Transcript Driving Sustainable Long Term Performance Presentation

Dr Peter Johnston
 D i r e c to r
Re s e a r c h a n d A n a l y s i s
 D e p a r t me n t o f
C o r r e c t io n s
Who?
What?
In what form?
How to measure?
Value?
THE OUTCOME: REDUCING RE-OFFENDING
Corrections’ target:
Between 2011 and June 2017, reducing re-offending by 25%
(Target) 2017
16,220
2016
2015
2014
2013
2012
2011
16,546
17,941
18,763
20,491
21,627
Reconvicted offenders (over 12 months to March 31)
Who?
What?
In what form?
How to measure?
Value?
CORE QUESTIONS IN THE REHAB
“INVESTMENT EQUATION”
Who are the right “targets” for rehabilitation?
What
should rehabilitation focus on?
How best to deliver services?
Are
services producing positive impacts?
Is the result “value for money”?
Who?
What?
In what form?
How to measure?
WHO TO TARGET?
8,700
More than 450,000
Prisoners
person records
35,000
Community-sentenced
Value?
Who?
What?
In what form?
How to measure?
THE CENTRALIT Y OF RISK
8,700
Prisoners
35,000
Community-sentenced
Value?
Who?
What?
In what form?
How to measure?
Value?
VARIABILIT Y IN RISK
Birth
Age 26
Line colour = offence type
Homicide ………………
Sexual Offences …..
Volence related …..
Burglary ………………..
Drugs …………………….
Robbery ………………..
Good order etc ..…..
Traffic ……………………
LEGEND
Line thickness = seriousness score
Score <= 40 ……………
Score <= 100 ………….
Score <= 1000 …….….
Score > 1000 ….………
Line thickness = sentence type
Fine
Community Service
Prison …………………..
Who?
What?
In what form?
How to measure?
Value?
VARIABILIT Y IN RISK
Assault
(first offence)
Community
sentence
Robbery
offence
Sentenced
to imprisonment
Custody remand
Line colour = offence type
Homicide ………………
Sexual Offences …..
Volence related …..
Burglary ………………..
Drugs …………………….
Robbery ………………..
Good order etc ..…..
Traffic ……………………
Released
onto Parole
LEGEND
Line thickness = seriousness score
Score <= 40 ……………
Score <= 100 ………….
Score <= 1000 …….….
Score > 1000 ….………
Line thickness = sentence type
Fine
Community Service
Prison …………………..
Parole period
ends
Who?
What?
In what form?
How to measure?
T YPICAL OFFENDER TIMELINES
Recidivist burglars
Lower risk offenders
Value?
Who?
What?
In what form?
How to measure?
Value?
OFFENDING-RELATED NEEDS
Pro-criminal thinking patterns
Low self-control
Drug and alcohol dependence/abuse
Lack of employment skills
Violence propensity
Criminal associates
Alienation from cultural roots
Low literacy
Housing and financial difficulties
Sexual deviance
75
75
60
50
40
40
30
30
15
5
Who?
What?
In what form?
How to measure?
Value?
PROGRAMMES AND SERVICES
AVAILABLE TO OFFENDERS
High Risk/High Intensity
• Special treatment unit rehabilitative
programmes (STURP)
• violent offenders
• personality-disordered offenders
• child sex offenders
• adult sex offenders
• Young offenders programme
• Saili Matagi (for violent Pacific offenders)
Medium Risk/Medium Intensity
• Medium-intensity rehabilitative
programme (MIRP)
• Kowhiritanga (MIRP for women)
• Mauri Tu Pae (Maori Therapeutic Prog)
• Family violence programmes
• Short motivational programme
• Short rehabilitative programme
• Te Tirohanga (Maori focus units)
• Pacific focus unit
Alcohol & Drug
• Brief & intermediate interventions
• Drug treatment units (DTUs - 3m & 6m
formats)
Who?
What?
In what form?
How to measure?
PROGRAMMES AND SERVICES
AVAILABLE TO OFFENDERS
Education
• Literacy/numeracy ed.
• School qualifications
• Trade training qualifications
• Tertiary-level study
In-prison industry placements
• Farming / forestry
• Joinery / timber
• Manufacturing
• Engineering
• Catering / kitchen
Reintegration support services
(mainly for prisoners at time of release)
• “Self-care” pre-release unit placement
• Release-to-work
• “Out of gate” prisoner navigation/support service
• Supported accommodation (short-term)
• “Work and living skills” courses
• Drivers’ license
Value?
Who?
What?
In what form?
How to measure?
Value?
REHABILITATION QUOTIENT (RQ) METHOD
 statistical measure of programme effects (reduced reoffending)
 conducted separately on specific rehab programmes and services
 based on rates of reconviction and re-imprisonment for new offences
within 12 months of release or programme end
 compares reconviction and re-imprisonment rates (RI) of a comparable
group of offenders who did/did not not receive that rehabilitative
programme or intervention
 percentage-point differences in rates between treatment and
comparison group = programme “effect size”
 can also be calculated based on re-offending seriousness
Who?
What?
In what form?
How to measure?
Value?
CALCULATING REOFFENDING RATES
1/04/2013
1/04/2014
1/04/2015
30/06/2015
SCD
Release period
Programme
Grace
period
Possible follow-up period
Offender
released
Grace period
12 month follow-up period
Offender
released
Grace period
12 month follow-up period
Offender
released
12 month follow-up period
Grace period
Who?
What?
In what form?
How to measure?
Value?
RQ PROPENSIT Y SCORE MATCHING
Offenders are matched on risk:
age, sex, ethnicity
age at first conviction
number of previous convictions
total number of sentences
total time in prison to date
summed seriousness of all
offences
• (for sex offender programmes)
type of offending
• actuarial risk score (“RoC*RoI”)
•
•
•
•
•
•
Offenders released in a year
Completed Did not complete
programme
programme
Who?
What?
In what form?
How to measure?
PRISONS RQ RESULTS (2014)
STURP
12.9
Pacific Focus Unit
8.8
DTU (6m)
5.9
Child Sex
5.8
Short Motiv.
5.7
MIRP
5.7
Employment: Internal
4.3
DTU (3m)
3.4
Employment: Timber
Employment: Engineer.
Kowhiritanga
Employment: Primary…
2.8
2.3
2
0.8
Value?
Who?
What?
In what form?
How to measure?
COMMUNIT Y RQ RESULTS
Short Rehab
Short Motivational
8.3
4.8
Value?
Who?
What?
In what form?
How to measure?
Value?
ARE THE BENEFITS WORTH THE COST?
COST/BENEFIT ANALYSIS
Sample programme:
•
•
•
•
•
•
•
annual cost of delivery
number of participants
number of completers
RQ effect size
fewer offenders re-imprisoned
future Corrections costs avoided (5y)
total future costs avoided (5y)
$4,024,000
600
431
-6.8
30
$7,441,869
$12,359,330
(total = Police+Justice+Corrections+Health+victim-related)
Benefit/cost ratio
(total costs avoided/costs of delivery)
3.2
Who?
What?
In what form?
How to measure?
Value?
COST/BENEFIT RATIOS
Prisons
Prisoner Employment
7.91
Short Motivational Programme
5.58
Drug Treatment Unit (3m)
4.04
Drug Treatment Unit (6m)
3.07
Community
Short Rehabilition Programme
1.3
Who?
What?
In what form?
How to measure?
Value?
MAKING BEST USE OF RESULTS
- Maintain/expand rehabilitation services that are effective
- Strengthen
programmes where outcomes are variable
- Improve matching of offenders to programme type
- Discontinue
- Pilot
ineffective services
new services to plug gaps …
Questions?