The Impact of Reentry Services on Juvenile Offenders’ Recidivism Presented by: Jeffrey A. Bouffard, Ph.D. Co-Authored with Kathleen J. Bergseth All opinions in this presentation are those of the authors and do not represent the official position of the agencies participating in the evaluations.
20
Embed
The Impact of Reentry Services on Juvenile Offenders’ Recidivism Presented by: Jeffrey A. Bouffard, Ph.D. Co-Authored with Kathleen J. Bergseth All opinions.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The Impact of Reentry Services on Juvenile Offenders’ Recidivism
Presented by:Jeffrey A. Bouffard, Ph.D.
Co-Authored with Kathleen J. Bergseth
All opinions in this presentation are those of the authors and do not represent the official position of the agencies participating in the evaluations.
Juvenile Aftercare and Reentry Current models call for a combination of
“restraint” and “intervention”
Primary models IAP program (Altschuler & Armstrong, 1994) SVORI (Winterfield & Brumbaugh, 2005) Common characteristics
Coordination of case management and rehabilitation over three phases
Client assessment and individualized case planning Continuity of services
Reentry and Recidivism
Restraint alone is not effective (Petersilia & Turner, 1993)
Mixed evidence for restraint combined with services
Most studies found no difference, but some studies found positive impact
IAP demonstration site study found improvements in some intermediate outcomes, but few significant differences in recidivism (Weibush et al., 2005)
Research plagued with null findings, small sample sizes, implementation difficulties, and little consistency in implementation, or methodology
Community-Based Mentoring Mentoring research finds positive effects
Dubois et al., 2002: mean effect size of .14 to .18 for average program, greater effects for programs with certain characteristics
Mixed research for system involved youth
Blechman et al., 2000: negative impact
Barnoski, 2002: beneficial, but NS impact
Research on AIM program indicates beneficial impact (Jarjoura, 2003; AIM, 2004)
Evaluation Plan Process and Outcome Evaluation
Youth in reentry program with strong mentoring component Compared to similar youth in neighboring county (no reentry services)
All youth returning after 3+ weeks in an “Out of Home” Placement
Youth in both groups receive traditional Probation Supervision
Reentry program
Transitional Coordinators (TC) with Small Caseloads
3 Phase Design; Assessment & Individualized Case Planning; Integration of Supervision & Services
TCs focus on Service Brokerage, Mentoring & Surveillance
Program Structure and Process 2 TCs work closely with 4 existing Juvenile Probation Agents
Assessments: YLS/CMI completed at 4 intervals, before & during program MAYSI-II used to identify potential Mental Health problems
Transitional Case Plans matched to Risks/Needs & Strengths
Transitional Coordinators collaborate with Other Service Providers
Services & Referrals emphasize Education & Family Issues
Flex Funds used for Services, Items & Activities
6-Month Program Duration
Traditional Probation Services continue for Reentry Participants
Sample CharacteristicsTotal Sample
N=112Reentry Services
N=63Traditional Probation
N=49
Age at Referral – Mean (SD) 16.50 (1.39) 16.32 (1.42) 16.75 (1.32)
% Non-White 58.9% 55.6% 63.3%
% Male 72.3% 71.4% 73.5%
Urban Hometown** 57.1% 68.3% 42.9%
Behavior -- Most Recent Charge†
Other Property Persons
34.8%
42.0%
23.2%
28.6%
41.3%
30.2%
42.9%
42.9%
14.3%
Any Prior Official Contact † 90.2% 85.7% 95.9%
# of Prior Contacts -- Mean (SD)*** 5.59 (3.37) 4.40 (2.62) 7.12 (3.63)
Outcomes – Risk/Needs ScoreChange in YLS/CMI Risk/Needs by Domain
Domain % Change Intake to Return
% Change Return to 6 Months
Prior / Current Offenses 21% 18%
Family / Parenting 12% -25%
Education / Employment -7% -43%
Peer Relations 3% -12%
Substance Abuse 3% -26%
Leisure / Recreation 0% -36%
Personality / Behavior -7% -42%
Attitudes / Orientation 12% -32%
Overall 3% -26%
Service Delivery Reentry services
Clients averaged 7 months in program TCs averaged 46 hours of Direct Contact per Client
45% of TC-Client events were ‘Supervisory’ 45% were ‘Mentoring’ 10% were direct ‘Treatment’
Level of Contact: Contacts per week on Probation
No significant difference in base contact levels (PO only) with Youth, Parents or Other Agencies’ Personnel
Program (PO + TC) represents a significant increase in contact levels 292% increase in contact with Youth*** 137% increase in contact with Parents** 65% increase in contact with Other Agencies’ Personnel*
Drug Testing OutcomesUrinalysis within 6 Months of Release
Traditional Probation
Reentry Services Percent Difference
aPercent of Tests that were Positive* 62.17% 34.27% -44.88%
Percent tested*** 30.60% 74.06% 142.03%
aNumber of tests – Mean (SD)***1.53 (1.06)
3.13 (2.11)
104.58%
aNumber of Positive Tests – Mean (SD).87
(.99)
1.11
(1.45)27.59%
† p < .10, * p < .05, ** p < .01, *** p < .000. a Among 15 comparison and 47 reentry services clients receiving at least one drug test in the first six months bSample size 61cSample size 46
Outcomes – 6 Months Post-Release
Total SampleN=112
Reentry Services N=63
Traditional Probation
N=49
Any Recidivism
Has Official Contact 42.0% 36.5% 49.0%
# of Official Contacts – Mean (SD)*.69
(1.06).48 (.76)
.96 (1.31)
Criminal Recidivisma
Has Criminal Contact 34.8% 28.6% 42.9%
# of Criminal Contacts – Mean (SD) †.46
(.82)
.35
(.63)
.61
(1.0)
Days in Restrictive Placement – Mean (SD)23.86
(38.36)
23.46
(37.07)24.37 (40.33)
† p < .10, * p < .05, ** p < .01, *** p < .000. a Excludes status and traffic offenses
Survival Analysis (Cox Regression)
Any Reoffense within 6 Months Criminal Reoffense within 6 Months
Wald 2 Exp(B) Wald 2 Exp(B)
Age at release 1.49 .87 2.20 .84
Non-White 2.67 1.70 1.29 1.49
Male 1.92 1.64 1.64 1.69
Urban hometown .02 .96 .34 .82
# of prior official contacts .30 1.03 .12 .98
Any persons charge 5.25* .50 6.31* .43
RSPa .94 .72 2.07 † .58
-2LL = 409.22, 2 (7, N=112) = 11.54, p = .12
-2LL = 339.62, 2 (7, N=112) = 13.60, p = .06
† p < .10, * p < .05, ** p < .01, *** p < .000aReentry Services effects tested with one-tailed significance
Survival PlotSurvival Proportions up to Six Months Post Release
Days to First Criminal Reoffense
200150100500
Prop
ortio
n no
t Reo
ffen
ding
1.0
.9
.8
.7
.6
.5
Type
Reentry Services
Comparison
Number of Official Contacts 6 Months Post-Release
Count Model (Overdispersed Poisson)
Official Contacts per Week at Risk Criminal Contacts per Week at Risk
# of prior official contacts -.09 (.08) -1.18 -.20 (.13) -1.57
Any persons charge -.98 (.47) -2.07* -1.23 (.77) -1.60
RSPa -1.05 (.51) -2.08* -1.10 (.77) -1.43 †
LL = -74.44 LL = -31.97
† p < .10, * p < .05, ** p < .01, *** p < .000aReentry Services effects tested with one-tailed significance
Supplemental Analyses Findings limited by Short Follow-up Period & Absence of Controls for
Other Factors (Risk/Needs Scores)
Repeated our analyses Survival (Any Recidivism & Criminal Recidivism) Number of New Contacts (Any Offenses and Criminal Offenses)
Control for YLS/CMI risk/need score (N = 95) Support for Reentry Services even stronger controlling for Risk/Need scores Significant beneficial effects for RSP in 3 of 4 outcomes, marginal in 4th
Follow-up to 1 year post-release (N = 95) Reentry youth continue to survive longer, but NS at one year post-release. Significant differences in number of later contacts (any and criminal) remain
to one year post-release.
Summary Service Delivery
High number of Referrals to needed Community-Based Services TC’s engage in a number of Mentoring & Supervisory activities Program increased contact with Youth, Parents, & Other Agencies
Intermediate outcomes More frequent Drug Testing in Reentry Program, but
Significantly lower rates of positive testing
Reentry Program lead to improvements over time in Risk/Need Scores
Recidivism After 6 months: Lower risks of Recidivism, Longer time to 1st Reoffense,
& Fewer New Offenses Even Stronger Support when controlling for Risk/Need levels Several promising results remained 1 year post-release