their abilities and motivation (responsivity principle),
and target their criminogenic needs or changeable risk
factors for recidivism (need principle). Indeed, the
effectiveness of programs is positively associated with
the number of criminogenic needs (e.g., attitudes
supportive of crime) they target relative to noncrimino-
genic needs (i.e., disturbances that impinge on func-
tioning in society, such as anxiety).
Although the surveillance model of supervision still
dominates community corrections, empirical support
for the RNR principle is helping a hybrid model of
supervision gain ascendance in some progressive
agencies. There here has long been tension in commu-
nity corrections between the goals of protecting com-
munity safety (“control”) and promoting offender
rehabilitation (“care”). The surveillance model focuses
exclusively on control, whereas hybrid models blend
control and care. A growing body of research demon-
strates the effectiveness of hybrid models relative
to surveillance models. For example, a recent meta-
analysis indicated that RNR programs significantly
reduced recidivism risk (r=.25), whereas surveillance
programs that applied sanctions without attending to
risk or needs did not (r=−.03).
Studies of intensive supervision programs (ISPs) also
suggest that rehabilitative efforts should be included in
supervision. ISPs were created to reduce prison and jail
crowding by having officers with reduced caseloads
closely supervise relatively serious offenders in the
community with prison-like controls. Traditional ISPs
emphasize monitoring virtually to the exclusion of ser-
vices for offenders. Evaluations of these ISPs robustly
indicate that they do not reduce recidivism and some-
times exacerbate (rather than alleviate) prison crowding.
For example, in an experiment that involved 14 diverse
jurisdictions, offenders were randomly assigned to
either traditional supervision or ISP supervision. A
meta-analysis of these data indicates that, after exclud-
ing the one site in which ISP had a positive effect, ISP
increased the likelihood of offenders’ rearrest by 94%.
Offenders in ISP were particularly likely to return to
prison on technical violations. One might argue that
detecting and sanctioning technical violations is an
index of the surveillance model’s success in preventing
crime. However, there was no evidence that sanctioning
technical offenses prevented new arrests.
Unlike traditional ISPs, hybrid ISPs yield positive
effects. One meta-analysis indicated that ISPs that
incorporated treatment (hybrids) reduced recidivism
by 22%, whereas ISPs that did not (surveillance) had
no effect on recidivism. Based on a matched sampleof 480 parolees, Mario Paparozzi and Paul Gendreau
found that those supervised in a hybrid ISP program
received significantly more social services (e.g., pub-
lic assistance) than those in a traditional parole pro-
gram. Hybrid ISP parolees were substantially less
likely to have new convictions (19% vs. 48%) and
revocations (38% vs. 59%) than traditional parolees.
There is increasing recognition that the manner in
which officers implement supervision has powerful
effects. For example, Paparozzi and Gendreau classi-
fied 12 ISP officers’ supervisory orientation into sur-
veillance, treatment, and hybrid categories. Within ISP,
parolees with hybrid officers (19%) were remarkably
less likely to have their probation revoked than those
with both surveillance (59%) and treatment (38%) offi-
cers. In fact, officers’ orientations toward supervision
affected parolees’ outcomes more strongly than the
particular supervision program applied (i.e., ISP vs.
traditional).Tailoring Responses to
Supervisees With Mental Disorder
The process of supervision may be especially impor-
tant for probationers and parolees with mental disor-
ders (PMDs). Both PMDs and their officers describe
the quality of their relationship as coloring every
interaction and strongly affecting outcomes. There is
some support for this notion. In a study of 90 PMDs,
Jennifer Skeem and colleagues developed and vali-
dated the revised Dual Role Relationship Inventory
(DRI–R) to capture relationship dimensions such as
caring, fairness, and trust. DRI–R scores related
coherently to observers’ codes of officer-probationer
interactions during a supervision session and signifi-
cantly predicted violations and revocation over a
1-year follow-up period.
PMDs constitute a large and at-risk population.
The prevalence of major mental disorders is 4 to 8
times higher in corrections populations than in the
general population. Relative to their nondisordered
counterparts, PMDs are twice as likely to fail on pro-
bation or parole. PMDs are particularly likely to have
supervision revoked for technical violations, perhaps
because their reduced level of functioning makes it
more difficult for them to comply with standard con-
ditions such as maintaining employment. The vast
majority of PMDs have a co-occurring substance
abuse disorder, which elevates their risk of rearrest.
PMDs present a number of unique challenges to
supervising officers, given their pronounced need for102 ———Community CorrectionsC-Cutler (Encyc)-45463.qxd 11/18/2007 12:41 PM Page 102