Research on Pathways
to Desistance

Research Update Created for the
Fourth Annual Models for Change
National Working Conference
December 8-9, 2009

www.modelsforchange.net
An initiative supported by John D. and Catherine T. MacArthur Foundation

www.modelsforchange.net

www.macfound.org

Pathways to Desistance
A First Look at Emerging Findings
Juveniles can commit serious and sometimes violent
offenses, from felony burglary to murder. Their
crimes are appalling. But only a minority of these
serious offenders will go on to a lifetime of repeated
crime; the rest will have relatively little involvement
in illegal activity. Nevertheless, they present a
challenge to the juvenile justice system. And they
fuel an ongoing debate among professionals and in
society at large.
The sanctions applied to serious juvenile offenders
vary widely from one jurisdiction to the next. Even
within a given jurisdiction, different youths may
receive very different sanctions for similar offenses.
Some will spend time in a secure facility, where they
may (or may not) receive a range of services. Some
are put on probation, with or without a treatment
program. Others are sentenced to community service.
How do these sanctions and services affect the
trajectories of the offenders’ lives? And what
measures could put them on a more positive path?

The ongoing study already has published some important
findings:
•	 Adolescents who become involved in serious
crimes are not a particular “type” but a
heterogeneous group, much like their nonoffending peers. The differences among them
are seldom considered by courts, nor are they
translated by service providers into different types
of intervention.
•	 Nothing in the basic psychological or social
characteristics of these adolescents strongly
predicts which will go on to a lifetime of crime
and which will curtail their offending after
court involvement.
•	 Longer stays in juvenile facilities do not appear
to reduce offending. However, continued
probation supervision and community-based
services provided after a youth is released do
make a difference, at least in the six months
following release.
•	 Substance abuse is a major factor in continued
criminal activity. Treating substance abuse can
reduce subsequent offending.

Answering these questions could help us write better

Pathways is a unique study in the field of juvenile

policy, make better use of scarce resources, and better

justice: in its goals, its subjects, its longitudinal

serve both the young offenders and the communities

nature, its comprehensive scope, and its findings.

in which they live.

These and other aspects of the project are explored in

That is the impetus behind Pathways to Desistance,

more detail on the following pages.

a large, multi-site, collaborative project following
1,354 juvenile offenders for seven years after their
conviction.1 Pathways is the most intense look to date
at the outcomes of sanctions and services – feedback
that is rarely available to decision-makers in the
juvenile justice system.
1

What broad questions is Pathways exploring?
The investigators are trying to unravel what it is

They are an ethnically diverse group: 25 percent

that reduces the severity and frequency of criminal

Hispanic, and 2 percent “other.” While the majority

activity among the majority of young offenders –

are male, 13 percent are female.

White, 44 percent African-American, 29 percent

or in any individual youth. Is there a relationship
between the sanctions and services they receive and
the future trajectory of their lives? Do other factors,
internal or external, account for the difference?

Despite their involvement in serious crime, these
adolescents are not uniformly “bad” kids on the
road to a lifetime of criminal activity.

What can we learn about these adolescents that will
allow the juvenile justice system to respond more
effectively and prevent future criminal activity?
How could the results of the study be used?
The intent of the Pathways team is to provide

How is the study conducted?
Investigators interview the adolescents, their family
members, and their friends at specific time points
for seven years after their conviction. The team has
completed about 90 percent of the interviews with

policymakers and practitioners with analyses that can

the subjects – more than 25,000 in all. Sadly, a high

guide the development of a more rational, effective,

proportion of this group (about 4 percent) died during

and developmentally appropriate juvenile justice

these years, most of them victims of homicide.

system. Some findings will be relevant to decisions
made at disposition, such as what kinds of placement
or community-based services are most appropriate for
which individuals, and for how long. Some could be
used to improve institutional and community-based
services, or to determine at what point in the process
they should be offered. Some findings may suggest
the need to involve families, schools, and other
institutions in new or existing solutions.
Who are the subjects of the study?
The study has followed a sample of juveniles who

What sort of information is collected?
Over seven years, Pathways has constructed the
richest source of information currently available
about how serious adolescent offenders mature and
what happens to them as they make the transition
from adolescence to early adulthood.2,3 The interviews
cover a wide range of topics: psychological
development, mental health, behavior, attitudes,
family and community context, and relationships.
The investigators also gather month-by-month
information on significant life events such as living

committed major offenses when they were 14 to

arrangements, employment, education, romantic

17 years old. Their crimes were the most serious

relationships, and involvement with the legal and

felonies that come before the court, including murder,

social service systems.

robbery, aggravated assault, sex offenses, and
kidnapping. About 70 percent of the subjects have
had one or more prior petitions to court. Nearly onefifth were processed in the adult system.
2

Using official records from the Federal Bureau of

The other half received community-based sanctions or

Investigation and local courts, the investigators

services. In some cases, this meant simply checking

document arrests and the sanctions and services the

in with a probation officer. Others received more

young offenders receive.

intensive supervision. And others received a mix

As of August, 2009, the working group and its
collaborators had published about 30 papers and made
almost 100 presentations based on these data.
Are the serious offenders markedly different
from other young people?
Not necessarily. There is considerable variability

of probation and services such as group sessions,
substance abuse treatment, or family therapy.
Do most of them continue to commit crimes?
Despite their involvement in serious crime, these
adolescents are not uniformly “bad” kids on the road
to a lifetime of criminal activity. In fact, the majority

among these adolescents – as there is in any group

reported engaging in few or no illegal activities after

of adolescents – in their background characteristics,

court involvement.

their family environments, and their attitudes towards

In one set of analyses, the investigators identified

the law.

4,5,6

They appear, in many ways, very similar

five distinct groups that followed different patterns of

to other adolescents in their communities, both

change in their illegal activity over the first 36 months

psychologically and socially, and they develop along

of follow-up (see Figure 1).7

similar pathways.
Trajectories of Criminal Behavior

There is a higher incidence among these adolescents
of certain risk markers, such as substance abuse,
parents who offended, and being in special education.
But we know from other studies that rates of
offending among adolescents in general are much
higher than rates of being caught or convicted; many
of the adolescents in this study may be in the system

"C

8

~
0

D.

CI

~

.5

Gi

s::::

.,!. "C

en
....
0
Gi
>
CIl

~

6

4
2

..J

12

largely because, unlike their peers, they were caught.

18

24

30

36

onths After Initial Interview
Figure 1

What sanctions did these adolescents receive
for their offenses?
About half of the offenders were placed in juvenile or

Three of these groups showed different, but relatively

adult facilities. These ranged from jails and prisons,

at a high level of offending and dropped off to a very

to boot camps, to institutional settings that look much

low level. The fifth group – just 8.5 percent of the total

like a high school or college campus. Some of these

– started at a high level of offending and continued at

were therapeutically intense settings, while others

this high level.

stable, low levels of offending. A fourth group started

offered fewer services.
3

In short, the vast majority of the adolescents in the

Institutional placements are widely used with

study reported very limited involvement in illegal

this group of offenders. Even those adolescents

activity in the three years following their court

who reported very low levels of involvement in

involvement. A strikingly large proportion of those

subsequent antisocial activity nevertheless spent a

who had been offending at high levels dramatically

sizable proportion of the follow-up period (about 30

reduced their illegal activities during the first few

percent of this time) in institutional care. However,

years of the study.

among these “low level” offenders, institutional
placement raised the level of offending by a small, but

Is it possible to predict which adolescents
will fall into which group?
The ability to predict future criminal activity would
be of great interest to the courts where youths are
adjudicated, and the study sought to answer this
important question. Unfortunately, despite the
detailed information collected on these adolescents,
predictions were elusive.

statistically significant, amount.
The Pathways findings highlight the need to
reconsider the costs and benefits of locking
up adolescents, even those committing the
most serious crimes.
This may mean that expensive institutional
placements are often being used in cases where there

The initial baseline interview of Pathways gathers

is little need for such an investment – and where

far more information about juveniles’ personality,

it may in fact be counterproductive. It is worth

behavior, history, and life circumstances than is

considering whether the general pattern of locking

typically available to any court. Yet the investigators

up such a large proportion of these adolescents is

found that baseline characteristics – demographics,

producing many positive results.

psychosocial characteristics, attitudes, and even
differentiating the “persisters” from the “desisters.”

Given the prominent role of institutional
placement in the juvenile justice system, does
the Pathways study suggest any guidelines?

The likelihood of effectively distinguishing them at

It offers food for thought. A key question for juvenile

disposition is low.

justice policy is whether and how long to keep

prior offending history – are not very useful for

juvenile offenders in facilities in order to maximize
Does desistance from crime match the
treatment offenders receive?
It doesn’t seem to. The investigators found, to their
surprise, that persisters and desisters received about
the same type of treatment in the justice system: they
spent, on average, the same amount of time in the
same types of institutions.

4

both public safety and the therapeutic benefits to
the offender. The study addressed this question first
by grouping subjects with very similar background
characteristics and comparing those who had been
placed in institutions to those given probation. They
examined whether institutional placement led to any
reduction in the rates of re-arrest or self-reported

illegal activity, and found that both groups were
equally likely or unlikely to re-offend. On these
measures, at least, institutional placement appears to
have no advantage to over probation.

Is there anything we can learn about individual
offenders that might be relevant to deciding
what sanctions and services they receive?
Sanctions are often meted out on the basis of the
severity of the crime and the number of prior

The investigators also looked at the length of stay for

convictions. Courts do not generally assess individual

those in juvenile institutional care, to see if there was

needs in depth and match them to particular services.

any marginal gain from longer stays.8 Again, they
found no significant benefits from a longer stay (see
Figure 2).

The Pathways study suggests that it might be feasible
to do so. The study found that individual adolescent
offenders differ substantially from one another on
a number of relevant dimensions: parenting styles,
social development, the timing of psychological
development, mental health, attitudes toward the

~

1.4

" &!

1.0

.-----!1.18l-f-----1~L1.12
~61---g_-----,
~
~-~-

£~1.21
~~

:. ~ 0.8

~ ~
c:( 8

-.............

LO.

consideration of these differences by the courts and

92

service providers could lead to more tailored, more

~

0.2

+------------------------1
+------------------------1
+------------------------1
+------------------------1

0.0

+-

; >- 0.6
a.:=:
.,t!! § 0.4

law, and the level of substance abuse. Closer

effective services.

~----~-----~-------I
0-6

6-10

10-13

> 13

Is substance abuse treatment a good example
of the potential of tailored services?

Months in Juvenile Placement

Figure 2

It’s one of the clearest examples. Substance use is

Institutional placement is a very costly undertaking.

strongly related to continued criminal activity in this

The Pathways findings highlight the need to

group, and it makes sense to focus on this behavior

reconsider the costs and benefits of locking up

for intervention. In fact, the study shows that

adolescents, even those committing the most serious

treatment for substance use can reduce offending.

crimes. As part of that reconsideration, however, we
need to carefully examine the effects of treatments
provided in institutional care; determine which, if any,
make a difference in behavior for which offenders;
and consider whether they can be offered with equal
success in a community setting.

Levels of substance use and associated problems are
very high in these young offenders. More than onethird qualify for a diagnosis of substance use disorder
in the year prior to the baseline interview, and over
80 percent report having used drugs or alcohol during
the previous six months. Moreover, the level of
substance use walks in lockstep with illegal activity

Closer consideration of differences among
offending youths could lead to more tailored,
more effective services.

over the follow-up period: more substance use, more
criminal offending.9

5

Adolescent offenders with substance use disorders
don’t always receive treatment for these problems.
When they do, though, it appears to work. The

Ongoing substance use treatment for serious
juvenile offenders appears to pay off. The key
is including family in the intervention.

investigators examined treatment-related reductions
in alcohol and marijuana use, cigarette smoking,

The intensity of community-based services for the

and non-drug offending during the first year post-

returning offenders was generally low – supervision

treatment.10 Although the study doesn’t test any single,

was more common than involvement with treatment-

new intervention for substance use, it does provide an

oriented agencies. Nevertheless, the analyses showed

excellent opportunity to examine how well the standard

that when adolescents did receive supervision and

treatments affect later adjustment. Results indicate

were involved in community-based services, they

that drug treatment significantly reduced substance use

were significantly more likely to avoid further

for about six months, and that this reduction was more

involvement with the legal system and to attend

than simply an effect of the adolescents being locked

school or work more regularly. Continued aftercare

up in a controlled environment. Subsequent criminal

supervision and service involvement in the six months

offending also was reduced – but only when treatment

after institutional placement, as delivered in the real

included family involvement.

world, appear to have a positive effect.

The bottom line: ongoing substance use treatment for

These results highlight the importance of investing

serious juvenile offenders appears to pay off, but the

resources in community-based aftercare programs.

key is including family in the intervention.

Though institutional care in general seems to have a
limited impact on later criminal activity, establishing

What happens to young offenders after
they’re released from an institution? Can
services at that point make a difference?
The study indicates that aftercare services do make a
difference.
Because the project collects monthly data about
institutional placement, probation, and involvement
in community-based services, investigators were
able to examine the effects of aftercare services for
the six months after a court-ordered placement (the

a wider array of sanctions and services might well
produce more positive outcomes.
Where does the Pathways project go from here?
The findings presented here are just a first look at the
potential of the Pathways study. Much work is yet to
be done on the dynamic nature of these adolescents’
lives, the factors that promote positive adjustment
during late adolescence and early adulthood, and the
effectiveness of the juvenile justice system.

period for such services in many locales). They

The investigators believe that the greatest potential

looked specifically at the effects these services had on

lies in examining the diversity of the subjects – the

community adjustment: going to school or working,

differences in their backgrounds, personalities,

getting arrested or placed back in a facility, or self-

development, attitudes, and responses – and how

reported illegal activity.

those differences relate to various interventions and

11

6

outcomes. If serious offenders were a homogeneous

The Pathways project is doing more than answering

group, it would make sense to simply link sanctions

specific questions. It is creating a huge database – a

to the severity of the crime and hope for a uniform

research infrastructure – that will be freely accessible

reduction in future offending. But given their

to researchers outside the project. With continuing

diversity, there may be other ways to group young

progress, in policy and in research, we can have

offenders and apply interventions that produce better

a fairer, more effective, and more cost-effective

results, for them and for society.

juvenile justice system.

Another promising line of inquiry appears to be
identifying the positive life events that occurred
during the follow-up period. Preliminary results
suggest adolescents who are better integrated into
the community by a stable living situation and more
routine lives, by school or employment, or by a
positive romantic relationship are less likely to be
involved in illegal activity. Whether there is a causal
relationship remains to be tested.
For more information on Pathways to Desistance, please write to the project coordinator, Carol Schubert, at
schubertca@upmc.edu.
	 The study grew out of the efforts of the MacArthur Foundation Research
Network on Adolescent Development and Juvenile Justice. It is funded
by the John D. and Catherine T. MacArthur Foundation, the Office of
Juvenile Justice and Delinquency Prevention (OJJDP), the National
Institute of Justice (NIJ), the Robert Woods Johnson Foundation,
the William Penn Foundation, the William T. Grant Foundation, the
Pennsylvania Commission on Crime and Delinquency, the Arizona
Juvenile Justice Commission, and the National Institute on Drug Abuse.
A multidisciplinary working group of investigators have collaborated
on the study since its inception and continue to analyze data and publish
findings. In alphabetical order, they are Robert Brame, Ph.D., Elizabeth
Cauffman, Ph.D., Laurie Chassin, Ph.D., Jeffrey Fagan, Ph.D., George
Knight, Ph.D., Edward Mulvey, Ph.D., Sandra Losoya, Ph.D., Alex
Piquero, Ph.D., Carol Schubert, M.P.H., and Laurence Steinberg, Ph.D.
2
	 Mulvey, E. P., Steinberg, L., Fagan, J., Cauffman, E., Piquero, A. R.,
Chassin, L., et al. (2004). Theory and Research on Desistance from
Antisocial Activity among Serious Adolescent Offenders. Youth
Violence and Juvenile Justice, 2 (3), 213-236.
3
	 Schubert, C. A., Mulvey, E.P., Steinberg, L., Cauffman, E., Losoya, S.,
Hecker, T., Chassin, L., et al. (2004). Operational Lessons from the
Pathways to Desistance Project. Youth Violence and Juvenile Justice, 2
(3), 237-255.
4
	 Cauffman, E., Piquero, A. R., Kimonis, E., Steinberg, L., Chassin, L.,
& Fagan, J. (2007). Legal, individual, and environmental predictors of
court disposition in a sample of serious adolescent offenders. Law &
Human Behavior, 31, 519-535.
5
	 Steinberg, L., Blatt-Eisengart, I., & Cauffman, E. (2006). Patterns of
competence and adjustment among adolescents from authoritative,
1

authoritarian, indulgent, and neglectful homes: A replication in a sample
of serious adolescent offenders. Journal of Research on Adolescence,
16 (1), 47-58.
6
	 Piquero, A. R., Fagan, J., Mulvey, E. P., Steinberg, L., & Odgers, C.
(2005). Developmental trajectories of legal socialization among serious
adolescent offenders. The Journal of Criminal Law and Criminology,
96 (1), 267-298
7
	 Mulvey, E. P., Steinberg, L., Piquero, A. R., Besana, M., Fagan, J.,
Schubert, C. A., Cauffman, E. (in press). Longitudinal offending
trajectories among serious adolescent offenders. Development &
Psychopathology.
8
	 Loughran, T., Mulvey, E. P., Schubert, C. A., Fagan, J., Losoya, S.
H., Piquero, A. R. (2009). Estimating a Dose-Response Relationship
between Length of Stay and Future Recidivism in Serious Juvenile
Offenders. Criminology, 47, 699-740.
9
	 Mulvey, E.P., Schubert, C.A., Chassin, L. (in press) Substance use
and offending in serious adolescent offenders. Washington, DC:
United States Department of Justice, Office of Juvenile Justice and
Delinquency Prevention.
10
	Chassin, L., Knight, G., Vargas-Chanes, D., Losoya, S., Naranjo, D.
(2009). Substance use treatment outcomes in a sample of male serious
juvenile offenders. Journal of Substance Abuse Treatment, 36(2), 183-194.
11
	Chung, H. L., Schubert, C. A., Mulvey, E.P. (2007) An empirical
portrait of community reentry among serious juvenile offenders in two
metropolitan cities. Criminal Justice and Behavior, 34 (11), 1402-1426.

7