Risk, Needs, and Behavior: Moderating the Mediated Relationship Between Criminal Thinking, Institutional Infractions, and Recidivism With Time Spent in Prison

• Author: Glenn D. Walters
• DOI: http://doi.org/10.1177/00938548221131948
• Published date: 01 April 2023
• Date: 01 April 2023
• Subject Matter: Articles

CRIMINAL JUSTICE AND BEHAVIOR, 2023, Vol. 50, No. 4, April 2023, 480 –496.

DOI: https://doi.org/10.1177/00938548221131948

Article reuse guidelines: sagepub.com/journals-permissions

© 2022 International Association for Correctional and Forensic Psychology

480

RISK, NEEDS, AND BEHAVIOR

Moderating the Mediated Relationship Between

Criminal Thinking, Institutional Infractions, and

Recidivism With Time Spent in Prison

GLENN D. WALTERS

Kutztown University

The research question addressed in the current study asked whether a risk factor would moderate the relationship between a

needs factor and behavior and in so doing increase a person’s odds of reoffending. In this study, prison time was the risk

factor, proactive and reactive criminal thinking the needs factors, and institutional infractions the behavior. A sample of 1,101

men (mean age = 34.64 years) released from federal prison between 2003 and 2009 participated in a study where a moderated

mediation design was employed. The interaction between proactive criminal thinking and time spent in prison was found to

predict more infractions in prison, which, in turn, predicted increased levels of recidivism upon release from prison. These

results suggest that time spent in prison may provide a context in which proactive criminal thinking encourages antisocial

behavior in prison, which then puts the individual at risk for future offending.

Keywords: moderated mediation; criminal thinking; time in prison; institutional infractions

In a survey of persons incarcerated in a Norwegian prison, Kolstad (1996) noted that about

half (44%) reported growing more hostile and resentful toward society as a result of

incarceration and nearly all (92%) viewed prisons as universities of crime where inmates

learn attitudes and behaviors conducive to future offending. Correctional staff can be as

pessimistic about prisoner change as the prisoners themselves (Clarke et al., 2004), and

while the general public’s views on the rehabilitative effects of prisons tend to be more posi-

tive than those who work or are incarcerated there, the overall consensus is still mixed

(Cullen et al., 2000). This raises the prospect that prison may serve as a risk factor for future

offending, though not everyone agrees that prison exerts a uniformly negative impact on

adjustment and future recidivism (Bonta & Gendreau, 1990; Gendreau et al., 1999). Given

the dearth of information currently available on risk x needs interactions (interactive

AUTHOR’S NOTE: Correspondence concerning this article should be addressed to Glenn D. Walters,

Department of Criminal Justice, Kutztown University, Kutztown, PA 19530-0730; e-mail: walters@kutztown.edu.

1131948CJBXXX10.1177/00938548221131948Criminal Justice and BehaviorWalters / Criminal Thinking and Time in Prison

research-article2022

Walters / CRIMINAL THINKING AND TIME IN PRISON 481

protective effects being a notable exception: Farrington et al., 2016). Bonta and Andrew’s

(2017) risk–need–responsivity (RNR) model will be used to explain the criminogenic effect

of time spent in prison in relationship to certain needs factors like negative attitudes toward

conventional society utilizing a moderation or moderated mediation methodology, The goal

of the present study was to explore the interaction between time spent in prison (risk factor)

and criminal thinking (needs factor) in predicting future recidivism by way of its impact on

institutional infractions (mediating behavioral variable).

TIME SPENT IN PRISON AS A RISK FACTOR

In a review of studies published between 1972 and 2008, Nagin and colleagues (2009)

determined that compared to noncustodial alternatives, incarceration had either a null or mild

criminogenic effect on future offending. There were a few experimental investigations that

randomly assigned participants to conditions in this review, but the majority of studies relied

on matching to equate groups, control certain variables, and estimate dose-response relation-

ships. Results were uniformly disappointing regardless of methodology. One early study not

included in the Nagin et al. review compared convicted adult males randomly assigned to

serve 6 months in state prison or 6 months in a therapeutic bootcamp. Results showed that

while bootcamp participants displayed mild reductions in self-control over 6 months of con-

finement, persons randomly assigned to prison experienced much larger reductions in self-

control, reductions that were compounded by equally large reductions in anger management

and augmentation of antisocial attitudes and behaviors (MacKenzie et al., 2007). After match-

ing 406 pairs of individuals convicted of non-aggravated assault on prior prison status, prior

appearances, number of concurrent appearances, and bail status, Weatherburn (2010) discov-

ered that participants sentenced to prison experienced significantly higher rates of recidivism

than participants sentenced to probation. Several more recent studies have also observed

higher rates of recidivism in persons released from prison compared with persons completing

a probation sentence (Caudy et al., 2018; Harding et al., 2017).

A question that continues to confound researchers is whether the risk effect of incarcera-

tion is counterbalanced by positive effects or whether prisons produce only null and crimi-

nogenic effects as suggested by the Nagin et al. (2009) review. Bhati and Piquero (2007)

attempted to answer this question using trajectory analysis. Reviewing data on 38,624 peo-

ple released from prisons in 15 states, Bhati and Piquero identified a small group (4%) that

upon release from prison returned to trajectories of offending higher than those they were

on prior to incarceration, thus signaling a criminogenic effect. By contrast, 40% of releasees

returned to trajectories of offending lower than those they had been on prior to incarcera-

tion, a possible sign that they were less likely to engage in future offending. Walters (2016b)

also used trajectory analysis to distinguish between risk and promotive effects and obtained

a 2.6:1 ratio (72% vs. 28%) in favor of promotive effects, as opposed to the 10:1 ratio

reported by Bhati and Piquero (2007). Another unique aspect of the Walters (2016b) inves-

tigation was that while the single declining trajectory dropped only slightly from pre-prison

(prior offending) to prison (institutional infractions) to post-prison (recidivism), the three

accelerating trajectories (a small rapidly accelerating trajectory, a small moderately accel-

erating trajectory, and a moderately sized slightly accelerating trajectory) displayed incre-

ments in antisocial behavior from prison to post-prison that were somewhat larger than the

small reduction observed in the declining trajectory.