & Violent Juvenile
Oflenders: Risk Factors and Successfirl Interventions (pp. 13-29). Thousand
Oaks: Sage.

Loeber, R. & Stouthamer-Loeber, M. (1986). Family factors as correlates and predictors
of juvenile conduct problems and delinquency. In M. Tonry & N. Morris (Eds),
Crime and Justice (pp. 29-149). Chicago: University of Chicago Press.

Loury, G. (1987). The family as context for delinquency prevention: Demographic
trends and political realities. In J. Q. Wilson and G. C. Loury (Eds), From
Children to Citizens (Vol III): Families, Schools, and Delinquency Prevention
(pp. 3-26). New York: Springer-Verlag.

155

 

 

Lynskey, D. P., Winfree, Jr., L. T., Esbensen, F ., & Clason, D. L. (2000). Linking
gender, minority group status and family matters to self-control theory: A
multivariate analysis of key self-control concepts in a youth-gang context.
Juvenile & Family Court Journal, 51(3), 1-19.

Lyons, K. S., Zarit, S. H., Sayer, A. G., & Whitlatch, C. J. (2002). Caregiving as a
dyadic process: Perspectives from caregiver and receiver. Journal of
Gerontology, 57B(3), 195-204.

Marsh, H. W. & Hau, K. (1999). Confinnatory factor analysis: Strategies for small
sample sizes. In R. H. Hoyle (Ed), Statistical Strategies for Small Sample
Research (pp. 251-284). Thousand Oaks, CA: Sage.

Martinson, R. (1974). What works? Questions and answers about prison reform. Public
Interest, 35, 22-54.

Martinson, R. (1979). Symposium on sentencing: Part H. Hofstra Law Review, 7(2),
243-258.

Matsueda, R. L. (1982). Testing control theory and differential association: A causal
modeling approach. American Sociological Review, 47, 489-504.

Matsueda,) R. (1988). The current state of differential association theory. Crime and
Delinquency, 34, 277-306.

Maxfield, M. G., Weiler, B. L., & Widom, C. S. (2000). Comparing self-reports and
official records of arrest. Journal of Quantitative Criminology, 16, 87-110.

May, D. C., Vartanian, L. R., & Virgo, K. (2002). The impact of parental attachment and
supervision on fear of crime among adolescent males. Adolescence, 37(146),
267-287.

Mazerolle, P., Brame, R., Paternoster, R., Piquero, A., &-Dean, C. (2000). Onset age,
persistence, and offending versatility: Comparisons across gender. Criminology,
38(4), 1143-1 172.

McCarthy, E. D., Gersten, J. C. & Langner, T. S. (1982). The behavioral effects of father
absence on children and their mothers. Social Behavior and Personality 10, 11-
23.

McCord, J. (1989). A forty year perspective on effects of child abuse and neglect. Child
Abuse & Neglect, 7, 265-270.

McCord, J. (1991). Family relationships, juvenile delinquency, and adult criminality.
Criminology, 29: 243-266.

156

 

 

 

Mednick, S. & Christiansen, K. O. (1977). Biosocial Bases or Criminal Behavior. New
York: Gardner Press. .

Menard, S. (1995). Applied Logistic Regression Analysis. Thousand Oaks, CA: Sage.

Mihalic, S., Irwin, K., Elliott, D., Fagan, A., & Hansen, D. (2001). Blueprints for
violence prevention. Washington DC: Office of Juvenile Justice and Delinquency
Prevention.

Miller, K. S. & Knutson, J. F. (1997). Reports of severe physical punishment and
exposure to animal cruelty by inmates convicted of felonies and by university
students. Child Abuse & Neglect, 21 , 59-82.

Moffitt, T. E. (1993). Adolescence-limited and life-course persistent antisocial behavior:
A developmental taxonomy. Psychological Review, 100, 674-701.

Moon, M. M., Sundt, J. L., Cullen, F. T., & Wright, J. P. (2000). Is child saving dead?
Public support for juvenile rehabilitation. Crime & Delinquency, 46(1), 38-60.

 

Nagin, D. & Paternoster, R. (1991). On the relationship of past and future participation
in delinquency. Criminology, 29, 163-190.

Needle, R., Su, S., Doherty, W., Lavee, Y., & Brown, P. (1988). F amilial, interpersonal,
and intrapersonal correlates of drug use: A longitudinal comparison of adolescents

in treatment, drug using adolescents not in treatment, and non-drug using :
adolescents. International Journal of the Addictions, 23, 1211-1240.
1
I
l

Nye, F. I. (1958). Family Relationships and Delinquent Behavior. New York: Wiley.

Olweus, D. (1993). Bullying as School: What we know and what we can do.
Cambridge, United Kingdom: Blackwell.

OJJDP (1998). Serious and violent juvenile offenders. Washington DC: Office of
Juvenile Justice and Delinquency Prevention.

Palmer, T. (1991). The effectiveness of intervention: Recent trends and current issues.
Crime & Delinquency, 3 7(3), 330-346.

Palmer, T. (1995). Programmatic and nonprogrammatic aspects of successful
intervention: New directions for research. Crime & Delinquency, 41 (1), 100-131.

Park, R. & Burgess, E. (1924). Introduction to the Science of Sociology. Chicago: Univ.
Press.

Patchin, J. W., Bynum, T. S., Huebner, B. M., McCluskey, J. D., & Varano, S. P. (2003).

157

Deviance, decay, of disadvantage? Disaggregating social disorganization theory.
Unpublished manuscript, Michigan State University.

Patternoster, R. & Brame, R. (1998). The structural similarity of processes generating
criminal and analogous behaviors. Criminology, 36, 633-670.

Patterson, G. R. (1980). Children Who Steal. In T. Hirschi and M. Gottfredson (Eds),
Understanding Crime: Current Theory and Research (pp. 73-90). Beverley Hills:
Sage.

Patterson, G. R. (1982). Social Learning Approach to Family Intervention. Eugene, OR:
Castilia Publishing Company.

Patterson, G. R., DeBaryshe, B. 8L Ramsey, E. (1989). A developmental perspective on
antisocial behavior. American Psychologist, 44, 329-35.

Patterson, G. R, Reid, J. B., & Dishion, T. J. (1992). Antisocial Boys. Eugene, OR:
Castalia Publishing Company.

Perkins-Dock, R. E. (2001). Family interventions with incarcerated youth: A review of
the literature. International Journal of Oflender Therapy and Comparative
Criminology, 45(5), 606-625.

Piquero, A. R. & Brezina, T. (2001). Testing Moffitt’s account of adolescent-lirnited
delinquency. Criminology, 39(2), 353-370.

Pratt, T. C. & Cullen, F. T. (2000). The empirical status of Gottfredson and Hirschi’s
general theory of crime. Criminology, 38, 931-964.

Quesnel, S., McArdle, P., Brinkley, A., Wiegersma, A., et a1 (2002). Broken home or
drug using peers: “Significant relations”? Journal of Drug Issues, 32(2), 467-490.

Rankin, J. H. (1983). The family context of delinquency. Social Problems, 30, 466-479.

Rankin, J. H. & Kern, R. (1994). Parental attachments and delinquency. Criminology,
32, 495-514.

Raudenbush, S. W. & Bryk, A. S. (2002). Hierarchical Linear Models: Applications and
Data Analysis Methods. Thousand Oaks: Sage Publications.

Raudenbush, S. W. & Bryk, A. S., Cheong, Y. F., 8: Congdon, R. T. (2000). HLM5:
Hierarchical linear and non-linear modeling. Chicago: Scientific Software
International.

Rebellion, C. J. (2002). Reconsidering the broken homes/delinquency relationship and
exploring its mediating mechanisms. Criminology, 40(1), 103-135.

158

 

Reiss, A. (1951). Delinquency as the failure of personal and social controls. American
Sociological Review, 16, 196-207.

Rowe, D. & Osgood, D. W. (1984). Heredity and sociological theories of delinquency: A
reconsideration. American Sociological Review, 49, 526-540.

Sahnelainen, P. (1996). Child neglect: Its causes and its role in delinquency. Crime and
Justice Bulletin, 33, 1-14.

Sampson, R J. (1985). Neighborhood and crime: The structural determinants of personal
victimization. Journal for Research in Crime and Delinquency, 22, 7-40.

Sampson, R. J. & Groves W. B. (1989). Community structure and crime: Testing social
disorganization theory. American Journal of Sociology, 94, 774-802.

Sampson, R. J ., & Laub, J. H. (1995). Understanding variability in lives through time:
Contributions of life-course criminology. Studies on Crime and Crime
Prevention, 4(2), 143-158.

Sampson R. J. & Laub, J. H. (1993). Crime in the Making: Pathways and Turning Points
Through Life. Cambridge, MA: Harvard University Press.

Sherman, L. W., Gottfredson, D., MacKenzie, D., Eck, J ., Reuter, P., & Bushway, S.
(1997). Preventing Crime: What Works, What Doesn ’t, and What 's Promising?
Washington DC: National Institute of Justice.

Singleton, R. A. & Straits, B. C. (1999). Approaches to Social Research. New York:
Oxford University Press.

Skinner, B. F. (1953). Science and Human Behavior. New York: Macmillan.

Snyder, H. N. (2001). Epidemiology of official offending. In R. Loeber & D. P.
F arrington (Eds), Child Delinquents: Development, Intervention, and Service
Needs (pp. 25-46). Thousand Oaks: Sage.

Sokol-Katz, J ., Dunharn, R., & Zimmerman, R. (1997). Family structure versus parental
attachment in controlling adolescent deviant behavior: A social control model.

Adolescence, 32, 199-215.

Sutherland, E. H., Cressey, D. R., & Luckenbill, D. F. ( 1992). Principles of Criminology
(11th Ed). Dix Hills, New York: General Hall.

Steffensmeier, D. J ., Allan, E. A., Harer, M. D., & Streifel, C. (1989). Age and the
distribution of crime. American Journal of Sociology, 94, 803—83 1.

159

 

Tate, D. C., Reppucci, N. D., & Mulvey, E. P. (1995). Violent juvenile offenders:
Treatment effectiveness and implications for future action. American
Psychologist, 50(9), 777-781.

Thomberry, T. P. (1987). Toward and interactional theory of delinquency.
Criminology, 25, 863-887.

Thomberry, T. P. & Krohn, M. D. (2000). The self-report method for measuring
delinquency and crime. In Measurement and Analysis of Crime and Justice,
Criminal Justice 2000 (pp. 33-84). Washington, DC: National Institute of

Justice. F1

Thomberry, T. P., Smith, C. A., Rivera, C., Huizinga, D. & Stoutharner-Loeber, M.
(1999). Family Disruption and Delinquency. Juvenile Justice Bulletin, Office of
Juvenile Justice and Delinquency Prevention, U. S. Department of Justice.

Tienda, M., & Angel, R. (1982). Female headship and extended family composition: .
Comparisons of Hispanics, Blacks, and non-Hispanic whites. Social Forces, 61, .3
508-531.

 

Toby, J. (1957). Social disorganization and stakes in conformity: Complementary factors
in the predatory behavior of hoodlums. Journal of Criminal Law, Criminology,
and Police Science, 48: 12-17.

Tolan, P. H. & Gorrnan-Smith, D. (1998). Development of serious and violent offending
careers. In R. Loeber & D. P. F arrington (Eds), Serious & Violent Juvenile
Offenders: Risk Factors and Successful Interventions (pp. 68-85). Thousand
Oaks: Sage.

Tolan, P. H., Guerra, N. G., & Kendall, P. C. (1995). A developmental-ecological
perspective on adolescents: Toward a unified risk and intervention framework.
Journal of Consulting and Clinical Psychology, 63(4), 579-584.

Vazsonyi, A. T. & F lannery, D. J. (1997). Early adolescent delinquent behaviors:
Associations with family and school domains. The Journal of Early Adolescence,
17(3), 271-293.

Veltman, M. W. & Browne, K. D. (2001). Three decades of child maltreatment research:
Implications for the schools years. Trauma Violence and Abuse: A Review
Journal, 2(3), 215-239.

Wahler, R. G. (1987). Contingency management with oppositional children: Some
critical teaching issues for parents. In 1. Q. Wilson and G. C. Loury (Eds), From
Children to Citizens (Vol III): Families, Schools, and Delinquency Prevention
(pp. 112-131). New York: Springer-Verlag.

160

Waitrowski, M. D., Griswold, D. B., Roberts, M. K. (1982). Social control and
delinquency. American Sociology Review, 46, 525-541.

Wallace, J. M. Jr., & Bachman, J. G. (1991). Explaining racial/ethnic differences in
adolescent drug use: The impact of background and lifestyle. Social Problems,
38, 333-357.

Warr, M. (2002). Companions in Crime: The Social Aspects of Criminal Conduct.
Cambridge: Cambridge University Press.

Wasserman, G. A. & Miller, L. S. (1998). The prevention of serious and violent juvenile
offending. In R. Loeber & D. P. F arrington (Eds), Serious & Violent Juvenile
Oflenders: Risk Factors and Successful Interventions (pp. 197-247). Thousand
Oaks: Sage.

Wasserman, G. A., Miller, L. S., & Cothern, L. (2000). Prevention of serious and violent
juvenile oflending. Washington DC: Office of Juvenile Justice and Delinquency
Prevention.

Wells, L. E. & Rankin, J . H. (1988). Direct parental controls and delinquency.
Criminology, 26, 263-285.

Wells, L. E. & Rankin, J. H. (1991). Families and delinquency: A meta-analysis of the
impact of broken homes. Social Problems, 28(1), 71-83.

Westrnan, J. C. (1996). The rationale and feasibility of licensing parents. Society, 34(1),
46-60.

White, J. L, Moffitt, T.E., Earls, F., Robins, L., & Silva, RA. (1990). How early can we
tell? Predictors of childhood conduct disorder and adolescent delinquency.
Criminology, 28, 507-533.

Whitehead, J. T. & Lab, S. P. (1989). A meta-analysis of juvenile correctional treatment.
Journal of Research in Crime and Delinquency, 26(3), 276-95.

Wiatrowski, M. D., Griswold, D. B., & Roberts, M. K. (1981). Social control theory and
delinquency. American Sociological Review, 46, 525-541.

Wilson, J. Q. & Herrnstein, R. (1985). Crime and Human Nature. New York: Simon
and Schuster.

Wolfgang, M. E., Figlio, R. M., & Selling, T. (1972). Delinquency in a Birth Cohort.
Chicago, IL: University of Chicago Press.

Wright, J. P., Cullen, F. T., & Miller, J. T. (2001). Family social capital and delinquent
involvement. Journal of Criminal Justice, 29, 1-9.

161

 

 

11111111511111111I