American Academy of Political and Social Science

American Academy of Political and Social Science
Scared Straight and Other Juvenile Awareness Programs for Preventing Juvenile
Delinquency: A Systematic Review of the Randomized Experimental Evidence
Author(s): Anthony Petrosino, Carolyn Turpin-Petrosino and John Buehler
Source: Annals of the American Academy of Political and Social Science, Vol. 589, Misleading
Evidence and Evidence-Led Policy: Making Social Science More Experimental (Sep., 2003), pp.
41-62
Published by: Sage Publications, Inc. in association with the American Academy of Political and
Social Science
Stable URL: http://www.jstor.org/stable/3658560 .
Accessed: 08/02/2014 11:52
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Sage Publications, Inc. and American Academy of Political and Social Science are collaborating with JSTOR
to digitize, preserve and extend access to Annals of the American Academy of Political and Social Science.
http://www.jstor.org
This content downloaded from 193.205.136.30 on Sat, 8 Feb 2014 11:52:22 AM
All use subject to JSTOR Terms and Conditions
Scared Straight
and Other
Juvenile
Awareness
Programs
for
Preventing
Juvenile
Delinquency:
A
Systematic
Review of the
Randomized
Experimental
Evidence
By
ANTHONY
PETROSINO,
CAROLYN TURPIN-
PETROSINO,
and
JOHN
BUEHLER
Scared
Straight
and other
programs
involve
organized
visits to
prison
facilities
by juvenile delinquents
or at-risk
kids to deter them from
delinquency. Despite
several
research studies and reviews
questioning
their effective-
ness,
they
remain in use and have now been tried in at
least six nations. The authors
report
here on the results
of a
systematic
review of randomized
experimental
tests
of this
program.
Studies that tested
any program
involv-
ing
the
organized
visits of
delinquents
or at-risk children
to
penal
institutions were included. Each
study
had to
have a no-treatment control condition with at least one
outcome measure
of"postvisit"
criminal behavior.
Using
extensive search
methods,
the authors located nine ran-
domized trials
meeting eligibility
criteria. After describ-
ing
the studies and
appraising
their
methodological
quality,
the authors
present
the narrative
findings
from
each evaluation. A
meta-analysis
of
prevalence
rates
indicates that the intervention on
average
is more harm-
ful to
juveniles
than
doing nothing.
The authors con-
clude that
governments
should institute
rigorous pro-
grams
of research to ensure that well-intentioned
treatments do not cause harm to the citizens
they pledge
to
protect.
Keywords:
Scared
Straight; juvenile
awareness;
delin-
quency prevention; meta-analysis;
ran-
domized
experiments
Background
In the
1970s,
inmates
serving
life sentences at
a New
Jersey prison began
a
program
to "scare"
at-risk or
delinquent
children from a future life
of crime.' The
program,
known as Scared
Straight,
featured as its main
component
an
aggressive presentation by
inmates to
juveniles
visiting
the
prison facility.
The
presentation
bru-
tally depicted
life in adult
prisons
and often
Anthony
Petrosino is coordinator and
Steering
Commit-
tee
memberfor
the
Campbell
Collaboration Crime and
Justice
Group (www.aic.gov.au/campbellcj).
He is also
research
consultantfor
the
Study
on Decisions in Educa-
tion at Harvard
University. Anthony workedfor
several
yearsfor statejustice agencies
in New
Jersey
and Massa-
chusetts,
completed
his Ph.D. in criminal
justice
at
Rutgers University
in
1997,
and
accepted
a
Spencer
DOI: 10.1177/0002716203254693
ANNALS, AAPSS, 589,
September2003
41
THE ANNALS OF THE AMERICAN ACADEMY
included
exaggerated
stories of
rape
and murder
(Finckenauer 1982).
A television
documentary
on the
program
aired in 1979 and
provided
evidence that sixteen or
the seventeen
delinquents
interviewed in the film remained
law-abiding
for three
months after
attending
Scared
Straight,
a 94
percent
success rate
(Finckenauer
1982).
The
program
received considerable and favorable media attention and was
soon
replicated
in more than
thirty jurisdictions
nationwide,
resulting
in
special
congressional hearings
on the
program
and film
by
the U.S. House Subcommittee
on Human Resources
(U.S.
House
1979).
The
underlying theory
of
programs
like Scared
Straight
is deterrence.
Program
advocates and others believe that realistic
depictions
of life in
prison
and
presenta-
tions
by
inmates will deter
juvenile
offenders
(or
children at risk of
becoming
delin-
quent)
from further involvement with crime.
Although
the harsh
presentation
in
the earlier New
Jersey
version is the most
famous,
inmate
presentations
are now
Post-Doctoral
Fellowship
in Evaluation at the Harvard Children's Initiative. His current
research includes a
systematic
review
of
"what works" to
prevent
crime
against
the commercial
airline
industry.
Carolyn Turpin-Petrosino
is associate
professor
in the
Department of Sociology
and Criminal
Justice
at
Bridgewater
State
College
in Massachusetts. She received her B. S.
from
Howard Uni-
versity,
a M. S.W.
from Rutgers
School
of
Social
Work,
and a Ph.D. in criminal
justice from
Rutgers University. Topics of
her recent articles include
parole
decision
making,
hate
crimes,
and
community policing.
She
recently
won the Fiore World
Justice
Award to
study
codification of
hate crime
legislation
in
penal
codes across the
globe.
John
Buehler is a Ph.D. candidate in the Harvard
University Department of
Statistics. He has a
B.A. in
mathematicsfrom
Williams
College
and an A.M. in statistics
from
Harvard
University.
For 5
years
he was a research assistant at the
Centerfor
Evaluation at the American
Academy of
Arts and Sciences. He coauthored a
paper ("Comparing
Results
of Large
Clinical Trials to Those
of Meta-Analysis,"
Statistics in Medicine
2002)
with
Professor
Emeritus Lincoln Moses
of
Stan-
ford University
and
Professor
Emeritus Frederick Mosteller
of
Harvard
University.
His current
research interests involve
estimating
variance
parameters
in Gaussian Hierarchical Models.
NOTE: This review was
completed
as a
pilot
test for the
newly
formed international
Campbell
Collaboration
(C2)
using
the
existing
infrastructure of the Cochrane Collaboration
Develop-
ment,
Psychosocial,
and
Learning
Disorders Editorial
Group.
The full text of this review was
published
in the Cochrane
Library (beginning
in
2002,
issue
2)
and will also be made available in
the
Campbell
Collaboration Reviews
of
Interventions and
Policy
Effects (C2-RIPE)
database.
This review was
partially supported by grants
from the Smith-Richardson Foundation
(to
the
University
of
Pennsylvania),
the Mellon Foundation
(to
the Center for
Evaluation,
Initiatives for
Children
Program
at the American
Academy
of Arts and
Sciences),
and the U.K. Home Office
(to
Cambridge University
Institute of
Criminology).
The Criminal
Justice
Collection at
Rutgers
University's
Center for Law and
Justice
and the Gutman
Library
at the Harvard Graduate School
of Education facilitated
interlibrary
loan
requests.
We thank
Phyllis
Schultze and Carla Lillvik
for their
expertise, patience,
and assistance. We
appreciate
all of the
help
of the Cochrane Devel-
opmental, Psychosocial,
and
Learning
Disorders
Group:
the overall
guidance by
Dr.
Jane
Den-
nis, coordinator;
guidance
and editorial criticisms
by
Professor Geraldine MacDonald, coordi-
nating
editor;
additional comments
by
Dr. Stuart
Logan
and other Editorial
Group
members;
additional searches run
by Jo
Abbott,
trials
coordinator,
and Celia
Almedia, researcher;
and the
statistical criticisms
by Julian Higgins.
Comments
by
Professor Robert
Boruch,
Sir Iain
Chalmers,
Dr. Phoebe
Cottingham,
Professor
Lyn
Feder,
and Professor
Joan
McCord on earlier
drafts of this work have
helped.
42
SCARED STRAIGHT AND OTHER
JUVENILE
AWARENESS PROGRAMS
sometimes
designed
to be more educational than confrontational but with a similar
crime
prevention goal
(Finckenauer
and Gavin
1999;
Lundman
1993).2
It is not
surprising why
such
programs
are
popular: they
fit with common notions
by
some
on how to
prevent
or reduce crime
(by "getting tough"), they
are
very inexpensive
(a
Maryland program
was estimated to cost less than
U.S.$1
per participant),
and
they provide
one
way
for incarcerated offenders to contribute
productively
to soci-
ety by preventing youngsters
from
following
down the same
path
(Finckenauer
1982).
A randomized controlled trial of the New
Jersey program
in
1982, however,
reported
no effect on the criminal behavior of
participants
in
comparison
with a
The
goal of
this review is to assess the
effects
of
programs comprising organized
visits to
prisons ofjuvenile delinquents (officially
adjudicated
or convicted
by
a
juvenile
court)
or
predelinquents
(children
in trouble but not
officially adjudicated
as
delinquents)
aimed
at
deterring
them
from
criminal
activity.
no-treatment control
group
(Finckenauer 1982).
In
fact,
Finckenauer
(1982)
reported
that
participants
in the
experimental program
were more
likely
to be
arrested. Yet belief in the
program's efficacy
continued. Finckenauer called the
process by
which
policy
makers,
practitioners,
media
reporters,
and others some-
times latch onto
quick,
short-term,
and
inexpensive
cures to solve difficult social
problems
the "Panacea Phenomenon." Other randomized trials
reported
in the
United States also
questioned
the effectiveness of Scared
Straight-type programs
in
reducing subsequent criminality
(Greater Egypt
Regional Planning
and Devel-
opment
Commission
1979;
Lewis
1983).
Consistent with these
findings,
reviewers
of research on the effects of crime
prevention programs
have not found deter-
rence-oriented
programs
like Scared
Straight
effective
(Sherman
et al.
1997;
Lundman
1993;
Lipsey
1992).
In
fact,
the
University
of
Maryland's well-publicized
review of more than five hundred crime
prevention
evaluations listed Scared
Straight
as one
program
that does not work
(Sherman
et al.
1997).
43
Despite
this
seeming convergence
of evidence, Scared
Straight-type programs
remain
popular
and continue to be used
(Finckenauer
and Gavin
1999).
For exam-
ple,
a
program
in Carson
City,
Nevada,
brings juvenile delinquents
on a tour of an
adult Nevada state
prison (Scripps
1999).
One
youngster
claimed that the
part
of
the tour that made the most
impact
on him was "all the inmates
calling
us for sex
and
fighting
for our
belongings" (Scripps
1999).
The United
Community
Action
Network
(U-CAN)
has its own
program
called Wisetalk in which at-risk
youth
are
locked in a
jail
cell for more than an hour with four to five
parolees. They
claim that
only
ten of three hundred
youngsters exposed
to this intervention have been re-
arrested
(U-CAN 2001).
In
2001,
a
group
of
guards-apparently
without the
knowledge
of
administrators-strip-searched Washington,
D.C.,
students
during
their tours of a local
jail
under the
guise
of that
they
were
using
"a sound
strategy
to
turn around the lives of
wayward kids"-claiming
the
prior
success of Scared
Straight
(Blum
and Woodlee
2001).
Scared
Straight
and other "kids visit
prison" programs
have been used in several
other nations. For
example,
it is called the
"day
in
prison"
or
"day
in
gaol"
in Austra-
lia
(O'Malley, Coventry,
and Walters
1993),
"day
visits" in the United
Kingdom
(Lloyd
1995),
and the "Ullersmo
Project"
in
Norway
(Storvoll
and Hovland
1998).
Hall
(1999)
reported positively
on a
program
in
Germany designed
to scare
straight
young
offenders with ties to neo-Nazi and other
organized
hate
groups.
The
pro-
gram
has been also tried in Canada
(O'Malley, Coventry,
and Walters
1993).
In a
different variant in the United
Kingdom,
a
program
was initiated that
employed
ex-prison guards
to re-create a
prison atmosphere
in
public
schools,
with the
goal
of
deterring any potential
lawbreakers
(Middleton, Lilford,
and
Hyde
2001).
In
1999,
Scared
Straight:
20 Years Later was shown on U.S. television and
reported
similar results to those of the 1979 film
(see
also Muhammed
1999).
The
1999 version
reported
that ten of the twelve
juveniles attending
the
program
have
remained offense-free in the three-month
follow-up
(Muhammed 1999).
As in the
1979 television
program,
no data on a control or
comparison group
of
young people
were
presented.
Positive
reports
and
descriptions
of Scared
Straight-type pro-
grams
have also been
reported
elsewhere
(e.g.,
in
Germany
[Hall 1999]
and in
Florida
[Rasmussen
and Yu
1996]),
although
it is sometimes imbedded as one
component
in a
multicomponent juvenile
intervention
package (Trusty
1995;
Ras-
mussen and Yu
1996).
In
2000,
Petrosino and his
colleagues reported
on a
prelimi-
nary analysis
of a
systematic
review,
drawing
on the raw
percentage
differences in
each
study. They
found that Scared
Straight
and like interventions
generally
increased crime between 1 and 28
percent
when
compared
to a no-treatment con-
trol
group.
This article
updates
that review and utilizes more
sophisticated
meta-
analytic techniques
to
analyze
the data.
Method
The
goal
of this review is to assess the effects of
programs comprising organized
visits to
prisons
of
juvenile delinquents (officially adjudicated
or convicted
by
a
44
JUVENILE
AWARENESS PROGRAMS
juvenile
court)
or
predelinquents
(children
in trouble but not
officially adjudicated
as
delinquents)
aimed at
deterring
them from criminal
activity.
Eligibility
criteria
We included
only
randomized or
quasi-randomized
(i.e.,
alternation
assign-
ment
procedures
such as
assigning every
other case to
treatment)
controlled
trials,
provided they
had a no-treatment control
group. Only
studies
involving juveniles,
that
is,
children
aged
seventeen or
younger,
were included.
Participants
were
delinquents
or
predelinquents.
Studies that contain
overlapping samples
of
juve-
niles and
young
adults
(e.g., ages
thirteen to
twenty-one)
were also included. The
intervention had to feature a visit
by program participants
to a
prison facility
as its
main
component.
The interest of
citizens,
policy
and
practice
decision
makers,
media,
and the research
community
is in whether Scared
Straight
and other "kids
visit
prison" programs
have
any
crime deterrent effect on the kids
participating
in
them. We therefore focused on crime measures: each
eligible study reported
on at
least one outcome measure of
subsequent criminality (e.g.,
arrest, conviction,
police
contact,
self-reported criminality).
We also list other measures
reported by
investigators
in case
subsequent
reviewers
focusing
on these and other outcomes
require
them to
identify potentially eligible
studies.
Search
strategy for identification of
studies
To minimize
publication
bias or the
possibility journals
are more
likely
to
pub-
lish
findings
that
reject
the null
hypothesis
(and find
programs
to be more effective
than
unpublished
literature
generally
does),
we conducted a search
strategy
designed
to
identify published
and
unpublished
studies. We also conducted a com-
prehensive
search
strategy
to minimize
discipline
bias;
that
is,
evaluations
reported
in
criminological journals
or indexed in
field-specific abstracting
databases
might
differ from those
reported
in
psychological, sociological,
social
service,
public
health,
or educational sources.
First,
randomized
experiments
were identified from a
larger
review of field tri-
als in crime reduction conducted
by
the first author. Petrosino
(1997)
used the fol-
lowing
methods to find more than 300 randomized
experiments
(and
analyze
150):
(1)
"handsearch"
(i.e.,
visually inspecting
the entire
contents)
of
twenty-nine
lead-
ing criminology
or social science
journals;
(2)
checking
the cites
reported
in the
Registry of
Randomized
Experiments
in Criminal Sanctions
(Weisburd, Sherman,
and Petrosino
1990); (3)
detailed electronic searches of
CriminalJustice
Abstracts,
Sociological
Abstracts and Social
Development
and
Planning
Abstracts
(Sociofile),
Education Resource Information
Clearinghouse
(ERIC)
and
Psychological
Abstracts
(PsycInfo);
(4)
searches
by
information
specialists
(before
some of these
became available
online)
of
eighteen bibliographic
databases,
including
the
spe-
cialized collection maintained
by
the National Criminal
Justice
Reference Service
(NCJRS);
(5)
an extensive mail
campaign
with more than two hundred researchers
and one hundred research
centers; (6)
published
solicitations in association news-
45
letters; (7)
tracking
of references in more than
fifty
relevant
systematic
reviews and
literature
syntheses;
and
(8)
tracking
of references in relevant
bibliographies,
books, articles,
and other documents. More detail about these search methods can
be found in Petrosino
(1995, 1997).
The citations found in Petrosino
(1997)
cov-
ered literature with a
publication
date between 1
January
1945 and 31 December
1993. Seven randomized trials
meeting
the
eligibility
criteria were identified from
this
sample.
Second,
we
augmented
this work with searches
designed
to find
experiments
possibly
overlooked
by
Petrosino's
(1997)
search methods and to cover more
recent literature
(1994-2001).
These methods included
(1)
broad searches of the
Campbell
Collaboration
Social,
Psychological,
Educational and
Criminological
Trials
Register
(C2-SPECTR)
developed by
the U.K. Cochrane Centre and now
supervised by
the
University
of
Pennsylvania
Graduate School of Education
(Petrosino
et al.
2000); (2)
check of citations from more recent
systematic
or tradi-
tional reviews
(e.g.,
Sherman et al.
1997;
Lipsey
and Wilson
1998); (3)
citation
checking
of studies and other
reports
on the
program (e.g.,
Finckenauer and Gavin
1999); (4)
e-mail
correspondence
with selected
researchers;
and
(5)
broad
searches of the Cochrane Controlled Trials
Register
[CENTRAL]
in the Cochrane
Library. By
broad
searches,
we mean that we tried to first
identify
studies relevant
to crime or
delinquency
and then we
visually
scanned the citations
(and
abstracts in
most
cases)
to see if
any
were relevant.
Third,
we decided to conduct a more
specific
search of fourteen available elec-
tronic databases relevant to the
topic
area.
Many
of these include
published
and
unpublished
literature
(e.g.,
dissertations or
government reports).
Searches were
done online
using
available Harvard
University
resources or other databases
freely
searchable via the Internet. Several
trips
were made to the
University
of Massachu-
setts to use
CriminalJustice
Abstracts and other
bibliographic
databases not acces-
sible at Harvard or via the Internet.
Jane
Dennis,
Jo
Abbott,
and Celia Almedia of
the Cochrane
Developmental, Psychosocial,
and
Learning
Disorders
Group
repeated
the searches on most of these databases to cover literature added in 2001.
The
bibliographic
databases and the
years
searched were Criminal
Justice
Abstracts,
1968
through September
2001;
Current
Contents,
1993
through
2001;
Dissertation
Abstracts,
1981
through August
2001;
Education Full
Text,
June
1983
through
October
2001; ERIC,
1966
through
2001;
GPO
Monthly
(Government
Printing
Office),
1976
through
2001;
MEDLINE
(Medical
Literature
Analysis
and Retrieval
System
Online),
1966
through
2001;
National
Clearinghouse
on
Child Abuse and
Neglect
database,
through
2001;
NCJRS, through
2001;
Political
Sciences
Abstracts,
1975
through
March
2001;
PAIS International
(Public
Affairs
Information
Service),
1972
through
October
2001;
PsychInfo,
1987
through
November
2001;
Social Sciences Citation
Index,
February
1983
through
October
2001;
and
Sociofile,
January
1963
through September
2001.
We
anticipated
that the amount of literature on Scared
Straight
would be of
moderate size and that our best course of action would be to
identify
all citations
relevant to the
program
and screen them for
potential
leads to
eligible
studies. This
removed the need to include
keywords
for
identifying
randomized trials
(e.g.,
ran-
46
JUVENILE
AWARENESS PROGRAMS
dom
assignment)
in our searches. After several trial
runs,
we found that
nearly
all
documents used
phrases
like Scared
Straight
or
juvenile
awareness in the title or
abstract of the citation.
Therefore,
the
following
searches were run in each relevant
database to
identify
relevant citations: Scared
Straight; (prison orjail
or
reforma-
tory
or
institution)
and
(orientation
or visit or
tour);
prisoner
run or
offender
run or
inmate
run;
prison
awareness or
prison
aversion or
juvenile
awareness;
and
(rap
session or
speak
out or
confrontation)
and
(prisoner
or
lifer
or inmate or
offender).
Finally,
we conducted searches of the Internet and World Wide Web
using
the
above terms in two
popular
search
engines:
Hotbot and Altavista. We later
updated
this with another search
using Google.
Selection
of
trials
The search methods above
generated
more than five hundred citations
(most
had
abstracts).
Anthony
Petrosino
(AP)
screened these
citations,
determining
that
thirty
were to evaluation
reports.
AP and
Carolyn Turpin-Petrosino
(CTP)
inde-
pendently
examined these citations and were in
agreement
that eleven were leads
to
potential
randomized trials. Seven had
already
been retrieved in AP's earlier
review
(Petrosino 1997).
We determined that the full-text
reports
for four should
be
pursued.
These were obtained either
through interlibrary
loan or
by
visits to
area libraries.
Upon inspection
of the full-text
reports,
we determined that two
studies should be excluded. Dean's
(1982)
randomized trial
studying
the
impact
of
Project
Aware in a Wisconsin state
prison
did not include
any postprogram
mea-
sure of
offending,
and our
attempts
to find the author or retrieve this data from
any
other
reports by
the Wisconsin
Department
of Corrections have been unsuccess-
ful.
Chesney-Lind's
(1981)
study
of
Stay Straight
in Hawaii was also
excluded,
as
matching
rather than random
assignment
to conditions was used. After all exclu-
sions,
we were left with nine randomized trials for
analysis.3
We did not find
any
reports
of
ongoing
trials.
Data
management
and extraction
AP extracted data from each of the nine main
study reports using
a
specially
designed
instrument. The data collection instrument was
adapted
from his earlier
study
(Petrosino 1997)
and included such items as
year published, type
of docu-
ment,
location of
study,
evidence that randomization
successful,
randomization
problems, type
of
prison setting (e.g.,
adult,
reformatory),
number in treatment
group,
number in control
group, average age
of
participants
in
study, percentage
of
white
participants
in
study, severity
of
prior
offence record of
participants, study
attrition
problems,
number of
follow-up
measurements of outcome
variable,
type
of outcomes collected: noncriminal
(e.g.,
educational, health)
and
criminal,
and
specific
crime outcome data for
experimental
and control
groups
at each interval.
In cases in which outcome information was
missing
from the
original reports,
we
made
attempts
via e-mail and
regular
mail
correspondence
to retrieve the data for
the
analysis
from the
original investigators.
We were unsuccessful in
obtaining any
47
additional data from
investigators, although
in two
cases,
we retrieved
unpublished
(fugitive
or
"grey"
literature)
master's theses from
university
libraries to see if
they
contained more detailed
reporting
(Cook 1990;
Locke
1984).
They
did not. We ran
statistical
analyses using
Cochrane Collaboration's MetaView statistical
software,
a
component
of Review
Manager
Version 4.1
(RevMan).
These were
repeated,
and
additional
analyses
run,
using
Meta
Analyst
software created
by
Dr.
Joseph
Lau of
the New
England
Cochrane Center. One of us
(John
Buehler)
also created meta-
analytic
formulae in Excel to double-check three of the
analyses.
Results were
identical.
Description
of Studies
Collectively,
the nine studies were conducted in
eight
different
states,
with
Michigan
the site for two studies
(Yarborough
1979;
Michigan Department
of Cor-
rections
1967).
No set of researchers conducted more than one
experiment.
The
studies
span
the
years
from 1967 to 1992. Five studies were
unpublished
and were
disseminated in
government
documents or
dissertations;
the
remaining
four were
found in academic
journal
or book
publications.
The
average age
of the
juvenile
participants
in each
study ranged
from fifteen to seventeen.
Only
the New
Jersey
study
included
girls
(Finckenauer 1982).
Racial
composition
across the nine
exper-
iments was
diverse,
ranging
from 36 to 84
percent
white. Most of the studies dealt
with
delinquent youth already
in contact with the
juvenile justice system. Nearly
1,000
(946)
juveniles
or
young
adults
participated
in the nine randomized studies.
All of the
experiments
were
simple two-group experiments except
Vreeland's
(1981)
evaluation of the Texas Face-to-Face
program. Only
one
study
used
quasi-
random alternation
techniques
to
assign participants
(Cook
and
Spirrison
1992);
the
remaining
studies claimed to use
randomization,
although
not all were
explicit
about how such
assignment
was conducted.
Only
the Texas
study
(Vreeland 1981)
included data on
self-report
measures. In two studies
(Cook
and
Spirrison
1992;
Locke et al.
1986),
no
prevalence
rates were
reported.
Some of the studies that did
include
average
or mean rates did not include standard deviations to make it
possi-
ble to
compute
the
weighted
mean effect sizes.
Also,
the
follow-up periods
were
diverse and included measurements at
three, six, nine, twelve,
and
twenty-four
months. We summarize each
below,
in
chronological
order
by publication
date.
In an
internal,
unpublished government
document,
the
Michigan Department
of Corrections
(1967)
reported
on a trial
testing
a
program
that involved
taking
adjudicated juvenile boys
on a tour of a state
reformatory. Unfortunately,
the
report
is
remarkably
brief.
Sixty juvenile delinquent boys
were
randomly assigned
to
attend two tours of a state
reformatory
or to a no-treatment control
group.
Tours
included fifteen
juveniles
at a time. No other
part
of the
program
is described.
Recidivism was measured as a
petition
in
juvenile
court for either a new offense or
a violation of
existing probation
order.
The Scared
Straight program
at the Menard Correctional
Facility
in Illinois
started in 1978 and is described as a frank and realistic
portrayal
of adult
prison
life
48
JUVENILE
AWARENESS PROGRAMS
(Greater
Egypt Regional Planning
and
Development
Commission
1979).
The
researchers
randomly assigned
161
youth aged
thirteen to
eighteen
to attend the
program
or a no-treatment control. The
participants
were a mix of
delinquents
or
children at risk of
becoming delinquent.
In the
JOLT
(Juvenile
Offenders Learn
Truth)
program, juvenile delinquents
in
contact with one of four
Michigan county
courts
participated (Yarborough
1979).
Each
juvenile spent
five total hours
(half
of that time in the
rap
session)
in the facil-
ity.
After a tour of the
facility, they
were escorted to the
cell,
subjected
to interac-
tion with inmates
(e.g., taunting),
and then taken to a confrontational
rap
session
with inmates. In the
evaluation,
227
youngsters
were
randomly assigned
to
JOLT
or to a no-treatment control.
Participants
were
compared
on a
variety
of crime out-
comes collected from
participating
courts at three- and six-month
follow-ups.
Whether
relying
on the actual data
reported
or measures
of
statistical
significance,
the nine
trials do not
yield evidencefor
a
positive
effectfor
Scared
Straight
and other
juvenile
awareness
programs
on
subsequent delinquency.
The Insiders
Program
in
Virginia
was described as an
inmate-run,
confronta-
tional intervention with verbal intimidation and
graphic descriptions
of adult
prison
life
(Orchowsky
and
Taylor
1981). Juveniles
were locked in a cell fifteen at a
time and told about the
daily
routine
by
a
guard. They
then
participated
in a two-
hour confrontational
rap
session with inmates.
Juvenile
delinquents
from three
court service units in
Virginia participated
in the
study.
The
investigators randomly
assigned eighty juveniles aged
thirteen to
twenty
with two or more
prior adjudica-
tions for
delinquency
to the Insiders
Program
or a no-treatment control
group.
Orchowsky
and
Taylor
(1981)
reported
on a
variety
of crime outcome measures at
six-, nine-,
and twelve-month intervals.
The Face-to-Face
program
in Texas included a thirteen-hour orientation ses-
sion in which the
juvenile
lived as an inmate.
Counseling
followed.
Participants
were fifteen to seventeen
years
of
age
and on
probation
from Dallas
County
Juve-
nile
Court;
most
averaged
two to three offenses before the
study.
One hundred
sixty boys
were
assigned
to four conditions:
prison
orientation and
counseling,
ori-
49
entation
only, counseling only,
or a no-treatment control
group.
Vreeland
(1981)
examined official court records and
self-reported delinquency
at six months.
The New
Jersey
Lifers'
Program began
in 1975 and stressed confrontation with
groups
of
juveniles, ages
eleven to
eighteen,
who
participated
in a
rap
session.
Finckenauer
(1982)
randomly assigned eighty-two juveniles,
some of whom were
not
delinquents,
to the
program
or to a no-treatment control
group.
He then fol-
lowed them for six months in the
community, using
official court records to assess
their behavior.
The California
SQUIRES (San
Quentin
Utilization of Inmate
Resources,
Expe-
rience and
Studies)
program,
which
began
in
1964,
was the oldest in the United
States. The
SQUIRES
program
included male
juvenile delinquents
from two Cali-
fornia counties between the
ages
of fourteen and
eighteen,
most with
multiple
prior
arrests. The intervention included confrontational
rap
sessions with
rough
language, guided
tours of
prison
with
personal
interaction with
prisoners,
and a
review of
pictures depicting prison
violence. The intervention took
place
one
day
per
week over three weeks. The
rap
session was three hours
long
and
normally
included
twenty youngsters
at a time. In the
study,
108
participants
were
randomly
assigned
to treatment or to a no-treatment control
group.
Lewis
compares
them on
seven crime outcomes at twelve months.
The Kansas
Juvenile
Education
Program (JEP)
was an intervention to educate
children about the law and the
consequences
of
violating
it. The
program
also tried
to
roughly
match
juveniles
with inmates based on
personality types. Fifty-two juve-
nile
delinquents aged
fourteen to nineteen from three Kansas counties were ran-
domly assigned
while on
probation
to
JEP
or a no-treatment control. The
investiga-
tors examined official
(from
police
and court
sources)
and
self-report
crime
outcomes at six months for
program
attendees and a no-treatment control
group.
The
Mississippi Project
Aware was a nonconfrontational educational
program
comprising
one five-hour session run
by prisoners.
The intervention was delivered
to
juveniles
in
groups numbering
from six to
thirty.
In the
study,
176
juveniles
between the
ages
of twelve and sixteen under the
jurisdiction
of the
county youth
court were
randomly assigned
to the
program
or to a no-treatment control. The
experimental
and control
groups
were
compared
on a
variety
of crime outcomes
retrieved from court records at twelve and
twenty-four
months.
Methodological Quality
of Included Studies
There are
many
factors on which to
grade
the
quality
of studies.
Complicating
any
assessment of methods is that review
teams,
by
and
large,
must
rely
on written
reports by investigators.
In some
cases,
methodology
sections
may
be
briskly
writ-
ten
(sometimes
because
ofjournal space requirements),
and
key
features of
design
and
analysis may
be deleted or
considerably
condensed. We determined that four
were most critical to
criminological experiments
and
practical
to extract from the
experimental reports.
These were
(1)
randomization
integrity,
(2)
attrition from
50
JUVENILE
AWARENESS PROGRAMS
initial
sample,
(3)
blinding
of outcome
assessors,
and
(4)
fidelity
of
program imple-
mentation. We discuss each of these below.
Randomization
integrity
Did the
investigators report
that
participants
to
experimental
or control condi-
tions
experienced any
violation or subversion of random
assignment procedures?
If
so, did
they report
how
many
cases were
incorrectly assigned
to conditions?
Only
one
study reported problems
with
randomization,
and
they
were dramatic
(Finckenauer 1982).
Only eight
of the eleven
participating agencies
that referred
troubled or
delinquent boys
to the
program correctly assigned
their cases.
Finckenauer
(1982)
did conduct additional
analyses
in an
attempt
to
compensate
for violation of randomization. We
agreed
that a
sensitivity analysis
should be done
to determine the influence of this evaluation on the
pooled analysis.
Attrition
Did the
investigators report major
attrition or loss of
participants
from the sam-
ple initially
randomized? If
so,
did
they report
how much attrition from the initial
sample
occurred? The
Virginia
Insiders
study reported
a
major
loss of
participants
from the initial randomization
sample (Orchowsky
and
Taylor
1981).
This was
reported,
however,
at the second and third
follow-up
intervals
(not
the
first,
at six
months).
Because there was a
paucity
of data
beyond
the first
follow-up
interval
across
studies,
we
only
conducted a
pooled analysis using
the this "first effect."
Therefore,
a
of the
impact
of this later attrition was not
per-
formed. The
Michigan JOLT study
did
report
a
large
number of
no-shows,
but
they
were deleted from the
analysis.
The
problem
is that we do not know how
many par-
ticipants
were
initially assigned,
and we have no assurances from
investigators
that
the
remaining sample
was similar to the initial
sample.
We determined to also con-
duct a
to determine the influence of this
study
on the
pooled
analysis.
Outcome assessor
blinding
It is
nearly impossible
to "blind" either the
practitioners
or the kids to
experi-
mental
conditions,
particularly
when
evaluating
stark contrasts such as Scared
Straight
versus
nothing.
But it is
possible
to blind outcome assessors-who are col-
lecting
data on success and failure-to treatment
assignment. Only
one
study
reported blinding
outcome assessors
(Michigan Department
of Corrections
1967).
Nonetheless,
we believe the likelihood of
corruption by any
bias from outcome
data assessors is
very
low. The reason is that in several of the
experiments reported
here,
a law enforcement
agency
(a
state criminal records
agency
or the Federal
Bureau of
Investigation)
far removed from the treatment
setting
did the criminal
records check. We cannot
imagine
that these
agencies
would be
given anything
51
more than the
participants'
names and other
identifying
information-not what
group they
were
assigned
to.
Program implementation
Did the
investigators report
that the
program
was so
poorly implemented
that
the evaluation was not an accurate assessment of the effectiveness of the interven-
tion? These
programs appear
to be
relatively simple
and short-term and therefore
pose
few
problems
for
implementation.
Not one
investigator reported any imple-
mentation
problems. By
and
large,
the kids received what
they
were
designed
to
receive.
Results
Narrative
findings
Whether
relying
on the actual data
reported
or measures of statistical
signifi-
cance,
the nine trials do not
yield
evidence for a
positive
effect for Scared
Straight
and other
juvenile
awareness
programs
on
subsequent delinquency.
For
example,
the
Michigan Department
of Corrections
(1967)
study
found that 43
percent
of the
experimental group
recidivated,
compared
to
only
17
percent
of the control
group.
No statistical test is
reported. Curiously,
more attention is not
provided
to this
large
negative
result in the
original
document.
In
Illinois,
the outcomes are
statistically insignificant
but
negative
in
direction,
with 17
percent
of the
experimental participants being
recontacted
by police
in
contrast to 12
percent
of the controls
(Greater
and
Development
Commission
1979).
The authors
concluded,
"Based on all available
findings
one would be ill advised to recommend continuation or
expansion
of the
juvenile prison
tours. All
empirical findings
indicate little
positive
outcome,
indeed,
they may actually
indicate
negative
effects"
(ibid., 19).
Researchers
report
no effect for the
program
on attitude measures
(Jesness Inventory,
Piers Harris
Self-Concept
Scale).
In
contrast,
interview and mail
surveys
of
participants
and
their
parents
and teachers indicated unanimous
support
for the
program
(ibid.,
12).
Researchers also note how
positive
and enthusiastic the adult inmates were
about their efforts.
The second
Michigan study
also
reported very
little difference between the
intervention and control
group (Yarborough
1979).
The
average
offense rate for
program participants,
however,
was .69
compared
to .47 for the control
group.
Yarborough
(1979,14)
concluded that "the
inescapable
conclusion was that
young-
sters who
participated
in the
program, undergoing
the
JOLT experience,
did no
better then their control
counterparts."
The
only positive findings, though statistically insignificant,
were
reported
in
Virginia (Orchowsky
and
Taylor
1981).
At six
months,
the results
slightly
favored
52
JUVENILE
AWARENESS PROGRAMS
the control
group
(39
percent
of controls had new court intakes versus 41
percent
of
experimental participants),
but
they
favored the
experimental participants
at
nine and twelve months. The
investigators
noted, however,
that the attrition rates
in their
experiment
were dramatic. At nine
months,
42
percent
of the
original
sam-
ple dropped
out,
and at twelve
months,
55
percent dropped
out. The
investigators
conducted
analyses
that seemed to indicate that the constituted
groups
were still
comparable
on selected factors.
A
study
of the Face-to-Face
Program
in Texas also
reported
little effect for these
interventions. Vreeland
(1981)
reported
that the control
participants outper-
formed the three treatment
groups
on official
delinquency
(28
percent delinquent
versus 39
percent
for the
prison
orientation
plus counseling,
36
percent
for the
Despite
the
variability
in the
type of
intervention
used,
on
average
these
programs
result in an increase in
criminality
in the
experimental group
when
compared
to a
no-treatment control.
According
to these
experiments, doing nothing
would have
been better than
exposing juvenile
to the
program.
prison only,
and 39
percent
for the
counseling only).
The
self-report
measure,
how-
ever,
showed the reverse. None of these
findings
were
statistically significant.
There were
discrepancies
between the
self-report
and official
data;
some who were
officially charged
did not
self-report
the offense and vice versa.
Viewing
all the
data,
Vreeland concluded that there was no evidence that Face-to-Face was an
effective
delinquency prevention program.
He found no effect for Face-to-Face
on several attitudinal
measures,
including
the "Attitudes toward
Obeying
Law
Scale."
Finckenauer
(1982)
reported
that 41
percent
of the kids who attended the Scared
Straight program
in New
Jersey
committed new
offenses,
while
only
11
percent
of
controls
did,
a difference that was
He also
reported
that the
program participants
committed more serious offenses. He also
reported
no
53
impact
of the
program
on nine attitude measures
except
one:
experimental partici-
pants
do much worse on a measure called "attitudes toward crime." His concerns
about randomization
integrity
are treated in a later
sensitivity analysis.
Additional evidence of a
possible
harmful effect can be found in the evaluation
of the
SQUIRES
program.
Lewis
(1983)
reported
that 81
percent
of the
program
participants
were arrested
compared
to 67
percent
of the controls. He also found
that the
program
did worse with
seriously delinquent youths, leading
him to con-
clude that such kids could not be "turned around
by
short-term
programs
such as
SQUIRES
... a
pattern
for
higher
risk
youth suggested
that the
SQUIRES
pro-
gram may
have been detrimental"
(p.
222).
The
only
data
supporting
a deterrent
effect for the
program
was the
average length
of time it took to be rearrested: 4.1
months for
experimental participants
and 3.3 months for controls. Data were
reported
on
eight
attitudinal
measures,
and Lewis
reported
that the
program
favored the
experimental group
on all of them.
Locke and his
colleagues
(1986)
reported
little effect of
JEP
in the Kansas state
prison.
Both
groups improved
from
pretest
to
posttest,
but the
investigators
con-
cluded that there were no differences between
experimental
and control
groups
on
any
of the crime outcomes measured.
Investigators
also
reported
no effect for the
program
on the
Jesness
and Cerkovich
psychological
tests.
Finally,
little difference was found between
experimental
and control
partici-
pants
in the
Mississippi Project
Aware
study
(Cook
and
Spirrison
1992).
For exam-
ple,
the mean
offending
rate for controls at twelve months was 1.25 for control
cases versus 1.32 for
Project
Aware
participants.
Both
groups improved
from
twelve to
twenty-four
months,
but the control mean
offending
rate was
slightly
lower than the
experimental group.
The
investigators
concluded that
"attending
the treatment
program
had no
significant
effect on the
frequency
or
severity
of
subsequent
offenses"
(ibid., 97).
The
investigators
also
reported
on two educa-
tional measures: school attendance and
dropouts. Curiously, they reported
that
Project
Aware reduced school
dropout
rates but noted that "it is not clear how the
program
succeeded in
reducing dropout
rates"
(ibid., 97).
Meta-analysis
For each
study,
we extracted all of the relevant crime outcome data. Our initial
plan
(the
Cochrane/Campbell protocol)
included an
organization
of
analyses by
examining
official
reports
(from
government
administrative
records)
distinct from
self-reported criminality
(obtained
from
investigator-administered survey ques-
tionnaires).
Given that we
expected
a diverse number of measures of crime to be
reported,
we
thought
it would be best to
organize
it into four indexes that would be
most relevant to
policy
and
practice: prevalence
rates
(i.e.,
what
percentage
of each
group
failed or
succeeded?),
average
incidence rates
(i.e.,
what was the
average
number of offenses or other incidents
per
individual in each
group?),
offense
severity
rates
(i.e.,
what was the
average severity
of offenses
per
individual in each
group?),
and
latency
(i.e.,
how
long
was the
average
return to crime or failure
delayed per
individual in each
group?).
54
JUVENILE
AWARENESS PROGRAMS
Where
possible,
we converted the outcome measures to Cohen's effect size
d,
which is also known as the standardized mean difference
(between
the
experimen-
tal and control
groups
divided
by
the
pooled
standard
deviation).
Using
d-Stat
meta-analytic
software created
by
Blair
Johnson
of the
University
of
Connecticut,
d can be
approximated using prevalence
data,
test
statistics,
and other information
when the means and standard deviations are not available. Table 1
provides
the dis-
tribution of these effect
sizes,
which indicate the size and direction of effect for the
program.
Bold
type
indicates
positive
effect sizes
(where
the intervention had a
positive impact
on criminal
behavior).
As Table 1
demonstrates,
nearly
all of the
positive
effects for Scared
Straight
are
generated by
a
single report by Orchowsky
and
Taylor
(1981).
Given that few outcome measures of crime were
reported
across the
studies,
we
were limited to a
single meta-analysis.
We
report
the crime outcomes for official
measures at "first effect"
(and
usually
the
only
effect
reported).
Each of the
analy-
ses focuses on the
prevalence
data,
as the outcomes
reporting
means or
averages
are
sparse
and often do not include standard deviations.
Thus,
because the data
rely
on dichotomous
outcomes,
both
analyses report
odds ratios
(ORs)
for each
study
and their 95
percent
confidence intervals
(CIs).
Because there is some dis-
agreement
in the literature about
this,
we tested the data
assuming
both random
and fixed effects models for
weighting
the treatment effects across the studies.
The forest
graph
in
Figure
1
plots
the ORs for the seven studies
reporting preva-
lence rates. We assume a random effects model
(that
the studies do not come from
some
single underlying population). Figure
1 shows that intervention increases the
crime or
delinquency
outcomes at the first
follow-up period.
The mean OR is 1.72
(95
percent
CI = 1.13 to
2.62)
and is
statistically significant.4
The intervention
increases the odds of
offending
about 1.7:1
(1.7
treatment kids offend for
every
control
participant
who
offends).
There is
always
a
question
about whether or not the results are
being
driven
by
experiments
that
reported methodological problems.
To test for
this,
we con-
ducted a
sensitivity analysis (Figure
2).
Specifically,
we excluded the two studies
identified in our
methodological
assessment as
having potentially threatening
flaws: the Finckenauer
(1982)
experiment
because of concerns about randomiza-
tion
breakdown,
and the
Yarborough
(1979)
study
because of the deletion of no-
shows
(which
could indicate a
potential
for
large
attrition from the initial
study
sample).
We
again
ran
analyses assuming
both random and fixed effects models
(which
did not
differ).
The deletion of these studies did not alter the results: these
programs, according
to the evidence
reported
here,
have an overall
negative
(and
statistically significant) impact
on
subsequent offending.
Discussion
These randomized
trials,
conducted over a
quarter century
in
eight
different
jurisdictions
and
involving nearly
one thousand
participants, provide
evidence that
Scared
Straight
and other
"juvenile
awareness"
programs
are not effective as a
55
TABLE 1
FULL ARRAY OF OUTCOME MEASURES TRANSFORMED TO COHEN'S EFFECT SIZE
(d)
Three Months Six Months
Study Type
of Data P I S L P I S L
MI DOC 1967
GERP&DC 1979
Yarborough
1979
Orchowsky
and
Taylor
1981
Vreeland 1981
Finckenauer 1982
Lewis 1983
Locke et al. 1986
Court
petition
or
probation
violated
Police contacts
New offenses
Petitions
Mean
days
in detention
New intakes
Officially
recorded
delinquency
Self-reported delinquency
Complaints,
arrests,
and
adjudications
Correction for randomization failure
Arrests
Juvenile
court records
-.03
.03
-.17
.52
-.60
-.14
-.14 -.13 -.04
.04
-.04
-.17
.21
-.71
-.33
-.18 -.09 .10
-.02
.15 .29
-.60
F =
.75,
unknown
direction
E does
worse,
but no data
provided
Self-reported offending
Reoffending
(F
test of E vs. C over
time)
Cook and
Spirrison
1992
Twenty-
Nine Months Twelve Months
Four
P I S L P I S L Months
MI DOC 1967
GERP&DC 1979
Yarborough
1979
Orchowsky
and
Taylor
1981
Vreeland 1981
Finckenauer 1982
Lewis 1983
Locke et al. 1986
Cook and
Spirrison
1992
Court
petition
or
probation
violated
Police contacts
New offenses
Petitions
Mean
days
in detention
New intakes .50 .68 .82
Officially
recorded
delinquency
Self-reported delinquency
Complaints,
arrests,
and
adjudications
Correction for randomization failure
Arrests
Juvenile
court records
Self-reported offending
Reoffending
(F
test of E vs. C over
time)
.44 .83 .83
-.32
E2.1, E4.1m,
C2.2 ns C3.3m
(s)
-.09 I
NOTE: Bold
type
indicates
positive
effect sizes. P =
prevalence (percentage
of each
group failing);
I = incidence
(average offending per person);
S =
severity (average severity
score or
percentage
of
group committing
"serious"
offenses);
L
=
latency (average
amount of time to first offense
per
offender);
E =
experimental group;
C = control
group;
m =
months;
s =
MI DOC =
Michigan Department
of
Corrections;
GERP&DC =
Greater
and
Development
Commission.
CA
--.1
FIGURE 1
FIRST EFFECT OF
INTERVENTION,
OFFICIAL
CRIME
MEASURES,
RANDOM-EFFECTS MODEL
Treatment Control OR
ni nlM
(95CIRandom)
Finckenaer 1982
GERP&DC1979
Lewis1983
ichigan
D,O,C1967
Orchowsky&Taorl1981
Vreeland 1981
Yarborough1979
19146
16194
43153
12128
16139
14139
271137
Total(95%CI)
147/436
Test for
heterogney ssq
re=8.50 d6
p02
Test
for
overalleffec
z=2.55
p=001
4/35
8/67
37/55
5130
16141
11/40
17/90
981358
.1 .1
Favourstreatment
I )
I
I
I
I
I-
Weight
OR
%
(95%CIRandom)
9,8
14,7
15.3
9,5
15.2
13.9
21,6
5.4511,65,18,02]
1.5110.61,3.77]
2.09[0,86,5.09]
3.7511.11,12.67]
1,09[0,44,2,66]
1.4810,57,3.83]
1.0510,54,2.07]
100,0
1.72[1.13,2.62]
i lb 6 10
Favour control
NOTE: n = number of
participants reoffending;
N = number
assigned
to
group;
OR = odds
ratio;
CI = confidence
interval;
weight
=
amount of
weight given
to
study
in
analysis;
GERP&DC =
Greater
and
Development
Commission;
D.O.C. =
Department
of
Corrections.
stand-alone crime
prevention strategy.
More
important, they provide empirical
evidence-under
experimental
conditions-that these
programs likely
increase
the odds that children
exposed
to them will commit another
delinquent
offense.
Despite
the
variability
in the
type
of intervention
used,
on
average
these
programs
result in an increase in
criminality
in the
experimental group
when
compared
to a
no-treatment control.
According
to these
experiments, doing nothing
would have
been better than
exposing juveniles
to the
program.
We note that the other two trials that did not
report prevalence
data for the
meta-analysis
also
reported
no effect for the intervention
(Cook
and
Spirrison
1992;
Locke et al.
1986). Indeed,
the mean data from the
Mississippi study
is also
negative
in
direction,
and the Kansas
investigators reported
that the
self-report
data showed a
negative impact.
Our
findings
are also
supported by
a
meta-analysis
of
juvenile prevention
and treatment
programs by Lipsey
(1992),
who indicated
that the effect size for eleven "shock incarceration and Scared
Straight programs"
was -.14
(or
produced
about 7
percent higher
recidivism rates in
experimental par-
ticipants
than controls
assuming
a 50
percent
baseline).
Study
. . .
58
JUVENILE
AWARENESS PROGRAMS
FIGURE 2
SENSITIVITY ANALYSIS
(EXCLUDING
FINCKENAUER 1982
AND YARBOROUGH
1979),
RANDOM-EFFECTS MODEL
Odds Ratio 95%/ Cl
J0 1 0oCe 06 0.1 02 05 1 2 5 10 20 50 100
SNDY YEAR tPTs . . '
, ... ... . , , ......
I
I
1 GERP&DC
1879
161
*
t
2 Leoas 193 10 '
3
IAcigan
DOC 1957 55
*
4
OrdoATayb
131 80
*
5 Vredand 1231 7 ,
4
OVERALL 486 z=2. 2P= O017 -.-
Favors Treatment Favors Control
NOTE: PTs = number of
participants;
CI =
confidence
interval;
GERP&DC = Greater
Egypt
Regional Planning
and
Development
Commission;
DOC =
Department
of Corrections. The
DerSimonian and Laird
procedure
(1986)
was used to create the random effects model.
Given the
strong suggestion
here that these
programs
have a harmful
effect,
they
raise a dilemma for
policy
makers.
Criminological
interventions,
when
they
cause
harm,
are not
just
toxic to the
participants. They
result in increased
misery
to
ordinary
citizens that come from the "extra" criminal victimization
they
create
when
compared
to
just doing nothing
at all.
Policy
makers should take
steps
to
build the kind of research infrastructure within their
jurisdiction
that could
rigor-
ously
evaluate
criminological
interventions to ensure
they
are not harmful to the
very
citizens
they
aim to
help.
Implications for practice
We note the
following irony. Despite
the
gloomy findings reported
here and
elsewhere,
Scared
Straight
and its derivatives continue in
use,
although
a random-
ized trial has not been
reported
since 1992. As Finckenauer and Gavin
(1999)
noted,
when the
negative
results from the California
SQUIRES study
came
out,
the
response
was to end evaluation-not the
program. Today
the
SQUIRES pro-
gram
continues,
evaluated
by
the testimonials of
prisoners
and
participants
alike.
Some
may argue
that these
trials,
with the most recent
reported
in
1992,
do not
apply
to the "newer" Scared
Straight-type programs.
We believe that our review
places
the onus on
every jurisdiction
to show how their current or
proposed pro-
gram
is different from the ones studies here. Given
that,
they
should then
put
in
place rigorous
evaluation to ensure that no harm is caused
by
the intervention.
Some literature indicates the
program
can have a
positive
effect on the inmate
providers,
and that
argument
is sometimes used to
legitimize
use of the
program.
Others claim that the
program by
itself is of little value but could be effective as
part
59
of an overall
multicomponent package
of interventions delivered to
youth.
This
raises ethical issues of whether such
programs
should still be
permitted
if
they
do
increase
criminality
in
juvenile participants.
Despite
these
findings
here and our earlier
report
(Petrosino,
Turpin-Petrosino,
and Finckenauer
2000),
we still
get inquiries
about how to
get
someone's
son,
daughter,
or friend into a "Scared
Straight" program. Many
of these
people
are
understandably looking
for
any program
than can
help
"turn around" a
wayward
or
antisocial
youth. Unfortunately,
we found no evidence that would
support using
this
program
for a
particular type
of kid with a
special
constellation of
personality
or
other characteristics.
Implicationsfor
research
One of the critical
questions
raised
by
this review is
why
the
program
has a
criminogenic
effect. Some
investigators presented
theoretical or
post
hoc rationale
for such
results,
but we did not find
any
of the
theory-driven
or
theory-based
evalu-
ations that would have allowed
investigators
to test causal mechanisms or media-
tors that could
provide
clues as to
why
Scared
Straight
fails. Future research stud-
ies,
including experimental
trials,
ought
to formulate a causal model
diagramming
how the
program
is theorized to work-and then test critical variables that can be
operationalized,
measured,
and tested
(Petrosino 2000).
The
type
of broad search we undertook also should allow us to do a more in-
depth study
of the effects of
using
different evaluative
designs
to test
criminologi-
cal interventions. We now have a
good
collection of
empirical
studies on Scared
Straight
and
plan
to examine these for
experimental, quasi-experimental,
and
nonexperimental
evaluative
design types.
Notes
1.
Programs featuring
inmates as
speakers
who describe their life
experiences
and the current
reality
of
prison
life have a rather
long history,
in the United States at least
(Michigan Department
of Corrections
1967).
2. Some of these
programs
featured interactive discussions between the inmates and
juveniles,
also
referred to as
"rap
sessions."
3. The more detailed version of this review was
published
in the Cochrane
Library
(issue 2, 2002)
and is
forthcoming
in
Campbell
Collaboration Reviews
of
Interventions and
Policy Effects (C2-RIPE).
Readers are
invited to consult these
publications
for more details on the nine included
studies,
including methodological
features. It includes a full list of studies excluded from this
synthesis
and the rationale for such exclusions.
4.
Assuming
a fixed-effects model
(that
the studies come from one
underlying population)
did not
change
these
findings.
The odds ratio was
1.68,
and this was also
References
Blum
J.,
and Y. Woodlee. 2001.
Trying
to
give
kids a
good
scare.
Washington
Post,
3
June,
p.
C01.
Chesney-Lind,
M. 1981. 'Ike Na Pa'ahao: The
experience of
the
prisoners.
A
juvenile
awareness
program.
Report
no. 258. Manoa:
University
of Hawaii at
Manoa,
Youth
Development
and Research Center.
60
JUVENILE
AWARENESS PROGRAMS
Cook,
D. D. 1990. Effects of a non-confrontational
prisoner-run juvenile delinquency
deterrence
program.
Unpublished
master's thesis,
Department
of
Psychology, Mississippi
State
University.
Cook,
D.
D.,
and C. L.
Spirrison.
1992. Effects of a
prisoner-operated delinquency
deterrence
program:
Mississippi's Project
Aware.
Journal
of
Offender Rehabilitation 17:89-99.
Dean,
D. G. 1982. The
impact
of a
juvenile
awareness
program
on select
personality
traits of male clients.
Journal of
Offender
Counseling,
Services and Rehabilitation 6
(3):73-85.
DerSimonian, R.,
and N. Laird. 1986.
Meta-analysis
in clinical trials. Controlled Clinical Trials 7:177-88.
Finckenauer,
J.
0. 1982. Scared
Straight
and the Panacea Phenomenon.
Englewood
Cliffs,
NJ:
Prentice Hall.
Finckenauer,
J.
O.,
and P. W. Gavin. 1999. Scared
Straight:
The Panacea Phenomenon revisited.
Prospect
Heights,
IL: Waveland Press.
Greater
and
Development
Commission. 1979. Menard Correctional Center:
Juvenile
tours
impact study.
Carbondale,
IL: Greater
and
Development
Commission.
Hall,
A. 1999.
Jailhouse
shock aims to scare
youths straight.
The
Scotsman,
26
October,
p.
12.
Lewis,
R. V. 1983. Scared
Straight-California style:
Evaluation of the San
Quentin SQUIRES program.
Criminal
Justice
and Behavior 10
(2):
209-26.
Lipsey,
M. W. 1992.
Juvenile
delinquency
treatment: A
meta-analytic inquiry
into the
variability
of effects. In
Meta-analysis for explanation,
edited
by
T. C.
Cook,
H.
Cooper,
D. S.
Cordray,
H.
Hartmann,
L. V.
Hedges,
R. L.
Light,
T. A.
Louis,
and F. M.
Mosteller,
83-127. New York: Russell
Sage
Foundation.
Lipsey,
M.
W.,
and D. B. Wilson. 1998. Effective intervention for serious
juvenile
offenders: A
synthesis
of
research. In Serious and violent
juvenile
offenders:
Riskfactors
and
successful
interventions,
edited
by
R.
Loeber and D. P.
Farrington,
313-45. Thousand
Oaks,
CA:
Sage.
Lloyd,
C. 1995. To scare
straight
or educate? The British
experience of day
visits to
prisonfor young people.
Home Office Research
Study
no. 149. London: Home Office.
Locke,
T. P. 1984. An
analysis
of the Kansas State
Penitentiary
Juvenile
Education
Program. Unpublished
master's thesis,
Department
of
Psychology, University
of Kansas.
Locke,
T.
P.,
G. M.
Johnson,
K.
Kirigin-Ram, J.
D.
Atwater,
and M. Gerrard. 1986. An evaluation of
ajuvenile
education
program
in a state
penitentiary.
Evaluation Review 10:281-98.
Lundman,
R.
J.
1993. Prevention and control
ofjuvenile delinquency.
2d ed. New York: Oxford
University
Press.
Michigan Department
of Corrections. 1967. A six
monthfollow-up ofjuvenile delinquents visiting
the lonia
Reformatory.
Research
Report
no. 4.
Lansing: Michigan Department
of Corrections.
Middleton,
John,
Richard
Lilford,
and Chris
Hyde.
2001.
Applying Campbell Principles.
Letter to the editor.
British Medical
Journal
323:1252.
Muhammed,
L. 1999. Kids and crooks revisited: Some were "Scared
Straight!"
USA
Today,
12
April, p.
4D.
O'Malley
P.,
G.
Coventry,
and R. Walters. 1993. Victoria's
Day
in Prison
Program:
An evaluation and
critique.
Australian and New Zealand
Journal
of Criminology
26
(2):
171-83.
Orchowsky,
S.,
and K.
Taylor.
1981. The Insiders
juvenile
crime
prevention program.
Richmond:
Virginia
Department
of Corrections.
Petrosino,
A.
J.
1995. The hunt for
experimental reports:
Document search and efforts for a "what works?"
meta-analysis. Journal of
Crime and
Justice
18:63-80.
. 1997. "What works?" revisited
again:
A
meta-analysis
of randomized
experiments
in
rehabilitation,
deterrence and
delinquency prevention.
Dissertation,
Rutgers University,
Ann
Arbor,
MI.
. 2000.
Answering
the
why question
in evaluation: The causal-model
approach.
Canadian
Journal of
Program
Evaluation 15
(1):
1-24.
Petrosino,
Anthony J.,
Robert F.
Boruch,
Cath
Rounding,
Steve
McDonald,
and lain Chalmers. 2000. The
Campbell
Collaboration
Social,
Psychological,
Educational and
Criminological
Trials
Register
(C2-
SPECTR)
to facilitate the
preparation
and maintenance of
systematic
reviews of social and educational
interventions. Evaluation Research in Education 14
(3/4):
293-307.
Petrosino
A.,
C.
Turpin-Petrosino,
and
J.
0. Finckenauer. 2000.
Well-meaning programs
can have harmful
effects! Lessons from
experiments
in Scared
Straight
and other like
programs.
Crime &
Delinquency
46:354-79.
Rasmussen,
D.
W,
and Y. Yu. 1996. An evaluation
ofjuvenilejustice
innovations in Duval
County,
Florida.
Tallahasse: Florida State
University.
61
Scared
Straight!
20
years
later. 1999. Broadcast 8
P.M.,
15
April,
on the UPN television
network,
host
Danny
Glover. Available from
http://www.contrib.andrew.cmu.edu/-aaron2/upn/upn-specials.html.
Scripps, J.
1999. Prison tour serves as a
wake-up
call. The
Forum,
27
October,
p.
1.
Sherman,
L.
W.,
D.
Gottfredson,
D.
MacKenzie,
J.
Eck,
P.
Reuter,
and S.
Bushway.
1997.
Preventing
crime:
What
works,
what
doesn't,
what's
promising.
A
report
to the United States
Congress. College
Park: Uni-
versity
of
Maryland, Department
of
Criminology
and Criminal
Justice.
Storvall,
A.
E.,
and A. Hovland. 1998.
Ullersmoprosjecktet:
"Scared
Straight"
I
Norge
1992-1996
(The
Ullersmo
Project:
Scared
Straight
in
Norway
1992-1996).
Nordisk
Tidsskriftfor
Kriminalvidenskab 85
(2):
122-35.
Trusty,
G. 1995. Sheriffs
office,
juvenile
court resources used for restarts. Louisiana Youth Care
Magazine
18
(5):
3-5.
United
Community
Action Network
(U-CAN).
2001. Services: Wisetalk
(Scared
Straight).
Retrieved 26
April
2001 from
www.ucan.av.org/services.htm.
U.S. House. 1979. Committee on Education and Labor.
Oversight
on Scared
Straight: Hearings before
the
House Subcommittee on Human Resources. 96th
Congress,
1st sess. 4
June.
Washington,
DC: Govern-
ment
Printing
Office.
Vreeland,
A. D. 1981. Evaluation of Face-to-Face: A
juvenile
aversion
program. Unpublished
doctoral dis-
sertation,
University
of
Texas,
Dallas.
Weisburd,
D.
L.,
L. W.
Sherman,
and A.
J.
Petrosino. 1990.
Registry
of
randomized
experiments
in criminal
sanctions,
1950-1983. Los
Altos,
CA: Sociometics
Corporation,
Data
Holdings
of the National Institute
of
Justice.
Yarborough, J.
C. 1979. Evaluation
of JOLT
as a deterrence
program. Lansing: Michigan Department
of
Corrections.
62

American Academy of Political and Social Science

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

American Academy of Political and Social Science

Uploaded by

Copyright:

Available Formats

American Academy of Political and Social Science

You might also like