Long-Term Effectiveness of a Brief Restorative Justice Intervention

Abstract

This research investigated the effectiveness of a brief Restorative Justice Intervention. Probationers who attended a Restorative Justice Intervention (n = 383) were compared with probationers receiving treatment as usual (n = 130) over a 2- to 6-year follow-up period. The proportion of individuals who recidivated in the control condition (n = 89, 68.46%) were higher compared with those who recidivated in the intervention condition (n = 127, 33.16%; z = 7.04, p < .001). In addition, among those who recidivated, those in the intervention condition did so less frequently. Qualitative analyses from a postintervention course evaluation given only to the intervention condition showed that 50% of probationers acknowledged an empathic understanding associated with participation. This brief intervention has a positive multilevel impact on restorative justice. Implications of these effects are discussed.

Keywords

recidivism restorative justice probation

Criminal recidivism in the United States remains at a high level, despite long-standing attempts to curb reoffending (Beck & Shipley, 1989; Durose, Cooper, & Snyder, 2014; Lagan & Levin, 2002). According to the most recently published national data, 76.6% of convicted offenders were rearrested within 5 years (Durose et al., 2014). Although the traditional method of incarceration intends to deter offending and subsequent reoffending, recidivism rates have remained high for several decades and a comprehensive solution has yet to be identified and implemented (Beck & Shipley, 1989; Durose et al., 2014; Lagan & Levin, 2002). To counter rising recidivism rates, researchers have examined what are known as Restorative Justice Interventions (RJIs) as both a supplement to and, in some cases, a replacement of the traditional method of incarceration (Bonta, Wallace-Capretta, Rooney, & McAnoy, 2002; Clamp & Doak, 2012; Dhami, Mantle, & Fox, 2009; Toews, 2013; Tsui, 2014).

Although most researchers agree that RJIs are designed to repair and reduce harm caused by criminal offending (Braithwaite, 2000; T. Marshall, 1999; Roche, 2001; Van Ness, 2004), the essential components of these programs are under debate. Initially, RJIs were designed to bring the victim and the offender together in a mediated meeting to repair harm caused by criminal offending (T. Marshall, 1999). However, questions soon arose about the outcome of these meetings and whether they were achieving the intended goals of reparation and reintegration (Roche, 2001). As a result, focus shifted toward testing such programs’ ability to repair, reintegrate, and reduce harm related to criminal offending. Researchers also suggested that the proposed mechanism of change results from a meeting between the victim(s) and the offender(s) (Daly, 2016). Although defining restorative justice and its associated principles is an important undertaking for a variety of reasons (see Daly, 2016 for a comprehensive review), it is not necessarily the aim of this article. Rather, the present study focused on testing the effectiveness of an alternative approach to traditional victim-offender mediated RJIs.

Another key point in the RJI debate is the mechanism through which this change occurs. As noted earlier, it has been proposed that the mechanism of action in a RJI is the meeting between the victim and the offender (Daly, 2016). Because of the importance of the victim’s perspective in this process, much of the research on RJIs examines the viewpoint of the victim (Androff, 2012). Although offenders are often included in the RJI process, typically for the benefit of the victim, their viewpoint and progress through the restorative process are often overlooked. We argue that offender habilitation is not only possible through RJIs but also essential to the community reparation process. If offenders are less likely to reoffend, the safety and security of the community and its members has been effectively increased.

Meta-analytic data examining the degree to which RJIs reduce recidivism among offenders show that most, but not all (Feder & Wilson, 2005; Redondo, Sanchez-Meca, & Garrido, 1999) are relatively effective at reducing recidivism over a 12 to 18 month period (Bergseth & Bouffard, 2013; Forgays & DeMilio, 2005). However, victim-offender moderated RJIs are labor intensive and limited by practical concerns involving cost-effectiveness and logistics of implementation. Some research also shows that less-complex RJIs (e.g., single-session, victim-absent designs) produce similar reductions in recidivism (Forgays & DeMilio, 2005; Koss, 2014). Despite these promising results, an aim of this study is to extend our knowledge on long-term effects by examining recidivism across multiple years. Moreover, we also aim to add to extant literature by examining the degree to which those who recidivate may do so less frequently. Such evidence would be suggestive of a potential and yet to be discovered harm-reduction effect of these interventions.

Another primary assumption of RJIs is that greater understanding and empathy developed by offenders will produce emotional and behavioral changes that lead to the lower likelihood of reoffending. Consistent with this assumption, RJIs sometimes include a component intended to increase the offender’s understanding of the victim’s experience of the crime (i.e., increased empathy, guilt, or shame; Calhoun & Pelech, 2010; Forgays & DeMilio, 2005; Jackson & Bonacker, 2006; Latimer, Dowden, & Muise, 2005; Martinez, Stuewig, & Tangney, 2014; Rodogno, 2008; Rossner, 2012; Saulnier & Sivasubramaniam, 2015; Sherman et al., 2005; Van Stokkom, 2002). One of the ways RJIs attempt to build offender empathy is through a variety of perspective taking activities (e.g., Victim Impact Statements [VISs]; Jackson & Bonacker, 2006). These activities are designed to help the offender achieve a greater understanding of how their crime negatively impacts the victim(s) and their community, which is an important component of empathy when situational understanding does not develop naturally (Wondra & Ellsworth, 2015). Because insight into the impact of the crime does not appear to develop naturally for most offenders (L. E. Marshall & Marshall, 2011; Robinson & Rogers, 2015), increasing perspective taking may be particularly beneficial in the context of an RJI for most offenders (Martinez et al., 2014; Pithers, 1999). We caution the use of such techniques with sexual offender populations because research suggests that increasing perspective taking when the training focuses on understanding another’s emotional state may be unhelpful or even detrimental to their rehabilitation (Covell & Scalora, 2002; Mann & Barnett, 2012; L. E. Marshall, Marshall, Serran, & O’Brien, 2009).

Despite the empathic enhancement assumption embedded within RJIs, current understanding of the role of empathy in the reduction of recidivism is limited. At best, there are mixed results related to the degree to which empathy can be enhanced within RJIs. For example, Jackson and Bonacker (2006) found no significant increases in guilt, shame, or empathy after a Victim Impact Training Program, a program designed to educate offenders about the human impact and consequences of crime. In contrast, Calhoun and Pelech (2010) found a significant increase in the areas of accountability (e.g., I know the incident was my fault), relationship repair (e.g., I put the victim and the people close to him or her through a lot), and closure (e.g., I’ve been able to say how hard this has been for me) after involvement in an RJI that sought to compare knowledge on these outcomes to those in a control condition (i.e., conventional court process). Because of the limited data on empathy enhancement effects resulting from RJIs, more research is needed to assess the influence of RJIs on empathetic understanding.

The present study examined the effectiveness of an existing single-session RJI routinely conducted in a statewide probation system within the United States. This convenience sample utilized probationers that did not complete the RJI (e.g., treatment as usual [TAU]) but were on probation at the same time as a comparison condition. The first aim of the study was to test the hypothesis that probationers who enrolled in a brief RJI would recidivate at lower rates compared with a cohort of probationers receiving TAU. Due to the staggered start times of probationers, follow-up time ranged between 2 and 6 years. The second aim of the study was to test the hypothesis that among recidivating probationers, those in the RJI condition would recidivate at a lower frequency (i.e., number of offenses during follow-up) than those in the TAU comparison condition. Finally, we explored the degree to which those in the RJI condition would acknowledge empathic understanding within a postintervention evaluation.

Method

Study Design

To address each hypothesis, the present study utilized a simultaneous multimethod design with a deductive theoretical drive consisting of a quantitative and a qualitative component (QUAN + qual; Morse, 2003). Mixing research methodologies aids in facilitating a greater understanding of the experiential aspects of human behavior, while enhancing the research paradigm (Morse, 2003; Teddlie & Tashakkori, 2002). The quantitative component of this study aimed to test the effectiveness of the RJI, while the postintervention qualitative component aimed to provide a description of the probationers’ experiences, as well as describe what aspects of the RJI the probationers found valuable or otherwise.

Participants and Procedure

Study questions were assessed between 2006 and 2010 using individuals assigned to probation of a Plains state in the United States. Probationers in the RJI (77 sessions; n = 383) were required to complete the intervention as a supplement to the terms of their probation. Because the assessment was not originally designed as a randomized clinical study and State probation was interested in the relative effectiveness of the RJI, they utilized probationers (n = 130) that underwent standard probation procedures and were never enrolled in the RJI as the TAU control. All records were collected at probation by certified probation officers and then deidentified before research staff obtained the data. The university institutional review board (IRB) approved the use of this deidentified data to examine the potential effectiveness of the RJI among probationers.

The aim of the quantitative component of the present study sought to assess the effectiveness of the RJI by comparing recidivism rates between those who had taken the RJI with a TAU control. Table 1 presents the sample statistics. For the qualitative component of the present study, anonymous data were derived from 608 responses to RJI evaluations (n_participants = 368; each probationer could generate multiple responses to each question). Due to the anonymous nature of the postintervention evaluations, we were unable to link these responses to the quantitative data set. To gather postintervention reactions to the RJI, probationers were asked four questions about the class: (a) How was the class helpful? (n_response = 590), (b) What did you like most about this class? (n_response = 590), (c) What did you like least about the class? (n_response = 589), and (d) Do you have additional comments? (n_response = 591) and one question about the effectiveness of the presenter (n_response = 590; scale = poor, fair, good, excellent).

Table 1.

Descriptive Statistics and Differences Between the TAU and RJI Group.

	TAU (n = 130)				RJI (n = 383)				z ^a	t
	n	%	M	SD	n	%	M	SD	z ^a	t
Age			27.87	8.76			32.24	10.6		−4.24**
Recidivate	89	68.46			127	33.16			7.04**
Female	32	24.62			131	34.20			−2.03*
Ethnicity
White	80	61.54			314	81.98			−4.77**
African American	36	27.69			55	14.36			3.44**
Native American	6	4.62			4	1.04			2.54**
Asian	0	0.00			7	1.83			−1.55
Other	8	6.15			3	0.78			3.65**
Reason for probation
Drug only	37	28.46			173	45.17			−3.35**
Nondrug only	86	66.15			203	53.00			2.61**
Both drug and nondrug	7	5.38			7	1.83			2.15*
Worst offense was
Felony	81	62.31			239	62.40			−0.02
Misdemeanor	49	37.69			144	37.60			0.02
LS/CMI Total Score			28.18	6.34			23.28	7.72		6.80**
Time spent on probation (in months)			28.43	14.00			28.46	15.92		−0.02
Time served in prison (in months)			4.73	22.18			2.02	7.70		1.36

Note. Drug only = charges related to alcohol or drugs, such as DUI and possession of paraphernalia. Nondrug only = charges unrelated to alcohol of drugs. TAU = treatment as usual; RJI = Restorative Justice Intervention; LS/CMI = Level of Service/Case Management Inventory; DUI = driving under the influence.

Difference in Proportions Test.

p < .05. **p < .01.

RJI

The intervention used in this study consisted of a single-session, 8-hr intervention designed to help offenders understand the impact of their crime. Facilitators trained in the areas of restorative justice and Victim Impact education conduct RJI sessions. The foundational component of the RJI used VISs to assist probationers in evaluating the extent of the crime(s) on their victim(s) and the community. In contrast to other RJIs (Sherman et al., 2005), the VISs used in the present intervention were not derived from the actual victims of probationers in the RJI. Rather, a collection of previously written or recorded VISs were selected to match the offenses of the RJI members. Although reading and discussing these statements, offenders were asked to document, in detail, how victims and communities were harmed by the crime (emotionally, physically, financially, and spiritually) with the intent of building offender empathy and understanding the damages associated with their crime. All RJI probationers read and discussed several VISs during the RJI.

Other components of the RJI included the following: (a) a discussion of the definition and purpose of Restorative Justice, (b) education about health and wellness needs of offenders and their emotional recognition skills while continuing to hold them accountable, and (c) a group discussion about the harm and damages caused to victims and communities. In addition to these core components, offenders were also required to complete three outside assignments within 10 days of the RJI. These assignments included (a) documentation of their direct and indirect victims and all the harm and damages caused to these parties, (b) reading and analyzing ten additional VISs, and (c) writing an apology letter to their direct and indirect victims. Apology letters were written, but not delivered to the actual victim. Rather, the purpose of this exercise was to develop the offender’s insight into the impact of their crime.

The present RJI was designed to address current components of restorative justice. By attempting to reduce reoffending, safety and security of community members is increased. By focusing on the offender’s viewpoint during the RJI process, we are respecting their needs as humans. By educating offenders about the impact of their crime and assisting them to become contributing members of society, we are acknowledging the values of inclusion. Finally, by centrally including the offenders in this process, we are attempting reparation of the crime and collaboration among community members. Although the focus of many RJIs is on the victim and their viewpoint of the crime, interventions designed to change the offender’s viewpoint appear to fit a variety of RJI definitions (Daly, 2016; T. Marshall, 1999; Roche, 2001; Van Ness, 2004).

Measurements

Recidivism was defined as re-arrest between RJI completion and the last follow-up period or TAU entry and the last follow-up. Trained probation officers (two) were given access to local, state, and national databases (National Crime Information Center [NCIC] and Nebraska Criminal Justice Information System [N-CJIS]) for tracking re-arrest. Recidivism data between 2006 and 2012 were collected at the end of the study to ensure complete data. A 6-year follow-up was chosen to capture most reoffending given that 75% of offenders recidivate within 5 years (Durose et al., 2014).

The Level of Service/Case Management Inventory (LS/CMI) was designed to measure risk and need factors of adult offenders (Andrews, Bonta, & Wormith, 2004). The LS/CMI assesses eight domains of risk and need: (a) criminal history (e.g., number of prior convictions), (b) education and employment, (c) family/marriage, (d) leisure and recreation, (e) companions (e.g., association with delinquent friends), (f) alcohol and drug problems, (g) procriminal attitude orientation, and (h) antisocial patterns. Research suggests that the LS/CMI has adequate reliability (α = .89 for males and α = .91 for females; Andrews et al., 2004; Andrews, Bonta, & Wormith, 2008) and that it predicts recidivism across gender, age, types of offender, and nationality (Vose, Cullen, & Smith, 2008). Probation officers conducted LS/CMI assessments with every probationer who entered the criminal justice system. For data analysis, deidentified scores were aggregated into an overall level need/risk score (higher scores indicate higher levels of need/risk) before given to researchers.

Results

Quantitative Analysis

Table 1 presents the descriptive statistics and group comparisons between TAU and RJI conditions. A Cox regression survival analysis was performed to evaluate the effectiveness of the RJI on recidivism. Results indicate that the RJI significantly reduced the incidence of recidivism by 66% compared with TAU (see Table 2, Step 1). Figure 1 presents the proportion of those who recidivated (inverse of the survival function) from the year they entered probation (Year 0) until Year 6. These data show that around 30% of those in the RJI condition recidivated by Year 2 compared with around 65% in the TAU condition. By the sixth year, approximately 70% of those in the TAU condition recidivated, while the RJI condition only showed a 35% recidivism rate. This effect remained after controlling for covariates related to recidivism such as gender, age, ethnicity (White vs. non-White), type of offense, and LS/CMI scores (see Table 2, Step 2). After accounting for RJI treatment effects, being younger and having worse LS/CMI scores were related to recidivism. Overall, the proportion of individuals who recidivated in the TAU condition (n = 89, 68.46%) were higher compared with those who recidivated in the RJI condition (n = 127, 33.16%; z = 7.04, p < .001).

Table 2.

Results of the Cox and Snell Regression Survival Analysis.

	b	SE	OR	p	R ²
Step 1
Restorative Justice Intervention	−1.078	0.140	0.340	<.001	.099
Step 2
Restorative Justice Intervention	−0.521	0.161	0.594	.001	.180
Female	−0.173	0.156	0.841	.267
Age	−0.020	0.008	0.980	.009
Non-White	0.179	0.155	1.196	.248
Reason for probation: drug vs. nondrug	0.126	0.153	1.135	.409
Reason for probation: drug vs. both	0.010	0.397	1.010	.979
LS/CMI Total Score	0.045	0.008	1.046	<.001

Note. Drug only = charges related to alcohol or drugs, such as DUI and possession of paraphernalia. Nondrug only = charges unrelated to alcohol of drugs. For Reason for probation, drug only group was the reference group. LS/CMI = Level of Service/Case Management Inventory; DUI = driving under the influence.

Figure 1.

Proportion of recidivism between RJI and TAU group across 6 years.

Group comparisons were performed to further understand recidivism in the RJI condition (see Table 3). Results indicate that, recidivating probationers in the RJI were typically younger, had a significantly higher LS/CMI total and subscale scores, and were charged with a felony as the worst charge at baseline. Among recidivating probationers in the TAU condition, no significant differences were observed between those who did and did not recidivate (see Table 4).

Table 3.

Restorative Justice Intervention Group Comparisons Among Those Who Recidivate (n = 127) and Did Not Recidivate (n = 256).

	Did not recidivate			Recidivate			t	z ^a
	M	SD	n (%)	M	SD	n (%)	t	z ^a
LS/CMI total score	14.52	12.19		22.61	10.33		−6.42**
Antisocial patterns	1.13	1.24		1.95	1.29		−6.04**
Procriminal attitudes	1.04	1.31		1.65	1.23		−4.33**
Alcohol/drug problems	3.31	3.15		5.06	2.68		−5.66**
Companions	1.73	1.64		2.65	1.46		−5.60**
Leisure/recreation	1.04	0.92		1.54	0.74		−5.71**
Family/marriage	1.44	1.45		2.08	1.42		−4.10**
Education/employment	2.2	2.5		3.5	2.69		−4.69**
Criminal history	2.63	2.38		4.22	2.04		−6.80**
Age	33.09	10.86		30.53	9.89		2.31*
Males			166 (64.84)			86 (67.72)		−0.56
Type of crime at baseline
Drug only			116 (45.31)			57 (44.88)		0.08
Nondrug only			134 (52.34)			69 (54.33)		−0.37
Both drug and nondrug			6 (2.34)			1 (0.79)		1.07
Worst offense at baseline
Felony			149 (58.20)			90 (70.87)		−2.41**
Misdemeanor			107 (41.80)			37 (29.13)		2.41**

Note. Drug only = charges related to alcohol or drugs, such as DUI and possession of paraphernalia. Nondrug only = charges unrelated to alcohol of drugs. LS/CMI = Level of Service/Case Management Inventory; DUI = driving under the influence.

Difference in Proportions Test.

p < .05. **p < .01.

Table 4.

Treatment as Usual Group Comparisons Among Those Who Recidivate (n = 89) and Did Not Recidivate (n = 41).

	Did not recidivate			Recidivate			t	z ^a
	M	SD	n (%)	M	SD	n (%)	t	z ^a
LS/CMI total score	27.39	6.12		28.55	6.45		−0.97
Antisocial patterns	2.83	1.05		2.64	1.03		0.97
Procriminal attitudes	2.10	1.28		2.24	1.19		−0.60
Alcohol/drug problems	5.24	2.38		5.72	2.02		−1.18
Companions	3.12	1.27		3.27	1.22		−0.63
Leisure/recreation	1.73	0.55		1.84	0.50		−1.14
Family/marriage	2.44	1.25		2.55	1.26		−0.47
Education/employment	4.78	2.34		5.06	2.46		−0.60
Criminal history	5.15	1.39		5.24	1.39		−0.34
Age	29.93	9.48		26.92	8.29		1.84
Males			29 (70.73)			69 (77.53)		−0.84
Type of crime at baseline
Drug only			13 (31.71)			24 (26.97)		0.56
Nondrug only			27 (65.85)			59 (66.29)		−0.05
Both drug and nondrug			1 (2.44)			6 (6.74)		1.01
Worst offense at baseline
Felony			25 (60.98)			56 (62.92)		−0.21
Misdemeanor			16 (39.02)			33 (37.08)		0.21

Note. Drug only = charges related to alcohol or drugs such as DUI and possession of paraphernalia. Nondrug only = charges unrelated to alcohol of drugs. LS/CMI = Level of Service/Case Management Inventory; DUI = driving under the influence.

Difference in Proportions Test.

p < .05. **p < .01.

Group comparisons also indicated that recidivating probations in the RJI condition (M = 1.89, SD = 1.24) had a significantly lower number of subsequent offenses (in-state and out of state misdemeanors and felonies) charged compared with those in the TAU condition who recidivated (M = 2.75, SD = 1.77; t = 3.98, p < .001).

Qualitative Analysis

A qualitative analysis of the probationers’ evaluation of the RJI was conducted to provide description of their experiences, as well as uncover aspects of the RJI the probationers found valuable. Content analysis was conducted following the suggestions of Elo and Kyngäs (2008). The analysis starts with open coding, wherein one of the authors and an undergraduate research assistant read and reread 2,360 responses from the postintervention evaluations while simultaneously generating categories. The categories generated in open coding were subsequently reduced and grouped under higher order headings. Another author and an undergraduate research assistant not involved in the open coding process then classified the qualitative data into the general categories previously generated. Intercoder agreement was excellent based on Fleiss’s (1981) kappa benchmark (Cohen’s κ = .855, p < .001). Categories that represented vague, ambiguous, and too general responses (e.g., “presenter was/was not effective,” “workshop was/was not effective,” and “it was good/bad”), and “no”/“none” responses were no longer reported.

Results of these qualitative analyses indicate that the responses can be classified under four general themes: (a) Empathy: “I understand the damage to my victims and community”; (b) Emotion: “I can identify, process, and express my feelings”; (c) Restorative Justice: “I understand what I need to do next”; and (d) Complaints about the class, program, and presenter. In this section, each of the themes will be discussed (see Table 5).

Table 5.

Themes From the Restorative Justice Intervention Class Evaluations.

	How was the class helpful? (n_response = 590)		What did you like the most? (n_response = 590)		What did you like the least? (n_response = 589)		Additional comments (n_response = 591)
	n	%	n	%	n	%	n	%
Empathy: I understand the damage to my victims and community	281	47.63	6	1.02	1	0.17	7	1.18
I can identify, process, and express my feelings	64	10.85	1	0.17	0	0	6	1.02
Restorative Justice: I understand what I need to do next	57	9.66	3	0.51	0	0	7	1.18
Complaints about the class, program, and presenter	1	0.17	10	1.69	389	66.04	34	5.75

Note. Responses and themes representing vague, ambiguous, or too general responses, and “no” or “none” responses were not reported.

Empathy: “I understand the damage to my victims and community.”

Responses under this theme included statements highlighting the detrimental effect (emotional, financial, or just in general) of the probationer’s actions on their victims and community. Most responses categorized under this theme were in response to the question “How was the class helpful?” Statements under this theme include: “Helped me understand the harm we caused to our victims,” “Helped me see that I hurt others,” and “Understanding the impact of crime on me and others around me.” In addition, responses in this category included statements suggesting how the program increased empathy toward victims, such as “Made me understand the victim’s point of view” and “I understand that everyone belongs and have value.”

Emotion: “I can identify, process, and express my feelings.”

Responses under this theme included statements highlighting probationer’s improved self-awareness. Most responses in this section were in response to the question “How was the class helpful?” Examples of this include: “Opened my eyes,” “I learned about myself,” and “Helped me understand my problems/situation.” Responses in this category also included behavioral components (not related to damage to victims or community) such as: “Helped me deal with my anger/emotion,” “Helped me process my feelings,” and “Helped me prepare for future mistakes.”

Restorative justice: “I understand what I need to do next.”

Responses under this theme included statements suggesting awareness of the reparations that probationers should do, such as: “I can write a meaningful and effective apology letter” or “I understand my obligation to my victim/community.” In addition, responses also included knowing more about and using the term “Restorative Justice.” Most of the responses in this theme were from the question “How was the class helpful?”

Complaints about the class, program, and presenter

Responses in this category included statements complaining about the class environment, schedule, or time. Examples included the following: “It was too long/short,” “Hard chairs,” “Other people in the workshop not participating,” and “It was too early in the morning.” Responses in this category also included statements complaining about program activities such as: “Ripping up the family cards” and “apology letter,” as well as presenter characteristics such as making them “Feel bad,” “Being called a liar,” and being “Too harsh.” Most of the responses under this category were from the question “What did you like least?” Although over 66% of RJI probationers had some sort of complaint about the class, indicating some sort of dissatisfaction, a significant reduction in recidivism was found with the RJI used in this study. Finally, probationers were also asked to anonymously rate the effectiveness of the RJI presenter. Results indicated that over 85% of RJI probationers rated the presenter as excellent.

Discussion

Consistent with outcomes of other RJI studies (Bergseth & Bouffard, 2013; Forgays & DeMilio, 2005; Koss, 2014; Rodriguez, 2005, 2007), recidivism rates in the RJI condition were lower than recidivism rates in the TAU condition. The present study extends current knowledge on the long-term durability of RJI’s effect by showing substantive reduced recidivism rate differences were identified at the 2-year follow-up and maintained across 6 years. Our data also showed novel RJI harm-reduction effects among re-offenders as compared with TAU. Among those who recidivated, those in the RJI condition did so less frequently than those in the TAU condition. This data suggest that the sole focus on recidivism in a dichotomized way may underestimate the benefit of an RJI in terms of reduced offender harm. Moreover, it remains to be seen whether repeated exposure to this brief RJI would transition the harm-reduction effects toward no recidivism across time. Within the present study, our results are suggestive of this possibility. The reduction in recidivism and the harm-reduction component found in the RJI condition highlight the importance of brief RJIs being routinely implemented within the criminal justice system.

Several studies have also examined the ability of RJIs to increase offender empathic knowledge (i.e., the impact of their crime on victims and society; Calhoun & Pelech, 2010; Jackson & Bonacker, 2006; Martinez et al., 2014; Pithers, 1999; Robinson & Rogers, 2015). Although the present study did not compare pre- and postintervention measures of empathy, our data showed that RJI probationers spontaneously responded with restorative justice concepts in a postintervention evaluation. More specifically, nearly 50% of postintervention responses included an empathic component. These response patterns are similar to other recidivism studies that measure empathy (53% of participants; Daly, 2002). However, an important distinction between the present study and others is that the present study appeared to be less influenced by potential response biases. For example, Calhoun and Pelech (2010) explicitly asked participants about empathy, whereas the present study had nonspecific questions (e.g., What did you like about the class?).

Despite these promising findings, we acknowledge certain limitations that temper our discussion. Due to the inherent selection bias in the present study, covariates were added in analyses to address potential differences between the TAU and RJI conditions. In addition, the proportion of individuals in the treatment condition outnumbered those in the TAU condition. To offset these methodological challenges, lack of random assignment and unequal group size were accounted for in the analyses. One may also question the degree to which the effects seen in the treatment condition were due to one superior facilitator. Our excessively high excellent rating of the presenters (over 85%) suggests that all presenters were viewed as extremely effective. Finally, as noted earlier, certain considerations should be made when working with certain types of offender populations (e.g., sexual offenders; Covell & Scalora, 2002; Mann & Barnett, 2012; L. E. Marshall et al., 2009). The present study was not designed to remove sexual offenders from the RJI that was preformed in probation, which raises some concern about inadvertent harm via increased empathy among this subgroup of probationers. Our data were not suggestive of such a risk given that sexual offenders did not necessarily show a greater than usual recidivism rate in the RJI condition.

In conclusion, the present study showed evidence that a brief RJI compared with TAU produced (a) long-term lower recidivism rates and (b) fewer reoffending among those that recidivated. Future studies should replicate this RJI intervention with a more stratified sample (vs. one Plains state in the United States) and with a randomized sample. Future researcher should also attempt to link empathy responses to subsequent recidivism rates and to show changes in empathy pre- to postintervention. Such changes will add to grow evidence for a more cost-effective RJI solution.

Footnotes

Acknowledgements

We would like to thank the Community Justice Center, James Jones, and Nebraska State Probation for their assistance with data collection and treatment delivery.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Andrews

D. A.

Bonta

Wormith

S. J.

(2004). The Level of Service/Case Management Inventory (LS/CMI). Toronto, Ontario, Canada: Multi-Health Systems.

Andrews

D. A.

Bonta

Wormith

S. J.

(2008). The Level of Service/Case Management Inventory (LS/CMI) supplement: A gender-informed risk/need/responsivity assessment. Toronto, Ontario, Canada: Multi-Health Systems.

Androff

D. K.

(2012). Reconciliation in a community-based restorative justice intervention. Journal of Sociology & Social Welfare, 39(4), Article 5.

Beck

A. J.

Shipley

B. E.

(1989). Recidivism of prisoners released in 1983. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Bergseth

K. J.

Bouffard

J. A.

(2013). Examining the effectiveness of a restorative justice program for various types of juvenile offenders. International Journal of Offender Therapy and Comparative Criminology, 57, 1054-1075.

Bonta

Wallace-Capretta

Rooney

McAnoy

(2002). An outcome evaluation of a restorative justice alternative to incarceration. Contemporary Justice Review, 5, 319-338.

Braithwaite

(2000). Survey Article: Repentance Rituals and Restorative Justice. The Journal of Political Philosophy, 8(1), 115-131.

Calhoun

Pelech

(2010). Responding to young people responsible for harm: A comparative study of restorative and conventional approaches. Contemporary Justice Review, 13, 287-306.

Clamp

Doak

(2012). More than words: Restorative justice concepts in transitional justice settings. International Criminal Law Review, 12, 339-360.

10.

Covell

C. N.

Scalora

M. J.

(2002). Empathic deficits in sexual offenders: An integration of affective, social, and cognitive constructs. Aggression and Violent Behavior, 7, 251-270.

11.

Daly

(2002). Mind the gap: Restorative justice in theory and practice. In von Hirsch

Roberts

Bottoms

A. E.

Roach

Schiff

(Eds.), Restorative justice and criminal justice: Competing or reconcilable paradigms? (pp. 219-236). Oxford, UK: Hart Publishing.

12.

Daly

(2016). What is restorative justice? Fresh answers to a vexed question. Victims & Offenders, 11, 9-29.

13.

Dhami

M. K.

Mantle

Fox

(2009). Restorative justice in prisons. Contemporary Justice Review, 12, 433-448.

14.

Durose

M. R.

Cooper

A. D.

Snyder

H. N.

(2014). Recidivism of prisoners released in 30 states in 2005: Patterns from 2005 to 2010. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

15.

Elo

Kyngäs

(2008). The qualitative content analysis process. Journal of Advanced Nursing, 62, 107-115.

16.

Feder

Wilson

D. B.

(2005). A meta-analytic review of court-mandated batterer intervention programs: Can courts affect abusers’ behavior? Journal of Experimental Criminology, 1, 239-262.

17.

Fleiss

J. L.

(1981). Balanced incomplete block designs for inter-rater reliability studies. Applied Psychological Measurement, 5, 105-112.

18.

Forgays

D. K.

DeMilio

(2005). Is teen court effective for repeat offenders? A test of the restorative justice approach. International Journal of Offender Therapy and Comparative Criminology, 49, 107-118.

19.

Jackson

A. L.

Bonacker

(2006). The effect of victim impact training programs on the development of guilt, shame, and empathy among offenders. International Review of Victimology, 13, 301-324.

20.

Koss

M. P.

(2014). The RESTORE program of restorative justice for sex crimes: Vision, process, and outcomes. Journal of Interpersonal Violence, 29, 1623-1660.

21.

Lagan

P. A.

Levin

D. J.

(2002). Recidivism of prisoners released in 1994. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

22.

Latimer

Dowden

Muise

(2005). The effectiveness of restorative justice practices: A meta-analysis. The Prison Journal, 85, 127-144.

23.

Mann

R. E.

Barnett

G. D.

(2012). Victim empathy intervention with sexual offenders: Rehabilitation, punishment, or correctional quackery? Sexual Abuse: A Journal of Research and Treatment, 25, 282-301.

24.

Marshall

L. E.

Marshall

L. E.

Serran

G. A.

O’Brien

M. D.

(2009). Self-esteem, shame, cognitive distortions and empathy in sexual offenders: Their integration and treatment implications. Psychology, Crime & Law, 15, 217-234.

25.

Marshall

L. E.

Marshall

W. L.

(2011). Empathy and antisocial behavior. The Journal of Psychiatry and Psychology, 22, 742-759.

26.

Marshall

(1999). Restorative justice: An overview (Occasional paper). London, England: Home Office.

27.

Martinez

A. G.

Stuewig

Tangney

J. P.

(2014). Can perspective-taking reduce crime? Examining a pathway through empathic-concern and guilt-proneness. Personality and Social Psychology Bulletin, 40, 1659-1667.

28.

Morse

J. M.

(2003). Principles of mixed methods and multimethod research design. In Tashakori

Teddlie

(Eds.), Handbook of mixed methods in social and behavioral research (pp. 189-208). Thousand Oaks, CA: Sage.

29.

Pithers

W. D.

(1999). Empathy: Definition, enhancement, and relevance to the treatment of sexual offenders. Journal of Interpersonal Violence, 14, 257-284.

30.

Redondo

Sanchez-Meca

Garrido

(1999). The influence of treatment programmes on the recidivism of juvenile and adult offenders: An European meta-analytic review. Psychology, Crime & Law, 5, 251-278.

31.

Robinson

E. V.

Rogers

(2015). Empathy faking in psychopathic offenders: The vulnerability of empathy measures. Journal of Psychopathology and Behavioral Assessment, 37, 545-552.

32.

Roche

(2001). The evolving definition of restorative justice. Contemporary Justice Review, 4(34), 341-353.

33.

Rodogno

(2008). Shame and guilt in restorative justice. Psychology, Public Policy, and Law, 14, 142-176.

34.

Rodriguez

(2005). Restorative justice, communities, and delinquency: Whom do we reintegrate? Criminology & Public Policy, 4, 103-130.

35.

Rodriguez

(2007). Restorative justice at work: Examining the impact of restorative justice resolutions on juvenile recidivism. Crime & Delinquency, 53, 355-379.

36.

Rossner

(2012). Emotions and interaction ritual: A micro analysis of restorative justice. British Journal of Criminology, 51, 95-119.

37.

Saulnier

Sivasubramaniam

(2015). Effects of victim presence and coercion in restorative justice: An experimental paradigm. Law and Human Behavior, 39, 378-387.

38.

Sherman

L. W.

Strang

Angel

Woods

Barnes

G. C.

Bennett

Inkpen

(2005). Effects of face-to-face restorative justice on victims of crime in four randomized, controlled trials. Journal of Experimental Criminology, 1, 367-395.

39.

Teddlie

Tashakkori

(2002). Major issues and controversies in the use of mixed methods in the social and behavioral sciences. In Tashakkori

Teddlie

(Eds.), Handbook of mixed methods in social and behavioral research (pp. 189-208). Thousand Oaks, CA: Sage.

40.

Toews

(2013). Toward a restorative justice pedagogy: Reflections on teaching restorative justice in correctional facilities. Contemporary Justice Review, 16, 6-27.

41.

Tsui

J. C.

(2014). Breaking free of the prison paradigm: Integrating restorative justice techniques into Chicago’s juvenile justice system. The Journal of Criminal Law and Criminology, 104, 635-666.

42.

Van Ness

D. W.

(2004, August 5-7). RJ City®: Contemplating a restorative justice system. Paper presented at Building a Global Alliance for Restorative Practices and Family Empowerment, Richmond, British Columbia, Canada.

43.

Van Stokkom

(2002). Moral emotions in restorative justice conferences. Theoretical Criminology, 6, 339-360.

44.

Vose

Cullen

F. T.

Smith

(2008). The empirical status of the Level of Service Inventory. Federal Probation: A Journal of Correctional Philosophy and Practice, 72(3), 22-29.

45.

Wondra

J. D.

Ellsworth

P. C.

(2015). An appraisal theory of empathy and other vicarious emotional experiences. Psychological Review, 122, 411-428.