Changes in Arrest Rate as a Function of Probation and Participant Criminal History Risk: Does Probation Work Best With Lower Risk Probationers?

• Author: Glenn D. Walters
• DOI: 10.1177/0887403417721605
• Published date: 01 June 2019
• Date: 01 June 2019
• Subject Matter: Articles

https://doi.org/10.1177/0887403417721605

Criminal Justice Policy Review

2019, Vol. 30(5) 748 –764

© The Author(s) 2017

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/0887403417721605

journals.sagepub.com/home/cjp

Article

Changes in Arrest Rate as

a Function of Probation

and Participant Criminal

History Risk: Does Probation

Work Best With Lower Risk

Probationers?

Glenn D. Walters1

Abstract

The purpose of this study was to determine whether criminal history risk moderates

the effect of probation on future reoffending. A sample of 327 participants from the

1997 National Longitudinal Survey of Youth (NLSY97) who had been on probation

were compared with 327 propensity score matched members of the NLSY97 who

had been arrested but not placed on probation. Probation and arrest data analyzed

between 1999 and 2008 failed to support the presence of an overall effect for

probation. When the sample was divided into higher criminal history risk (one or

more prior arrests) and lower criminal history risk (no prior arrests), however,

probation was found to reduce recidivism in the lower criminal history risk group

but not in the higher criminal history risk group. Accordingly, probation appeared to

have a small but significant ameliorative effect on future offending in lower criminal

history risk offenders.

Keywords

criminal history risk, evidence-based, probation, propensity matching, recidivism

In a recent national survey of probation departments and probationers, Kaeble and

Bonczar (2016) determined that approximately two thirds of all persons on probation

successfully complete their probation sentences. This is important because research

1Kutztown University, PA, USA

Corresponding Author:

Glenn D. Walters, Department of Criminal Justice, Kutztown University, Kutztown, PA 19530-0730,

USA.

Email: walters@kutztown.edu

721605CJPXXX10.1177/0887403417721605Criminal Justice Policy ReviewWalters

research-article2017

Walters 749

indicates that success on probation is associated with lower levels of subsequent

offending, with one study showing a 5% recidivism rate the first year after successful

completion of probation (Wilks & Nash, 2010). In examining the overall rate of proba-

tion completion, it could be argued that this rate is misleading in that misdemeanant

probationers—who usually make up about half the probation population (Kaeble &

Bonczar, 2016)—rarely reoffend (DeLisi & Conis, 2011). The same cannot be said of

felony probationers. In a classic study conducted for the Rand Corporation, Petersilia,

Turner, Kahan, and Peterson (1985) found that nearly two thirds of a group of 1,672

felony probationers serving time for property and violent offenses were rearrested

within 40 months. Even though a much lower rate of rearrest (i.e., 32%) was observed

in a national survey of felony probationers serving time for property offenses (Benedict

& Huff-Corzine, 1997), the results of the Rand study raised concerns among criminal

justice policy makers about the risk felony probationers pose to the general public.

MacKenzie, Browning, Skroban, and Smith (1999) conducted interviews with 107

probationers from several different counties in northern Virginia. In the year before

they began supervision, 6.5% reported committing an average of 67 burglaries, 18.7%

reported committing an average of 43 thefts, 15.0% reported committing an average of

251 forgeries, 8.4% reported committing an average of 14 robberies, 28.0% reported

committing an average of 16 assaults, and 37.4% reported committing an average of

1,003 drug deals. After being on probation for 8 months, the prevalence and frequency

of offending dropped but did not disappear. While on probation, 0.9% reported com-

mitting an average (annualized rate) of three burglaries, 1.9% reported committing an

average of 10 thefts, 2.8% reported committing an average of eight forgeries, 0.9%

reported committing an average of 15 robberies, 14.0% reported committing an aver-

age of three assaults, and 10.3% reported an average of 408 drug deals. MacKenzie

et al. concluded that while the supervision provided by probation significantly reduced

criminality, there was no evidence that additional probation conditions or actions on

the part of the probation officer had any effect on the self-reported criminality of this

sample of 107 probationers. The MacKenzie et al. study, while interesting, suffers

from two problems. First, the self-report of preprobation offending was retrospective

and occurred when the individual was already on probation. Second, like most studies

on probation and recidivism, there was no comparison group to control for factors like

history, maturation, and statistical regression.

Despite the public safety concerns raised by the Rand and MacKenzie et al. (1999)

studies, it is important that these results not be over-interpreted. What I mean by this

is that, probation, despite its limitations, may still achieve better results than more

restrictive and costly alternative sanctions like intensive probation and incarceration.

Using data gathered from a large sample of convicted felons in the state of Florida,

Cochran, Mears, and Bales (2014) compared the recidivism rates of individuals sen-

tenced to standard probation, intensive probation, jail incarceration, and prison incar-

ceration. In each comparison, participants from the standard probation group were

matched with participants from the counterfactual group with propensity scores based

on age, race, severity of the current offense, prior convictions, prior prison commit-

ments, judicial circuit, and a binary measure of prison eligibility. Recidivism was