2013. Vol.4, No.6A2, 1-7
Published Online June 2013 in SciRes (http://www.scirp.org/journal/psych) http://dx.doi.org/10.4236/psych.2013.46A2001
Copyright © 2013 SciRes. 1
A Selective Review of the Risk Factors for Antisocial Behavior
across the Transition to Adulthood
Joanne Savage1, Stephanie K. Ellis2, Kathryn Kozey1
1Department of Justice, Law and Society, American University, Washington DC, USA
2Department of Sociology and Criminal Justice, Marymount U n i ve rsity, Arlington, USA
Received April 9th, 2013; revised May 11th, 2013; accepted June 10th, 2013
Copyright © 2013 Joanne Savage et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited.
In this paper, we discuss the theory and research on a select set of risk factors for continuity in antisocial
behavior across the transition to adulthood. Several risk factors (e.g., early onset, intelligence, marriage,
employment) are based on Moffitt’s dual taxonomy and the age-graded theory of social control. In addi-
tion, we also review studies of impulsivity, school enrollment, educational attainment, academic achieve-
ment, abuse victimization, social support, poverty, deviant peers, drug and alcohol abuse, and criminal
Keywords: Transition to Adulthood; Risk Factors; Antisocial Behavior; Crime; Marriage
The transition to adulthood has been receiving increasing at-
tention as an important stage in development in recent years.
One essential focus of the literature is the persistence of antiso-
cial behavior into young adulthood. One reason this particular
outcome is important because, in some ways, it signifies a fail-
ure in certain developmental sequences. Another reason for its
importance is a practical one. Offending in adulthood is viewed
much differently in the criminal justice system than is juvenile
delinquency. In contrast with the juveile justice system, where
efforts are generally made to shepherd wayward youths back to
the fold, the adult criminal justice system, referred to anachro-
nistically as a “corrections” system, emphasizes accountability
and often includes an intentionally harsh punitive component,
reflecting a shift in societal expectations for adult behavior and
responsibility. Thus, understanding what causes some juvenile
delinquents to perpetrate criminal behavior after the age of 18
has significant practical implications, and is likely to be useful
for creating interventions and prevention programs that may
have the enhanced dividend of keeping some individuals from
ever going to adult prison.
In this paper we will review what is currently known about a
selection of risk factors for persistence in antisocial behavior
from adolescence into young adulthood. We say “persistence”
because few offenders initiate offending in adulthood, but r ather
continue offending from adolescence. Thus, predictors of “per-
sistence” are likely to be nearly identical to predictors of young
The most well-known theory about the development of per-
sistent antisocial behavior is Moffitt’s dual taxonomy. Moffitt
(1993) predicted that life-course-persistent offending is likely
to result from a combination of neuropsychological risks and
environmental risks. Her own review of the first ten years of
research suggests that support for this hypothesis has been quite
strong (Moffitt, 2006a). Her theory makes several predictions
related to stability of offending, early onset, and intellectual
function; these a r e among t h e r isk factors discussed below.
Crime and the Life Course
The life-course perspective in criminology emphasizes the
importance of both continuity and change in offending over
time. The most prominent contemporary theorists in this area,
Sampson and Laub (e.g., 1993), introduced the age-graded theory
of informal social control, emphasizing the role of social bonds
in the inhibition of antisocial behavior (expanding earlier work
by Hirschi, 1969). Robins (1966) had earlier observed that
while most antisocial adults were antisocial children, not all
antisocial children become antisocial adults, and Sampson and
Laub have used this fact to bolster their position that social
bonds change over time and important transitions in these
bonds can lead to dramatic changes in offending. In particular,
Sampson and Laub highlight the importance of transitions into
adulthood such as marriage, employment, or military service. If
the quality of the bonds established in these transitions is high,
they argue that these “turning points” are likely to lead to de-
sistance in antisocial conduct (Sampson & Laub, 1993). We
will review studies about the impact of work and marriage on
young adult offending below.
Other Risk Factors for Persistent Antisocial
Behavior from Adolescence to Adulthood
There are six other risk factors which we will discuss here
that do not emanate directly from the basic propositions em-
phasized in theoretical work by Moffitt or by Sampson and
J. SAVAGE ET AL.
Laub: impulsivity; school enrollment, attainment and achieve-
ment; abuse victimization; social support; poverty; deviant
peers; alcohol and drug use; and criminal justice intervention.
Measures of behavioral regulation have been used in many
studies to predict aggressive behavior and conduct problems,
including persistent conduct problems in childhood. It was
Gottfredson and Hirschi’s (1990) seminal book, propounding
the general theory of crime, which brought low self-control to
the attention of criminologists as a potential cause of long-term
antisocial behavior. As a result, “self-control” is probably the
most studied “personal factor” in the etiology of criminal be-
havior and the research on low self-control and persistent of-
fending will be discussed here. School is central to the life of
children and adolescents and may also yield an enormous im-
pact on young adult behavior. Abuse victimization in childhood
and adolescence is thought to have long-term consequences
through its traumatic effects and influence on social learning.
Social support in young adulthood might be expected to buffer
the adverse effects of stre ssors that cause offending, so it is also
included here. Situational factors, such as poverty, friendship
with deviant peers, and the abuse of alcohol or drugs are also
believed by many to play an important role in offending in the
emerging adult years.
Stability of Offending
The best predictor of current antisocial behavior is usually
some measure of previous aggressive or antisocial behavior.
Authors of virtually all the major, recent, longitudinal studies of
childhood aggression and later delinquency report evidence for
continuity over time.
Although stability has been seen by many as a “person” fac-
tor, due to a criminal “trait,” the reasons for stability are much
more complex. Some stability in antisociality is probably due to
genetic and biological factors that directly impinge on behavior
such as risk-taking. In addition, however, ongoing exposure to
an environment which elicits antisocial behavior can also cause
stable behavior (e.g., Patterson, Reid, & Dishion, 1992; Samp-
son & Laub, 1993). The exposure to environmental effects such
as weak social bonds, poor parenting, delinquent peers, and
high-crime communities is likely to be an ongoing problem
which influences criminal activity over time, regardless of indi-
vidual propensity. Moffitt (1993) suggested that persistent
criminality would arise from a combination of neuropsy-
chological deficits and a disadvantaged environment. Vila
(1994) argues that antisocial behavior is influenced by bio-
logical and developmental risk factors, but criminal propensity
is analogous to criminal “strategic styles” which take shape due
to the differential reinforcement of force, fraud, or stealthful
behavior. Finally, some authors intromit the reciprocal influ-
ence of behavior, reminding us that past behavior has an effect
on current contextual factors that might influence antisocial
behavior today (e.g., Laub & Sampson, 2003). For example
juvenile delinquency can cause parents to reject their children.
Expulsion from school can cause later unemployment. Violent
behavior can result in punishment through the criminal justice
system—which can attenuate prosocial bonds and foster
friendships with other delinquents. These contextual factors in
turn adversely influence later behavior. Wright, Caspi, Moffitt
and Silva (2001) add an interaction based on their “life-course
interdependence” view, arguing that prosocial ties, like educa-
tion, and antisocial ties, like association with deviant peers, are
li k el y t o have a greater influence on those high in criminality. The
role of factors that cause a person to be more vulnerable to risk is
likely to be significant in predicting continuing antisocial b ehavior.
Early Onset of Antisocial Behavior and Continuity across
the Transition to Adulthood
“Early onset” is the only risk factor for persistent offending
about which we conclude that a consensus has been reached.
Recognition that early onset is a strong and consistent predictor
of chronic offending has been present for a long time. That
conclusion has not changed in recent research. Fergusson, Hor-
wood, and Nagin (2000) concluded that early onset conduct
problems and early onset attention problems were associated
with chronic offending. Early acting out behavior, conduct dis-
order, age at first conviction and related problem behavior have
all been associated with later chronic offending in numerous
datasets (e.g., Blokland, 2005; Cottle, Lee, & Heilbrun, 2001).
More research is needed to assess the generalizability of the
effect across demographic groups.
Moffitt’s theory features “neuropsychological” factors and
she has operationalized these as cognitive tests including early
undercontrolled temperament, neurological abnormalities, de-
layed motor development, low intellectual ability, reading dif-
ficulties, poor memory, hyperactivity, and slow heart rate
(Moffitt & Le Blanc, 2003). Moffitt (2006b) cites a variety of
neurodevelopmental and neurocognitive factors that are dif-
ferentially associated with later membership in life-course-
persistent offending groups such as low intellectual ability,
reading difficulties, and poor scores on neuropsychological
tests of memory. She reports that persistent serious offenders in
her study showed the greatest deficits on standard neuropsy-
Many studies have reported a negative association between
intellectual function and chronic offending (e.g., Cottle et al.,
2001; Farrington, 2000; Raine, Loeber, Stouthamer-Loeber,
Moffitt, Caspi, & Lynam, 2005; Sampson & Laub, 2003). If we
limit ourselves to studies predicting offending in early adult-
hood from earlier indicators of intellectual function, there are
many whose findings are consistent with the theory that lower
intelligence predicts young adult offending (e.g., Farrington,
2000; Fergusson et al., 2000; Sampson & Laub, 1993; Werner
& Smith, 1992).
A meta-analysis by Pratt and Cullen (2000) suggests that the
association between indicators of self-control and offending is
frequently statistically significant, but weak in longitudinal st ud-
ies. We conclude that findings from studies focusing on young
adult offending are consistently supportive (e.g., DeLisi &
Vaughn, 2008; Piquero, Farrington, & Blumstein, 2007). De-
Lisi and Vaughn (2008) examined data from a sample of juve-
nile offenders and found that those identified as “career offend-
ers” (divided at the 90th percentile on the Career Criminality In-
dex) had lower self-control than other offenders. In their study,
“...low self-control was overwhelmingly the strongest predictor of
career criminality” and “far exceeded” the impact of other factors
including gender , race, and traumatic experience (p. 520) .
Lahey and Loeber (1997) point out that while several pro-
Copyright © 2013 SciRes.
J. SAVAGE ET AL.
spective longitudinal studies report that children with ADHD
exhibit antisocial behavior later in life, the analyses do not con-
trol for other conduct disorders. Because attention problems
and hyperactivity are frequently comorbid with childhood con-
duct problems, and these are fairly “stable” over time, the au-
thors concluded that studies of the independent effect of ADHD
had “failed to provide an unambiguous answer to this question”
(p. 56). In research on young adult criminality, at least two
studies have tested the predictive association of hyperactivity
on young adult criminal outcomes and found no empirical con-
nection when they controlled for conduct problems. These latter
findings suggest that the predictive value of impulsivity may be
low since the effects may be confounded with conduct prob-
lems already evident before young adultood.
Education and School
Many studies suggest that school factors are associated with
delinquency and it is common that offenders have very signifi-
cant school problems. School factors are potentially quite use-
ful because children in most industrialized countries are re-
quired to attend school, and it provides opportunities to identify
high risk individuals.
School Enrollment, Drop-Out and Educational Attainment
We might anticipate that school enrollment and attainment
would be associated with young adult offending for two princi-
pal reasons. First, being in school (post-secondary school in this
case) is likely to keep the individual occupied in a non-crimi-
nogenic environment, reduce routine activities associated with
criminal places, and foster associations with non-criminal f r ie n d s.
Second, greater school attainment is associated with many goods
related to socioeconomic status and employment opportunity,
and is likely to lead to association with non-deviant friends and
residence in a low-crime community. Being in school has been
associated with lower offending among young adults in the
Pittsburgh Youth Study (Stouthamer-Loeber, Wei, Loeber, &
Masten, 2004), and in a sample of convicted offenders (Horney,
Osgood, & Marshall, 1995). More studies would be needed to
draw a firm conclusion.
Dropout is of serious concern, but the findings have been
ambiguous. We found only a few studies of its impact on of-
fending in young adulthood and two of them support the idea
that academic attainment is significantly, negatively associated
with criminal behavior in young adulthood. At least three major
studies, however, have found contradictory evidence. For ex-
ample, in the Cambridge data, “junior school” attainment did
not distinguish high-rate chro nic offenders (Piquero et al., 2007).
More research is clearly needed to elucidate the associations
between dropout/attainment and persistence in offending.
Academic achievement has been negatively associated with
many criminal outcomes in a great many studies, including
those of chronic offending. Authors have reported significant
associations between placement in special education programs
and chronic offending (e.g., Cottle et al., 2001). Doing poorly
in school has been associated with classification into a high-rate
offending trajectory group by the end of high school in several
studies. Associations between grade point average, usually in
high school, and young adult offending have been negative and
significant in other studies as well (e.g., Arum & Beattie, 1999).
In data presented by Johnson, McGue, and Iacono (2009),
changes in antisocial behavior from age 17 to 24 were signifi-
cantly associated with age 17 GPA, in a conservative model
which also included parental SES, IQ, reading score, and age
24 educational attainment. Subsequent genetically-informed
analysis of the data suggests that age 17 GPA is largely due to
genetic factors and some non-shared environmental factors,
leading the authors to conclude that the association between
school performance and later antisocial behavior is due to indi-
vidual characteristics that exist by age 17. Because GPA is an
easy measure to acquire, its utility as a predictor is promising,
even if it may really be a proxy for genetic factors.
Severe abuse also features prominently in literature about se-
rious antisocial behavior. There is a general consensus that
maltreatment predicts delinquency and Pagani (2009) includes
corporal punishment and violence in families in her review of
family factors in the etiology of persistent criminality.
Findings on the association between abuse victimization and
young adult offending are consistent. A meta-analysis con-
ducted by Cottle et al. (2001) indicated that a history of physi-
cal or sexual abuse is associated with juvenile recidivism and
Widom (1989) found associations between early abuse and both
violent and nonviolent criminal behavior. Studies have shown
that childhood neglect and physical abuse are associated with
young adult offending, early onset and persistent delinquency
(e.g., Thornberry, Henry, Ireland, & Smith, 2010; Werner &
Smith, 1992) and that severe punishment is associated with
early onset offending. Adult violent offending has also been
associated with the extent of physical punishment in childhood
(e.g., Cohen, Kasen, Smailes, & Fagan, 2002).
Parents, families and conventional sources of support (part-
ners, friends, clergy, etc.) can wield an impact on antisocial
behavior as the adolescent becomes an adult. Social support has
many recognized benefits that may impact chronic offending.
For example, when provided on a consistent basis and in a pro-
social context, social support has the potential to strengthen
social bonds, improve psychological well being, provide access
to information and other resources, and perhaps most impor-
tantly, buffer the impact of stress and other negative life events
(e.g., Cullen & Wright, 1997; Ellis & Savage, 2009).
There are relatively few empirical studies on the impact of
social support on persistent offending, but the findings so far
suggest a promising direction for research. Bui and Morash
(2010) found that succe ssful adult female parolee s made c ha n g es
in their social networks after incarceration, dissolving relation-
ships that promoted criminal activities (including drug use), and
establishing new relationships or reestablishing ties with sup-
portive family members. Several other studies have reported
negative and significant associations between social support
and offending in young adulthood (e.g., Ellis & Savage, 2009;
Wiesner & Windle, 2004). We expect that social support might be
especially important for women and for vulnerable populations
such as re-ent ering off enders and young adults leaving foster care.
Poverty blocks access to the fundamental requirements of life
Copyright © 2013 SciRes. 3
J. SAVAGE ET AL.
(food, decent shelter, etc.) and creates severe stress. It is likely
to have an influence on offending in young adulthood, espe-
cially if the young adult does not have financial support from
family. Some authors have made the case that childhood pov-
erty can have a lasting impact on the developing child due to
financial stress in the family (reducing the amount or quality of
supervision of children and potentially lowering academic at-
tainment or achievement). Families in poverty are also often
plagued by other problems associated with delinquency such as
low parental education and residence in disadvantaged neighbor-
There are numerous studies that suggest that recidivistic,
chronic offending is most common among the poor and in poor
communities (e.g., Cottle et al., 2001; Fergusson et al., 2000;
Moffitt, 2003). In the Cambridge study, high-rate chronic of-
fenders were more likely to come from low income families
than high adolescence-peaked offenders and other offenders
(Piquero et al., 2007). Hoeve et al. (2007), however, did not
find an association between family SES and delinquency in the
long term. Poverty is one of the few factors that we might ex-
pect to predict adult-onset offending, for example in cases
where the young person is suddenly expected to be financially
independent and does not have adequate resources. This is most
likely to happen among children with weak ties to family or
We might expect that interaction with deviant peers would be
a very important influence on offending in the transition to
adulthood (e.g., Bui & Morash, 2010). A meta-analysis sug-
gests that delinquency of peers is associated with juvenile re-
cidivism (Cottle et al., 2001). Several studies of delinquency
trajectories suggest that antisocial peers are most common
among those on the highest or escalating trajectories in adoles-
cence (e.g., Ayers, Williams, Hawkins, Peterson, Catalano, &
Some scholars have proposed that peer influence is less im-
portant among chronic offenders than other types of offenders
(e.g., McGloin & Stickle, 2011; Moffitt, 1993). For example,
Moffitt proposes that life-course-persistent offenders begin of-
fending prior to the time that peers become highly influential.
However, peers may play an important role for persistent of-
fenders through co-offending, or group offending. McGloin and
Stickle (2011) observe that offending with others is often one of
the consistent markers of career criminals. Association with
deviant peers may also reinforce antisocial tendencies and may
lead to spontaneous offending, especially in the presence of
suitable targets (McGloin & Stickle, 2011). McGl oin and St ic kl e
(2011) hypothesized that chronic offenders would be “less
likely to offend because of delinquent peers, but ju st as likely to
offend with them” (p. 425). They found that chronic offenders
in their sample were less likely than nonchronic offenders to
report peer influence as the reason for their deviant behavior,
ranking it last. Yet chronic offenders were just as likely as
nonchronic offenders to engage in co-offending.
Life-course theorists have emphasized that employment
represents an important turning point in the transition to adult-
hood (Sampson & Laub, 1993). The empirical evidence on
work and young adult offending has been mixed. Most authors
anticipate a beneficial effect of employment, due to the likeli-
hood of reduced economic stress, prosocial bonds, and fewer
idle hours in the company of delinquent friends. However oth-
ers have argued that the unemployed are less likely to go out
and encounter criminal opportunities that spur spontaneous
criminal behavior. In some samples, high-rate offenders are less
likely to have a job than other subjects and job stability has
been negatively associated with recidivism, but Horney et al.
(1995) report that their subjects committed more property crime
while they were employed and one other study reports no sig-
nificant association between employment and desistance. In
light of these mixed findings, further research is clearly needed.
Important questions regarding the differential effects of em-
ployment on youth versus adults, and the potential confounding
influences of income level, criminal opportunities with co-
workers, and changes in routine activities associated with work
remain to be probed further.
In Sampson and Laub’s work, marriage is a key turning point
likely to influence desistance in young adulthood. Many au-
thors have reported a negative association between marriage
and measures of offending. Qualitative data collected by Laub
and Sampson (2003) suggest that many desisters credit mar-
riage as a major turning point in their offending careers (though
these interviews were carried out well beyond “emerging adult-
hood”). Divorce and separation were “conspicuously absent” in
their desister group. Whether this was due to an abiding bond,
as the authors have argued, or to greater supervision, is uncer-
tain (said one subject: “She won’t put up with any baloney”).
Though the preponderance of findings is supportive of Sampso n
and Laub’s hypothesis about marriage, the robustness of the
relationship has been called into question. For example, Warr
(1998) controlled for delinquent friends, and found that the
relationship between marriage and desistance was no longer
significant, bringing to light the importance of careful statistical
modeling for future research.
It is also likely that certain qualities of the marital bond af-
fect young adult offending in different ways. For example,
some authors have distinguished between early, hasty marriages,
and later marriages. The findings thus far are inconclusive on
that point. Werner and Smith (1992) found that a second mar-
riage “not infrequently” had a restraining effect, while a mar-
riage in response to a “hasty teen pregnancy” did not (p. 117).
In the Cambridge study, a recent analysis, using propensity
scores to control selection bias, found that a reduction in of-
fending occurred only among those who married in young
adulthood, ages 18 - 24, and not those who married later (Theo-
bald & Farrington, 2009). Sampson and Laub have come to
emphasize enduring marriage in their work, and there is an
emerging consensus that enduring marriage has a beneficial
effect on offending (e.g., Laub, Nagin, & Sampson, 1998); sig-
nificant associations have been found in several data sets.
Because marriage and “enduring marriage” are not randomly
distributed in the population, it is likely that any effects they
might exert are partly due to sample bias. Those who get mar-
ried and stay married may be more likely to desist from crime
due to a host of underlying characteristics or circumstances that
predict good outcomes and differ from their unmarried coun-
terparts. Thus, recent research on marriage and offending has
employed two approaches to address this problem. One way to
Copyright © 2013 SciRes.
J. SAVAGE ET AL.
reduce sample selection biases is to control for “propensity” to
marry (and stay married). Farrington and West (1995) used a
case control design on a small sample of men who got married
in the Cambridge Study who could be matched very closely on
number of prior convictions with other subjects who remained
unmarried for at least five years. Although the married and
unmarried men had incurred the same number of convictions
before marriage, those who married had significantly lower
convictions in the following five years than those who did not
get married. In fact, the conviction rate for the unmarried men
was close to two times that for the married individuals when
they extended the follow-up to age 32 (or until the married
individuals were separated from their spouses). The design is
clever, but does not eliminate the possibility that the married
and unmarried subjects differed in other important ways. In
another study, propensity score matching was used to disentan-
gle effects of selection bias with the same result (King, Masso-
glia, & MacMillan, 2007).
Another way to address the sample selection problem is to
use a within-subjects design and examine the effects of being
married or unmarried on individual offending over time. Stud-
ies using these designs have, on the whole, found that being
married is associated with lower offending in young adulthood.
The weakness of these designs is that they must eliminate rival
hypotheses linking “being married” in a given year and offend-
ing. While it is easy to control for factors such as income or age,
such studies have not adequately controlled for factors such as
“maturity” which might cause both successful marriage and
desistance from antisocial behavior. In a very complex analysis
of the Glueck and Glueck follow-up sample, employing “in-
verse probability of treatment weighting” (IPTW) to control for
confounds related to selection bias in married subjects as well
as within-subjects sources of bias, Sampson, Laub and Wimer
(2006) estimated an average reduction of approximately 35% in
the odds of committing a crime when a given subject was mar-
ried compared to not married. More research employing pro-
pensity score matching and within-subjects designs is needed to
test the tightly confined research question (is it really being
married that changes the behavior?). Nonetheless, marriage as
observed in the population, encompassing all the reasons a
person gets married and stays married, appears to be associated
with lower rates of offending.
Drug and Alcohol Abu se
Many authors have expected that drug and alcohol abuse in
the transition to adulthood would increase offending. Although
early reviews reported that the evidence on associations be-
tween drug use and criminality were not conclusive, emergent
research is leaning in the direction of an association between
alcohol and drug abuse and persistent offending (e.g., Benda et
al., 2001; Stouthamer-Loeber et al., 2004). The effect in this
particular life period may be especially potent since drug use
disorders peak during this time, drug use does not decline as
precipitously as other types of criminality, and, in the US and
many other cultures, emerging adults are awarded the legal
right to drink alcohol as freely as they wish.
In Laub and Sampson’s (2003) qualitative data, alcohol a buse
was common in the life histories of persistent offenders. Quan-
titative studies have reported statistically significant associa-
tions between heavy episodic drinking and fighting, alcohol use
and chronic offending, and adolescent drunkenness and young
Findings by Horney et al. (1995) indicate that the association
may be confounded. They used a retrospective month-by-month
survey to examine the association between certain life circum-
stances and offending in a sample of Nebraska serious offend-
ers. Controlling for other factors such as whether or not the
offender was on probation, in school, employed, and living
with a wife, “heavy drinking” was not significantly associated
with criminal behavior. Authors of at least one other recent
study have reported similar findings. Hussong, Curran, Moffitt,
Caspi, and Carrig (2004), who entitle their paper, “Substance
abuse hinders desistance...” used a time-varying covariate model
to predict the future trajectory of criminal non-drug-related
antisocial behavior, and estimate whether or not alcohol was
associated with an upward departure from that trajectory. They
found that men with more symptoms of alcohol abuse reported
significantly higher antisocial behavior than expected at ages 18
and 21; the effect was marginally significant at age 26 as well.
Thus, alcohol abuse is a risk marker for young adult offending,
but more research is needed to understand the amount of alco-
hol associated with the effect and whether the association is
fully mediated by other factors.
Illicit Drug Use
On the balance, the preponderance of studies have shown
fairly convincing support for a drug use-persistence association
(e.g., Horney et al., 1995; Morizot & Le Blanc, 2007; Schroe-
der, Giordano, & Cernkovich, 2007; Wiesner & Windle, 2004),
but some contradictory findings leave some questions unan-
swered. First, it is not yet clear whether drug use predicts on-
going serious crime, or is limited mainly to drug crime and
drug-related crime. It is also likely that different drugs and
different levels of drug use affect offending in young adulthood
in different ways, but such a small number of studies examine
specific drug types, or distinguish between heavy and casual
use, that no conclusions can be drawn at this time.
Behavior an d I t s C o ns e quences
Criminal Justice Intervention for Prior Delinque nc y
Unfortunately, while just about every response to juvenile
offending is intended to reduce the likelihood of future offend-
ing, evidence suggests that criminal justice responses often re su lt
in a greater likelihood of offending. Criminal justice interven-
tion has many “collateral consequences” including interruption
of education, possible loss of employment, and attenuation of
prosocial bonds (e.g., Laub & Sampson, 1993). A large number
of empirical studies over many decades have reported that offi-
cial sanctions or severity of sanctions are associated with a
greater likelihood of future offending. For example, in a two-
year follow-up of 414 adolescents (age 17 years), having prior
incarcerations was one of the best predictors of entry into the
adult correctional system (Benda et al., 2001). The effect is not
limited to serious sanctions like incarceration, but has even
included adverse effects of mere adjudication. Those most lik ely
to offend in young adulthood are those who committed offenses
in adolescence and we earlier discussed the many propositions
about stability. In contrast with factors that are largely out of
the control of policy makers such as family income and intel-
Copyright © 2013 SciRes. 5
J. SAVAGE ET AL.
lectual ability, criminal justice interventions are completely
amenable to manipulation. They should be used as a tool to
interrupt the continuity of offending, and yet the findings sug-
gest that, on average, they exacerbate the problem. Further
research is needed to develop a comprehensive understanding
of system effects, by intervention type, in order to identify areas
where policies can be changed to minimize future offending.
This review points to several factors consistently associated
with persistent antisocial behavior in emergent adulthood (e.g.,
previous antisocial behavior, early onset, intelligence, academic
achievement, abuse victimization, enduring marriage, peer devi-
ance, drug abuse, and criminal justice intervention). But it is
also clear that some important questions remain. We highlight
the ones of greatest concern to us. First, despite a great deal of
attention to the construct of low self-control, there is a strong
possibility that the findings have been confounded. Future re-
search must disentangle the effects of low self-control to ensure
that associations between impulsivity and offending are not
merely refle cting stabil ity in c onduct problems. The findings on
high school dropout and school attainment are also much less
consistent than we might expect. It could be the case that the
beneficial effect of education on offending tops out at some
point shortly after high school, but the association is not yet
clear. Findings on the role of drug and alcohol abuse suggest
that they slow desistance, though the mixed findings on the
topic also suggest that the extent of use and the type of drug
used may be important areas for further inquiry. Similarly,
while the association between “enduring marriage” and young
adult offending is fairly clear, other questions about marriage
and offending remain related to the age at which marriage can
have a beneficial effect, the impact of antisocial behavior of the
spouse, and race and sex differences. Further, controls for
maturation in within-subjects designs are needed as are more
studies correcting for selection biases.
The unexpected findings on work and young adult offending
should encourage further research on this topic. Numerous
studies, not reviewed here, show that adolescents who work are
more, not less, prone to offend, but theory predicting an asso-
ciation between having a good jo b and desi stance remains h i g hl y
compelling. Thus, examining age-by-age effects, across various
circumstances, would be of keen interest, as would understand-
ing the way criminal opportunities afforded by work may offset
Poverty is one of the few factors that we believe could help
explain adult-onset offending which we would predict might
occur for youths with weak ties to parents, such as juvenile
offenders and those exiting foster care. Social support might be
something that could be leveraged by the syste m to miti gate the
many stressors in those most vulnerable at this age.
We feel that a very important focus of future research on
persistent offending into adulthood should be on high risk popu-
lations such as youthful offenders and foster children. So far,
research suggests that, rather than serving as a deterrent, crimi-
nal justice interventions frequently have adverse effects on de-
linquents, so research shedding light on the effects of specific
interventions, or on protective factors (such as social support)
that mitigate the effects of involvement in the juvenile justice
system, is very impo rtant.
Arum, R., & Beattie, I. (1999). High school experience and the risk of
adult incarceration. Criminology, 37, 515-539.
Ayers, C. D., Willia ms, J. H., Hawkins, J. D., Peterson, P. L., Catalano,
R. F., & Abbott, R. D. (1999). Assessing correlates of onset, escala-
tion, deescalation, and desistance of delinquent behavior. Journal of
Quantitative Criminology, 15, 277-306.
Benda, B. B., Corwyn, R. F., & Toombs, N. J. (2001). Recidivism
among adolescent serious offenders: Prediction of entry into the cor-
rectional system for adults. Criminal Justice and Behavior, 28, 588-
Blokland, A. (2005). Crime over the lifespan: Trajectories of criminal
behavior in Dutch offenders. Leiden, Netherlands: Institute for the
Study of Crime and Law Enforcement.
Bui, H. N., & Morash, M. (2010). The impact of network relationships,
prison experiences, and internal transformation on women’s success
after prison release. Journal of Offender Rehabilitati on, 49, 1-22.
Cohen, P., Kasen, S., Smailes, E., & Fagan, J. (2002). Childhood ante-
cedents of adolescent and adult crime and violence, final report
(Grant Number 1999-IJ-CX-0029). Washington DC: National Insti-
tute of Justice.
Cottle, C. C., Lee, R. J., & Heilbrun, K. (2001). The prediction of cri-
minal recidivism in juveniles: A meta-analysis. Criminal Justice and
Behavior, 28, 367-394. doi:10.1177/0093854801028003005
Cullen, F. T., & Wright, J. P. (1997). Liberating the anomie-strain
paradigm: Implications from social support theory. In N. Passas, & R.
Agnew (Eds.), The future of anomie theory (pp. 187-206). Boston,
MA: Northeastern University Press.
DeLisi, M., & Vaughn, M. G. (2008). The Gottfredson-Hirschi cri-
tiques revisited: Reconciling self-control theory, criminal careers,
and career criminals. International Journal of Offender Therapy and
Comparative Criminology, 52, 520-537.
Ellis, S., & Savage, J. (2009). Strain, social support and persistent
criminality. In J. Savage (Ed.), The development of persistent crimi-
nality (pp. 71-89). New York: Oxford University Press, Inc.
Farrington, D. P. (2000). Psychosocial predictors of adult antisocial
personality and adult convictions. Behavior Sciences & the Law, 18,
Farrington, D. P., & West, D. J. (1995). Effects of marriage, separation
and children on offending by adult males. In Z. S. Blau, & J. Hagan
(Eds.), Current perspectives on aging and the life cycle: Vol. 4. De-
linquency and disrepute in the life course (pp. 249-281). Greenwich,
CT: JAI Press.
Fergusson, D. M., Horwood, L. J., & Nagin, D. S. (2000). Offending
trajectories in a New Zealand bir th cohort. Criminology, 38 , 525-552.
Gottfredson, M., & Hirschi, T. (1990). A general theory of crime.
Stanford, CA: Stanford University Press.
Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University of
Hoeve, M., Smeenk, W., Loeber, R., Stouthamer-Loeber, M., van der
Laan, P. H., Gerris, J. R. M., & Dubas, J. S. (2007). Long-term ef-
fects of parenting and family characteristics on delinquency of male
young adults. European Journal of Criminology, 4, 161-194.
Horney, J., Osgood, D. W., & Marshall, I. H. (1995). Criminal careers
in the short-term: Intra-individual variability in crime and its relation
to local life circumstances. American Sociological Review, 60, 655-
Hussong, A. M., Curran, P. J., Moffitt, T. E., Caspi, A., & Carrig, M. M.
(2004). Substance abuse hinders desistance in young adult’s antiso-
cial behavior. Development and Psychopathology, 16, 1029-1046.
Copyright © 2013 SciRes.
J. SAVAGE ET AL.
Copyright © 2013 SciRes. 7
Johnson, W., McGue, M., & Iacono, W. G. (2009). School performance
and genetic and environmental variance in antisocial behavior at the
transition from adolescence to adulthood. Developmental Psychology,
45, 973-987. doi:10.1037/a0016225
King, R. D., Massoglia, M., & MacMillan, R. (2007). The context of
marriage and crime: Gender, the propensity to marry, and offending
in early adulthood. Crimi nol og y, 45, 33-66.
Lahey, B. B., & Loeber, R. (1997). Attention-deficit/hyperactivity dis-
order, oppositional defiant disorder, conduct disorder, and adult an-
tisocial behavior: A life span perspective. In D. M. Stoff, J. Breiling,
& J. D. Maser (Eds.), Handbook of antisocial behavior (pp. 51-59).
New York: John Wiley.
Laub, J. H., Nagin, D. S., & Sampson, R. J. (1998). Trajectories of
change in criminal offending: Good marriages and the desistance
process. American Sociological Review, 63, 225-238.
Laub, J. H., & Sampson, R. J. (1993). Turning points in the life course:
Why change matters to the study of crime. Criminology, 31, 301-
Laub, J. H., & Sampson, R. J. (2003). Shared beginnings, divergent
lives: Delinquent boys to age 70. Cambridge, MA: Harvard Univer-
McGloin, J. M, & Stickle, W. P. (2011). Influence or convenience?
Disentangling peer influence and co-offending fo r chronic off enders.
Journal of Research in Crime and Delinquency, 48, 419-447.
Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent
antisocial behavior: A developmental taxonomy. Psychological Re-
view, 100, 674-701. doi:10.1037/0033-295X.100.4.674
Moffitt, T. E. (2003). Life-course-persistent and adolescence-limited
antisocial behavior: A 10-year research review and a research agenda.
In B. B. Lahey, T. E. Moffitt, & A. Caspi (Eds.), Causes of conduct
disorder and juvenile delinquency (pp. 49-75). New York: Guilford.
Moffitt, T. E. (2006a). A review of research on the taxonomy of life-
course persistent versus adolescence-limited antisocial behavior. In F.
T. Cullen, J. P. Wright, & K. Belvins (Eds.) Taking stock: The status
of criminological theory: Vo l. 15 (pp. 277-311). New Brunswick, NJ:
Moffitt, T. E. (2006b). Life-course persistent versus adolescence-li-
mited antisocial behavior. In D. Cicchetti, & D. Cohen (Eds.), De-
velopmental psychopathology: Vol. 3. Risk, disorder and adaptation
(pp. 570-598). New York: Wiley.
Morizot, J., & Le Blanc, M. (2003). Continuity and change in personal-
ity traits from adolescence to midlife: A 25-year longitudinal study
comparing representative and adjudicated men. Journal of Personal-
ity, 71, 705-755. doi:10.1111/1467-6494.7105002
Pagani, L. S. (2009). The influence of family context on the develop-
ment and persistence of antisocial behavior. In J. Savage (Ed.), The
development of persistent criminality (pp. 37-53). New York: Oxford
University Press. doi:10.1093/acprof:oso/9780195310313.003.0002
Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). A social learning
approach: Vol. 4. Antisocial boys. Eugene, OR: Castalia.
Piquero, A. R., Farrington, D. P., & Blumstein, A. (2007). Key issues in
criminal career research: New analyses of the Cambridge study in
delinquent development. Cambridge: Cambridge University Press.
Pratt, T. C., & Cullen, F. T. (2000). The empirical status of Gottfredson
and Hirschi’s general theory of crime: A meta-analysis. Criminology,
38, 931-964. doi:10.1111/j.1745-9125.2000.tb00911.x
Raine, A., Loeber, R., Stouthamer-Loeber, M., Moffitt, T. E., Caspi, A.,
& Lynam, D. (2005). Neurocognitive impairments in boys on the
life-course persistent antisocial path. Journal of Abnormal Psychol-
ogy, 114, 38-49. doi:10.1037/0021-843X.114.1.38
Robins, L. (1966). Deviant children grown up. Baltimore, MD: Wil-
liams & Wilkins.
Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways
and turning points through life. Cambridge, MA: Harvard University
Sampson, R. J., Laub, J. H., & Wimer, C. (2006). Does marriage reduce
crime? A counterfactual approach to within individual causal effects.
Criminology, 44, 465-508. doi:10.1111/j.1745-9125.2006.00055.x
Savage, J. (2009). Understanding persistent offending: Linking devel-
opmental psychology with research on the criminal career. In J.
Savage (Ed.), The development of persistent criminality (pp. 3-36).
New York: Oxford University Press.
Schroeder, R. D., Giordano, P. C., & Cernkovich, S. A. (2007). Drug
use and desistance proces s e s. Criminology, 45, 191-222.
Stouthamer-Loeber, M., Wei, E., Loeber, R., & Masten, A. S. (2004).
Desistance from persistent serious delinquency in the transition to
adulthood. Development and Psychopathology, 16, 897-918.
Theobald, D., & Farrington, D. P. (2009). Effects of getting married on
offending. European Journal of Criminology, 6, 496-516.
Thornberry, T. P., Henry, K. L., Ireland, T. O., & Smith, C. A. (2010).
The causal impact of childhood-limited maltreatment and adolescent
maltreatment on early adult adjustm ent. Journal of Adolescen t Health, 46,
Vila, B. (1994). A general paradigm for understanding criminal behav-
ior: Extending evolutionary ecological theory. Criminology, 32, 311-
Warr, M. (1998). Life-course transitions and desistance from crime.
Criminology, 36, 183-216. doi:10.1111/j.1745-9125.1998.tb01246.x
Werner, E. E., & Smith, R. S. (1992). Overcoming the odds: High risk
children from birth to adulthood. Ithaca, NY: Cornell University
Widom, C. S. (1989). Child abuse, neglect, and violent criminal behav-
ior. Criminology, 27, 251-271.
Wiesner, M., & Windle, M. (2004). Assessing covariates of adolescent
delinquency trajectories: A latent growth mixture. Journal of Youth
and Adolescence, 33, 431-442.
Wright, B. R. E., Caspi, A., Moffitt, T. E., & Silva, P. A. (2001). The
effects of social ties on crime vary by criminal propensity: A life-
course model of interdepe n d en c e. Criminology, 39, 321-351.