Social Support During Reentry: Family,Mentor,Religious,Parole Officer,and Social Service Roles

Abstract

Existing research on social support and reentry primarily focuses on a single dimension of support, such as family or community support. Informed by the social support perspective, this study assessed how combined support from family, mentors, religious groups, parole officers, and social service agencies influences reentry outcomes. Given that the social support one receives during reentry is time-variant rather than static, longitudinal multilevel modeling was employed to examine how temporal changes in social support were associated with temporal changes in recidivism and drug misuse among people on parole. Results indicated that simultaneous support from family, parole officers, and social service agencies exerted protective effects on reentry outcomes. However, community-based mentoring programs had no significant effect on recidivism, and religious and social service support increased substance misuse. Policy implications derived from this research are discussed to address the intersection of various sources of social support.

Keywords

differing forms of social support reentry recidivism drug use

Mass incarceration policy in the United States produced an unprecedented growth in both the prison and jail populations. More than 2.2 million people were incarcerated in jails or prisons in 2018, a sixfold increase since the 1970s (Cloud et al., 2020). The tremendous growth in the U.S. prison population was followed by a high volume of people seeking reentry after incarceration (Huebner & Shannon, 2022). According to a series of reports on prison populations by the Bureau of Justice Statistics, from 2010 to 2017, nearly 625,000 people were released from prison each year, which averaged around 12,000 per week (Bronson & Carson, 2019; Carson, 2014; Carson & Golinelli, 2012). COVID induced an even greater number of people to transition out of prison: we saw a 15% decrease in the incarcerated population during 2020, which was the largest single-year decrease recorded since 1926 (Minton et al., 2021). Various reentry initiatives have been established to provide reentering persons with support and resources regarding health, housing, and food insecurity to facilitate their successful reentry (Berghuis, 2018). However, studies using both official data (Alper et al., 2018; Durose & Antenangeli, 2021; Durose et al., 2014; O’Connell et al., 2019) and self-reported recidivism measures (Cartier et al., 2006; Farabee et al., 2014; Loughran et al., 2009) have consistently shown an alarmingly high rate of reentry failure. For example, by the fifth year after release, one study found that over three-fourths of formerly incarcerated people had been reincarcerated (LaCourse et al., 2019).

Studies explaining why some returning citizens achieve reentry success while others do not have underscored how support and resources play a vital role in severing from their criminal past (Harding et al., 2019; Wallace et al., 2016). Newly released individuals who had no access to social services and received little support from their family and community were at a higher risk of resorting to illegal activities and substance misuse to cope with traumatic experiences, such as food insecurity, homelessness, social isolation, and mental health struggles (Fries et al., 2014; Moschion & Johnson, 2019). In addition, social support studies have revealed that the emotional, relational, and material supports returning citizens receive came largely from family and friends, social service agencies, community organizations, and reentry programs (Bares & Mowen, 2020). While a plethora of studies were dedicated to understanding the impact of a single dimension of social support, such as family (Harding et al., 2019; Sugie & Augustine, 2020), parole officer (Bares & Mowen, 2020; Chamberlain et al., 2018), or a religious organization (Clear et al., 2000), we know very little about the impact that joint effects of different forms of social support may have on the reentry experience. Do individual- and community-related supports both affect reentry? Do formal and informal supports have equivalent weight in reentry success? With limited research regarding the comparative effects of these different forms of support, policymakers and practitioners are ill-equipped to develop evidence-based programs of effective social support for facilitating reentry.

This study sought to fill this gap and contribute to the emerging understanding of social support and reentry outcomes. It expanded prior work that focused primarily on single domains of social support to explore how combinations of different forms of support (family, community organizations, parole officers, and social services) influence reentry outcomes. This study also assessed the longitudinal change in a predictor’s effect on reentry outcomes. Prior research revealed that the transition from prison to society is a dynamic rather than static process. During reentry, individuals may tap into various forms of resources while they rebuild relationships with family and friends and carve out new daily routines. Consequently, the forms of support an individual receives may vary over time (Harding et al., 2019; Western, 2018; Western & Sirois, 2019). Support may also vary across individuals (Mowen & Boman, 2018). Thus, some people may have access to more sources of support and among those with the same sources of support, variations in the duration of support may contribute to different levels of reentry failure. To explore how these ways reentry are affected by forms and durations of social support, we adopted a longitudinal multilevel modeling strategy that used multiple waves of data.

Desistance and Social Support

Based on their pathbreaking analysis of Gluecks’s data, Sampson and Laub (1993) developed an age-graded theory of informal social control, which explained desistance from a control perspective. According to the theory, when mutual interdependencies are developed with pro-social social institutions, such as employment and marriage, individuals tend to veer away from crime: such attachments create obligations and expectations that are incompatible with a deviant lifestyle (Sampson & Laub, 1993). Their theory provided limited insight, however, on how individuals develop such prosocial attachments. After all, this only happens after the individual enters conventional social institutions. Drawing on insights from Coleman’s (1990) social capital perspective, Cullen (1994); Pratt & Cullen (2005); Wright & Cullen (2006) discovered that social support enabled individuals to enter conventional social institutions. Family, friends, and community organizations and agencies provided vital sources of formal and informal support that linked reentering persons to employment opportunities and transitional housing (Wright & Cullen, 2006). A returning citizen with sufficient social support thus would be more likely to have employment opportunities than someone who received little social support.

Differing Forms of Social Support

Social support is generally defined as the receipt of assistance from social connections (Sarason et al., 1987, p. 817). The assistance can be material, emotional, or both (Jacobson, 1986; Vaux, 1988). Social support has been measured by both the closeness of the individual’s relationships and the availability of different sources of support (Johnson Listwan et al., 2010; Longmate et al., 2021). The more sources of social support individuals can draw from and the stronger the connections are, the more likely they can cope with difficulties during life transitions (Tsai et al., 2012).

Reentry is unquestionably characterized by dynamic challenges. Newly released individuals draw on various sources of support to combat social isolation, addiction, mental health issues, food and housing insecurity, and joblessness (Bushway & Uggen, 2021; Western, 2018). Supportive relationships play a vital role. When postincarcerated persons receive social support, they find comfort and understanding when encountering adversity, which decreases their risk of resorting to criminal coping strategies (McKay et al., 2016; Mowen et al., 2019).

A key characteristic of social support is that it is time-variant rather than static (Cornwell, 2003; Martínez et al., 2011; Sugie & Augustine, 2020). Levels of social support fluctuate considerably over time, especially during life transitions, and these fluctuations affect an individual’s well-being in profound ways (Gariepy et al., 2016; Martínez et al., 2011). For example, an increase in social support alleviates mental health struggles, whereas a decrease exerts a deleterious effect (Kail & Carr, 2020).

During the transition from prison, there is typically an initial burst of time spent with family followed after a few weeks by less time (Sugie & Augustine, 2020). Informal emotional support from family and friends fluctuates in response to the accumulated burdens of helping returning loved ones (Liu & Visher, 2019; Western et al., 2015; Wyse et al., 2014). Formal support from social services, parole officers, and community organizations can also vary over time (Harding et al., 2019; Western, 2018; Western & Sirois, 2019). Therefore, the time-variant characteristics of social support cannot be ignored when studying reentry experiences of postincarcerated individuals.

While the vast majority of studies on social support during reentry focused primarily on family support, multiple other sources of support are pertinent, including personal (family members, prosocial friends, partners) and organizational (parole officers, mental health agencies, religious organizations) (Longmate et al., 2021; Skeem et al., 2009). To date, few reentry studies have assessed the collective impact of multiple sources of social support. Research on individual forms of social support, however, offer useful insights.

Prior studies on family support, for example, identified it as a key contributor to reentry success. These include both qualitative and quantitative studies based on range of population samples (Sugie & Augustine, 2020; Wallace et al., 2016). For returning citizens, family support has been found to suppress recidivism (Shollenberger, 2009), increase the chances of securing a job (Berg & Huebner, 2010), and alleviate mental health struggles (Wallace et al., 2016). However, other more recent studies have shown that family can be criminogenic and increase the odds of reincarceration (Liu & Visher, 2021; Mowen & Boman, 2019). When family members are themselves involved in offending or drug use and provide little support, reentering persons may face a higher risk of reentry failure (Liu & Visher, 2021). The criminogenic effect of family has been documented in analyses based on both the Returning Home and Serious and Violent Offender Reentry Initiative (SVORI) data sets, both large, multi-state data sets on reentry (Mowen & Boman, 2019; Mowen & Visher, 2015). These findings suggest the riskiness of the family environment should be included in future studies on reentry outcomes.

Parole officer support has also been shown to be critical in facilitating reintegration. When individuals on parole perceive their officer as supportive, they may be more likely to confide in the officer and communicate their needs (Bares & Mowen, 2020; Liu et al., 2021; Robinson, 2005); perceive the officers as legitimate; and believe the officer can guide them toward a prosocial life by setting sequential goals of reentry, moving beyond their comfort zone, and avoiding complacency (Lovins et al., 2018). In contrast, individuals who perceive their officers as controlling, coercive, and not invested in building a relationship will not experience a positive relationship with the officer or receive much benefit from it (Chamberlain et al., 2018).

Numerous studies on the link between parole officer support and reentry outcomes demonstrated a positive association between positive reentry outcomes and officers’ engagement with clients (Blasko et al., 2015; Vidal et al., 2015). For example, in a study of justice-involved juveniles, Vidal et al. (2015) discovered that respondents who perceived positive relationships with their parole officers had lower rates of reoffending. Similarly, Blasko et al. (2015) discovered that people on parole who perceived stronger relationships with officers had lower substance misuse and recidivism rates.

Support from religious communities has also received research attention, with mixed results. Some studies evinced an association between religious activities and reduction in prison misconduct and postrelease recidivism (Camp et al., 2008; Mowen et al., 2019). Camp and colleagues (2008), for example, reported a negative relationship between the religious involvement of incarcerated individuals and prison misconduct. Camp et al. (2008), Mowen et al. (2019), and O’Connor (2005) discovered links between engagement in religious communities and reductions in crime and substance misuse. Other studies, however (e.g., Johnson et al., 1997), reported that involvement in religious activities had no effect on reentry success.

Mentoring programs provide another source of support for returning individuals. In mentoring programs, reentering persons are assigned a mentor who uses their experience, expertise, and wisdom to guide and help them in the transition from prison to the community (Garcia, 2016). These programs are generally provided by nonprofit organizations and take a variety of forms (Victor et al., 2021). Some pair new returning citizens with community members or volunteers (Hucklesby & Wincup, 2014); others pair them with formerly incarcerated mentors to leverage the mutuality derived from shared life experiences (Buck, 2020; Hucklesby & Wincup, 2014). Studies of these programs have demonstrated that peer mentors helped returning citizens strengthen self-esteem and hopefulness (Cooke & Farrington, 2016; Lopez-Humphreys & Teater, 2018) and promoted desistance (Nixon, 2020). Mentees with formerly incarcerated mentors reported promising effects. They described enhanced self-efficacy (Marlow et al., 2015), better coping strategies (Maruna, 2015), and less personal stigma (Maruna, 2015).

Social welfare programs also play a vital role in an individual’s transition from prison to community. Access to shelter, health care, employment programs, counseling services, and financial aid may reduce the risk of homelessness, mental health struggles, substance misuse, and food insecurity (Wolff & Draine, 2004). Nonetheless, extant empirical studies on social welfare and recidivism are sparse and their findings are inconclusive. Some evidence suggests these formal forms of support not only facilitate economic reintegration but also inhibit recidivism. For example, Bullis and Yovanoff’s (2002) longitudinal study of more than 500 justice-involved youth in Oregon documented that those who had received mental health services were 4.8 times more likely to be working 1 year after release than those who did not receive the services. However, another study of more than 13,000 postincarcerated youth in Illinois reported that respondents who received social welfare benefits (e.g., child welfare, public assistance, and Medicaid-related services) had higher recidivism rates (Cusick et al., 2009).

The present study built on the literature by assessing the multi-dimensional nature of social support. Specifically, we examined the availability of support for reentering persons from family, corrections and social service agents, mentors, and community organizations, and how disparities in access to such support were associated with reentry outcomes. Since social support is a time-variant rather than static commodity, we also filled another gap in the literature by examining the effects of short- versus long-term support.

Method

Data and Sample

Data for this study came from the SVORI (Lattimore & Steffey, 2010). SVORI was a federally funded program that helped states develop programs and policies to smooth the transition from prison to the community and create better reentry outcomes for men, women and youth (Lattimore & Steffey, 2010). SVORI respondents were selected from 12 states receiving funding between July 2004 and November 2005. The SVORI evaluation included respondents who received in-prison SVORI programming and a comparison group that did not. Three postrelease in-person interviews were conducted 3, 9, and 15 months after reentry (hereafter T1, T2, and T3, respectively). For each survey, the interviews occurred regardless of whether the respondent had been re-incarcerated. Information was collected about reentry experiences (focusing on family relationships), social services received, involvement in community activities, and interaction with parole officers. The sample selected for the current study was limited to male respondents¹ who were on parole. Parole status was an eligibility criterion because the strength of parole officers’ support is a primary focus of the study and parole supervision is associated with sentence severity and criminality. We also opted to exclude SVORI respondents who were not on parole as a comparison group in the model. Including them would have required making the assumption that risk factors affect both groups in the same way, regardless of parole status. This is unlikely given that sentence severity and criminality are associated with release conditions. To avoid comparing apples to oranges, we thus excluded SVORI men who were not under parole from the study sample, which resulted in a sample of 817 respondents.

Dependent Variables

Two reentry outcomes were included as dependent variables: recidivism and postrelease substance misuse. T1, T2, and T3 recidivism were measured by an item asking, “Since release/last interview, have you committed any violent crimes, regardless of whether or not you were caught?” These three measures of self-reported recidivism (which excluded technical parole violations) captured the periods from release to T1 interview, T1 to T2 interview, and T2 to T3 interview, respectively. We used self-reported recidivism instead of official rearrest records for several reasons. Although the SVORI data included official rearrest data, it did not indicate whether the rearrest was resulted from a new offense or was a technical parole violation. Furthermore, arrest data are subject to bias resulting from disparate policing strategies across different types of neighborhoods (Matsuda et al., 2022). Individuals from disadvantaged neighborhoods may be more likely to be rearrested due to more aggressive policing practices. Respondents’ self-reported recidivism is thus a better proxy of reentry outcome. T1, T2, and T3 past 30-day drug use measured respondents’ use of any type of drug (sedatives, tranquilizers, stimulants, pain relievers, methadone, marijuana, hallucinogens, cocaine, heroin, amphetamines, and inhalants) over the 30 days prior to the T1, T2, and T3 interview.

Independent Variables

T1, T2, and T3 family support: Following the practice of prior SVORI studies on family support (Mowen & Boman, 2019; Mowen et al., 2019), we used four items to construct the latent concept of family support. These items asked respondents whether they agreed with the following statements: (1) I feel close to my family; (2) I have someone in my family to talk to about myself or my problems, (3) I have someone in my family to turn to for suggestions, and (4) I have someone in my family who understands my problems. Response options were on a 4-point Likert-type scale (strongly disagree, disagree, agree, and strongly agree; α = .86–.89 across all waves).

T1, T2, and T3 social service benefits received: These variables measured the scope of social service benefits respondents received. Questionnaire items asked about 15 categories of social services: public financial assistance, public health care, housing, free transportation, cloth/food, and social worker’s help to get driver’s license, to provide employment information, to put together resumes, to obtain documents, to find mentors, to enhance education, to enroll in programs to change criminal attitude, to manage anger, to handle personal relationships, and to enhance other life skills. Each was a yes/no question (0 = no, 1 = yes). Items were summed to form the final scale.

T1, T2, and T3 support from parole officers: A measure of officers’ support developed by the Urban Institute (Yahner et al., 2008) was used to construct support from parole officers. This scale was based on seven items, which asked the extent to which the respondent agreed that (1) the parole agent was helpful with his transition, (2) the parole agent seemed trustworthy, (3) the parole agent gave him correct information, (4) the parole agent acted too busy to help him, (5) the parole agent treated him with respect, (6) the parole agent acted professionally, and (7) the parole agent didn’t listen to him. The answers ranged from (1) strongly disagree to (4) strongly agree. The fourth and seventh questions were recoded so that a higher score indicated a higher level of support, and the items were summed to form the final scale. The items exhibited good internal reliability (α = .85–.87 across waves).

T1, T2, and T3 participation in community-based mentoring programs: T1, T2, and T3 participation in mentoring programs were measured by a question posed in all three postrelease interviews: “Now I’d like to ask you about any community activities you may have taken part in since your release from incarceration. Have you ever participated in peer mentoring programs?” Response options included (1) yes and (0) no.

T1, T2, and T3 involvement of religious organizations: These variables were measured by the question: “Now I’d like to ask you about any community activities you may have taken part in since your release from incarceration. Have you ever participated in the activities of a church, mosque, temple, or any other religious group?” Response options included (1) yes and (0) no.

Control Variables

To obtain unbiased results, a range of risk factors for recidivism and postrelease substance misuse was included in the models as controls. First, we controlled for the effects of prior mental well-being and substance misuse based on SVORI items on the T1 postrelease interview. Prerelease mental and emotional well-being was variable measured at prerelease interview, which was based on the question: In general, would you say your current emotional or mental health is excellent, very good, good, fair, or poor? Response options ranged from (1) poor to (5) excellent. Past substance misuse was a binary variable based on the question: Before the current incarceration, did you use illicit drugs? Second, we controlled for correctional programs in the model. SVORI participant was a binary (yes/no) covariate based on an SVORI item at the T1 interview indicating if the respondent had participated in the SVORI program while incarcerated (49% of the sample were SVORI participants). Third, given that risky family environment is a salient predictor of engagement in crime and drug use, we controlled for a range of family environmental factors based on SVORI items in the T1 interview asking if a family member was ever incarcerated (24% = yes), a family member was currently using drugs (4% = yes), a family member was currently engaging in illicit activities (1% = yes), and a family member used alcohol in respondents’ presence (16% = yes). These were binary (yes/no) variables. Finally, we controlled respondents’ demographic characteristics, including age, education attainment (in years), and race (1 = White, 2 = Black, 3 = Other). For the race variable, the Other category included Hispanic, Asian, Pacific Islander, Native American, and multi-racial, which were not isolated due to the small numbers of respondents in the sample. We concede the drawback of this coding strategy. Experiences in the criminal legal system vary by race and ethnicity. Grouping people into an Other category obscures their experiences. Therefore, we encourage future studies to avoid using this practice when the size of race and ethnic groups are more balanced in the data.

Missing Data

Panel studies inevitably suffer from respondent attrition, and missing data within the SVORI sample has been well-documented (Lattimore & Steffey, 2009). Of the 817 male SVORI respondents who were on parole, about 72% (585 respondents) participated in all three postrelease interviews. An attrition rate of 28% was not surprising given that returning persons may have difficulty participating if they are challenged by homelessness and food insecurity. In addition, 86 of the 585 respondents did not respond to items used to construct the study variables. Following the practice of prior panel studies using SVORI data (Boman & Mowen, 2018; Stansfield et al., 2018), we finalized the study sample using list-wise deletion, which resulted in 499 respondents. Researchers have consistently found that sample attrition in SVORI was random and therefore unlikely to introduce bias to study results (Mowen et al., 2018; Wodahl et al., 2021). The sensitivity tests conducted in this study confirmed this: we discovered no noticeable differences in age, race, SVORI participation, or education level between respondents included and not included in the study (these supplemental analysis results are available upon request).

Analytical Strategies

This study employed longitudinal multilevel modeling, an analytical approach widely used by criminologists to model panel data in which there are repeat measures of a respondent over several waves of data collection (Berg & Loeber, 2011; Janssen et al., 2021; Liu et al., 2022). In this modeling framework, each respondent was a cluster (a Level 2 unit) and repeated measures of this person were Level 1 units. Longitudinal multilevel modeling, which collectively implements arrays of both group- (at Level 1) and grand-mean centered (at Level 2) variables, produced “unbiased estimates of the associations between the predictors and outcomes” because it accounted for unobserved heterogeneity in respondents by eliminating preexisting differences from the within-person estimator (Raudenbush & Bryk, 2002, p. 183).

In this study, a respondent was considered a cluster whose repeated measures were Level 1 units within a cluster. A Level 1 predictor was person-specific-mean-centered. Its coefficient was at the person-period level: $x_{person - period}$ = $x_{j i}$ - $\bar{x_{j}}$ . In this equation, ji represented the ith measure of individual j. A Level 2 predictor was grand-mean centered. Its coefficient was at the between-person level: $x_{between - person}$ = $\bar{x_{j .}}$ - $\bar{x_{\cdot\cdot}}$ It measured the extent to which an individual j’s average value of a variable deviated from $\bar{x_{\cdot\cdot}}$ , the average value of the sample. Thus, both heterogeneity across respondents and the temporal dynamics within the same person were accounted for.

A two-part analytical strategy was implemented to fulfill our research purposes. First, we conducted descriptive analyses and generated binary visualizations of the preliminary results concerning the link between social support and reentry outcomes. Second, using longitudinal multilevel models, we simultaneously assessed the heterogeneity of reentry experiences across respondents as well as the temporal dynamics in social support. This approach provided a stringent test of the association between divergent sources of social support and reentry outcomes.

Results

Preliminary Results

Table 1 reports the descriptive statistics for all variables used in this analysis. The largest proportion of respondents was Black (50%), followed by White (38%) and Other race respondents (12%). Nearly, 55% of the respondents had been SVORI program participants while incarcerated. On average, respondents had 12 years of education, equivalent to a high school diploma. There was a noticeable prevalence of familial risk factors: about 24% of respondents lived with family members who had an incarceration history, 16% said their family members used alcohol in their presence, and 4.27% had family who currently used drugs.

Table 1:

Descriptive Statistics (N = 499)

Variables	M or percentage	SD	Minimum	Maximum
Dependent variables
T1 recidivism	17.24%
T2 recidivism	23.37%
T3 recidivism	30.27%
T1 substance misuse	18.77%
T2 substance misuse	29.89%
T3 substance misuse	37.55%
Independent variables
Person-period variables
Family support	−0.01	0.40	−1.73	1.41
Parole officer support	0.05	0.39	−1.63	1.33
Social services received	0.46	1.36	−4.67	5.00
Involvement in peer mentoring communities	0.00	0.27	−0.15	0.85
Involvement in religious communities	0.00	0.38	−0.33	0.67
Between-person variables
Family support	0.07	0.68	−2.73	0.84
Parole officer support	0.04	0.74	−2.36	1.33
Social services received	0.07	1.91	−1.60	8.40
Involvement in peer mentoring communities	0.00	0.24	−0.67	0.67
Involvement in religious communities	0.00	0.28	−0.67	0.67
Control variables
Used drugs before the current incarceration	94.05%
Prerelease emotional and mental health	2.44	1.10	1.00	5.00
Family members ever incarcerated	23.73%
Family members currently use drugs	4.27%
Family members currently engage in illicit acts	0.85%
Family members use alcohol in respondent’s presence	15.55%
SVORI participant	55.17%
Education attainment	12.18	2.02	6.00	18.00
Age	29.63	7.68	19.00	69.00
White	38.31%
Black	49.81%
Other	11.88%

Note. SVORI = Serious and Violent Offender Reentry Initiative.

We found nontrivial temporal changes in reoffending, substance misuse and the level of social support respondents received. During the three time periods—release—T1, T1 to T2, and T2 to T3—17.24%, 23.37%, and 30.27% of the respondents committed new offenses, respectively, and 18.77%, 29.89%, and 37.55% began to misuse substances. Recall that the recidivism and substance misuse measures were noncumulative. The upward trend of the two outcomes suggested that returning citizens were at a particularly high risk for reentry failure in the first 15 months after release (the T3 interview was conducted during the 15th month after release), echoing findings from past studies (Loughran et al., 2009; Paternoster et al., 2016). Social support variables exhibited varied temporal patterns: some supports decreased from T1 to T3 (viz., support from family and parole officers) while others fluctuated (social services and involvement in peer mentoring communities) or stayed stable after reaching a point (involvement in religious communities) (see Figure 1).

Figure 1:

Temporal Changes in Differing Sources of Social Supports

Multilevel Model Results on Social Support and Reentry Outcomes

Model 1 in Table 2 presents the longitudinal multilevel model results using various forms of social support to predict recidivism. We expected to find that with the between-person effects controlled, there would be a significant person-period effect of social support. In other words, the temporal changes in social support should have predicted temporal changes in recidivism over the three waves. The results largely supported this expectation, suggesting that multiple sources of social support exerted significant inhibitory effects on recidivism at the person-period level. When respondents received a one-unit increase in family and parole officer support, their concurrent odds of reoffending were 40% (OR = .60) and 51% (OR = .49) lower, respectively. When social service benefits increased by one unit, the concurrent risk of recidivism dropped by 14% (OR = .86). In contrast, involvement in religious communities or mentoring programs did not exert significant effects at the person-period level.

Table 2:

Hierarchical Model Results Predicting Recidivism and Drug Use During Reentry

	Model 1: Recidivism		Model 2: Substance misuse
Predictors	OR	95% CI	OR	95% CI
Level 1 (person-period) predictors
PP^a family support	0.60***	[0.44, 0.82]	0.66***	[0.48, 0.91]
PP parole officer support	0.49***	[0.36, 0.67]	0.57***	[0.41, 0.78]
PP social services received	0.86**	[0.78, 0.96]	0.96	[0.86, 1.07]
PP involvement in peer mentoring communities	0.86	[0.51, 1.46]	2.18***	[1.26, 3.77]
PP involvement in religious communities	1.06	[0.67, 1.70]	1.91***	[1.18, 3.1]
Level 2 (between-person) predictors
BP family support	0.60*	[0.4, 0.92]	0.37***	[0.24, 0.59]
BP parole officer support	0.85	[0.56, 1.28]	0.95	[0.61, 1.46]
BP social services received	0.95	[0.82, 1.09]	0.97	[0.83, 1.13]
BP involvement in peer mentoring communities	0.93	[0.37, 2.33]	1.12	[0.42, 2.99]
BP involvement in religious communities	0.53	[0.28, 1.01]	0.54	[0.27, 1.07]
Prerelease mental and emotional health	1.13	[0.92, 1.39]	0.92	[0.74, 1.15]
Used drugs before the current incarceration	11.80***	[2.64, 52.76]	7.70***	[2.2, 26.95]
Family members ever incarcerated	1.31	[0.72, 2.39]	1.23	[0.64, 2.35]
Family members currently use drugs	1.35	[0.41, 4.48]	0.91	[0.25, 3.33]
Family members currently engage in illicit acts	14.25*	[1.37, 147.89]	7.2	[0.87, 59.71]
Family members use alcohol in R’s presence	2.36***	[1.36, 4.07]	2.76***	[1.52, 5.00]
Age	0.93***	[0.9, 0.96]	0.97	[0.94, 1.01]
Black	0.59*	[0.35, 0.97]	0.97	[0.57, 1.67]
Other	0.74	[0.36, 1.53]	0.63	[0.29, 1.41]
Education level	0.94	[0.83, 1.05]	0.86*	[0.76, 0.97]
SVORI participant	0.64	[0.41, 1.01]	0.84	[0.74, 1.94]
Random effects
Person-period intercept	1.93***(0.11)		2.16***(0.12)
Between-person intercept	—^b		—
Fit indices
−2 log likelihood	1,384.26		1,448.53
AIC	1,430.26		1,498.53
BIC	1,535.49		1,612.81

Note. OR = odds ratio; CI = confidence interval; SVORI = Serious and Violent Offender Reentry Initiative; AIC = Akaike information criterion; BIC = Bayesian information criterion.

PP: person-period; BP: between-person. ^b Because the outcome is a dichotomous variable, the multilevel model does not provide a meaningful individual-level variance component. If the Level 1 model is conceived of in terms of a latent variable (Raudenbush & Bryk, 2002), then the Level 1 random effect can be assumed to have a standard logistic distribution with a mean of 0 and variance of π2/3.

p < .05. **p < .01. ***p < .001.

At the between-person level, family support appeared to exert a protective effect against recidivism. Respondents who received higher-than-average family support had a lower-than-average risk of recidivism (OR = .60). Furthermore, respondents with higher-than-average engagement in religious communities had lower odds of recidivism. However, the p value for this coefficient was .054, indicating the evidence was not strong. Among other time-invariant variables, a risky familial environment placed returning citizens at a high risk of reentry failure. Family members’ illegal acts (OR = 14.25) and alcohol use (OR = 2.36) were both significantly and positively associated with the odds of recidivism. It is worth noting that respondents who reunited with family members engaging in crime were 14 times more likely to reoffend than those who were not exposed to a risky familial environment. Returning citizens’ history of substance misuse also predicted their recidivism. Those who had substance misuse histories were 11 times more likely to reengage in criminal activity compared to their peers (OR = 11.80). Older respondents had significantly lower odds of recidivism than younger respondents (OR = .93), and Black respondents had significantly lower odds than their White peers (OR = .59).

Model 2 examined the predictors’ effects on the odds of substance misuse during reentry. As shown in Table 2, the effect of social support on substance misuse differed from its effect on recidivism. Some sources of social support exerted a significant protective effect at the person-period level. During their initial 15 months of reentry, respondents who received a one-unit increase in family support were significantly less likely to misuse substances during that period (OR = .66). Respondents who received a one-unit increase in parole officer support had 43% less risk of substance misuse during the 15-month period (OR = .57). In contrast, involvement in local church activities (OR = 1.91) or community-based mentoring programs (OR = 2.18) appeared to increase the risk of substance misuse.

At the within-person level, respondents who received higher-than-average family support following their release had a lower-than-average likelihood of misusing substances during the 15-month period (OR = .37). The other four forms of social support exerted no significant effect on substance misuse at the between-person level. Among the time-invariant variables, preincarceration substance misuse was the only significant predictor of substance misuse after release (OR = 7.70). Those who engaged in substance misuse before their current incarceration were 7.70 times more likely to engage in substance misuse during reentry. When past substance misuse history and other covariates were controlled for in the model, prior mental and emotional health did not reach significance. A risky familial environment emerged to be a predictor of postrelease substance misuse. Family members’ alcohol use (OR = 2.76) increased the likelihood of returning citizens’ engagement in substance misuse. Among the demographic factors, education achieved significance. Respondents with higher education had a significantly lower risk of drug use (OR = .86).

Discussion

Most prior research on social support and reentry has not assessed how support from persons and agencies exert distinct and independent effects on reentry outcomes. This study addressed this knowledge gap by simultaneously assessing support from family, parole officer, social services, and mentors received during reentry. It also extended our understanding of reentry by exploring the time-variant nature of social support. Using multiple waves of data, we compared consistent versus short-term support to assess whether weakening support is associated with an increase in reentry failure. Several major findings emerged from the study.

First, we found multiple pathways that promote reentry. Support from family members, parole officers, and social service agencies exerted protective effects on one or more dimensions of reentry outcomes. As social support theory underscores, support from family, correctional officers, and social services each play a distinct role in the reentry process (Cullen, 1994). The overwhelming majority of previous reentry studies focused on the role of family support and family bonds (Liu, 2020; Mowen et al., 2019; Mowen & Visher, 2015). This study revealed, however, that the salient role of family does not necessarily diminish when other forms of social support are provided.

Second, findings suggest that the stability and longevity of support matter for reentry outcomes. Social support operates in two ways. It explains variation across respondents as well as temporal changes in reoffending and substance misuse for individual persons. This dynamic aspect of social support during reentry underlines the importance of not using static measures of social support for research. The amount of support for reentering citizens from their family fluctuates, as does the relational and material support from parole and social service agencies. This finding echoes previous research that showed returning citizens’ relationship with family members can be both strengthened and strained over time (Sugie & Augustine, 2020), and that continued care and support from parole officers is a critical element for successful reentry (Lovins et al., 2018). Social support’s person-period as well as between-person effects on reentry outcomes suggests respondents who receive better or more varied social support fare better than their peers who do not. In addition, reentry outcomes were drastically better when respondents received a form of support that did not easily weaken over time. Those whose support from family and parole officers decreased over time were nearly twice are likely to reengage in crime and substance misuse.

Not all forms of support are beneficial, however. Findings showed that involvement in community-based mentoring or religious communities exacerbated substance misuse. Previous research reported that when mentors come from a middle-class background and bear little resemblance to their mentee, mentees may feel alienated, pressured, and insufficiently understood (Kenemore & In, 2020; Lopez-Humphreys & Teater, 2019). This can result in increased risk of substance misuse as a self-medicating strategy. Regarding religious activity, it is possible that returning citizens struggling with addiction and criminal stigma experience a substantial social distance between themselves and other church attendees, which triggers feelings of shame, further stigmatization, and acute stress. They may turn to drugs as a coping strategy. Unfortunately, the data did not permit testing this possibility. Future studies should investigate this further using detailed temporal measures of reentering persons’ experiences in religious groups.

We recognize a number of limitations to our study that provide directions for future research. First, our sample contained justice-involved males who reentered society after incarceration. Thus, the findings may not be generalizable to reentering youth and adult females, whose service needs after incarceration might be different from those of their adult male counterparts. Second, the majority of measures used in this study were based on self-reported information. For various reasons, respondents might have hesitated to reveal their familial risk environment and drug use. Future research should consider using multiple sources of data (e.g., official data, clinical diagnosis, and self-reported data) to cross-validate the findings. Third, although our findings suggest that social service support decreases recidivism, it is unclear if all types of social services have equal effects. For example, returning persons often face barriers to employment (Western & Sirois, 2019); it is possible that employment services may have more robust effects than other forms of social services. Future research should examine the effect specific social support services on a variety of reentry outcomes. Finally, it is entirely possible that the effect of social support is contingent on the familial and community environment as well as the relationship between the support provider and the reentering individual. Our data showed that family support is a primary inhibitor of recidivism. However, it also underscored the criminogenic effect of a risky family environment. Family support and risky family environment are not necessarily mutually exclusive; individuals may perceive strong emotional support from a family member who also uses drugs or participates in criminal activity (Mowen & Boman, 2019). Due to data limitations, it was beyond the scope of the study to compare reentry experiences of subgroups of individuals with divergent levels of family support and from diverse family environments. Future studies are needed to explore how risky family environments condition the effect of support from family members.

Given the portrait these results paint, some policy implications can be considered. First, officials should incorporate organized and formal social support for returning citizens into existing reentry policy. Most reentering individuals rely primarily on loved ones for social support, but that support is often fragile (Mowen et al., 2019). Correctional agencies should work with social service agencies to provide more sustainable support and ensure access to housing, transportation, financial assistance, and health care for returning citizens. In-prison programs should also be provided prior to release to help individuals develop the capacity and skills to secure social support from multiple sources. This would be particularly helpful for postincarcerated individuals who do not have a supportive family.

Second, given the little benefit that involvement in religious communities and community-based mentoring programs has in preventing substance misuse, policymakers might consider modifying extant forms of community activities for returning individuals. As recent studies have shown, a formerly incarcerated mentor who already achieved reentry success can help a person leaving prison navigate the transition and can provide more authentic empathy (Matthews, 2021).

Third, this study showed that supportive parole officers can significantly facilitate their clients’ reentry success. A synthetic approach—one that integrates supervision and rehabilitation—is widely observed in current parole practices (DeMichele & Payne, 2018; Hsieh et al., 2015; Miller, 2015). Policymakers should provide training programs to parole officers to promote parole strategies that integrate support into surveillance and rehabilitation. When clients perceive support from officers, they are more willing to obey the requests (Liu et al., 2020; Tyler, 2017). Resources should be provided to allow parole officers to engage with their clients’ families and communities, whose support could be harnessed to prevent clients from being exposed to risky places and activities. Furthermore, policies to address inappropriate parole officer practices should be implemented. Officers who ignore the rehabilitation tasks, use excessively punitive sanctions, and communicate poorly with clients should be held accountable.

In closing, this examination of various forms of social support illustrates multiple pathways toward reentry success: a supportive family, a helpful and empathetic correctional officer, and access to social services. Using a rigorous test that models multiple waves of reentry data, we have demonstrated that an increase in substance misuse and reoffending occurs when social support declines. The assumption that returning citizens can successfully reintegrate solely through willpower and the support of family and friends contradicts the reality of the reentry outcome data. It is time for us to highlight the vital role of organized and formal support from correctional and social service agencies and to construct a social-support-oriented reentry model.

Footnotes

ORCID iD

Lin Liu

Notes

Lin Liu is an assistant professor in the Department of Criminology and Criminal Justice at Florida International University. Her primary research interests include the heterogeneous needs of at-risk youth and adults, offending and co-occurring mental health and substance use disorders, and using machine learning and artificial intelligence to examine the path of social and economic reintegration among individuals transitioning from prison to the community.

Patricia Becker is an assistant professor of criminology at the College of New Jersey. Her current work seeks to explore and understand the various dimensions of crime and victimization within three main areas: (a) the role—and importance—of families in the reentry process of justice-involved individuals, (b) how parenting relates to recidivism and desistance, and (c) how social and structural dynamics and inequality influence the relationship between family and crime.

Thomas J. Mowen is an associate professor in the Department of Sociology at Bowling Green State University. Tom’s current research examines the influence of school security on student outcomes. Tom’s recent studies appear in Justice Quarterly, Journal of Research in Crime and Delinquency, and Crime & Delinquency.

References

Alper

Durose

M. R.

Markman

(2018). 2018 update on prisoner recidivism: A 9-year follow-up period (2005-2014). Bureau of Justice Statistics, U.S. Department of Justice.

Bares

K. J.

Mowen

T. J.

(2020). Examining the parole officer as a mechanism of social support during reentry from prison. Crime & Delinquency, 66(6–7), 1023–1051.

Berg

M. T.

Huebner

B. M.

(2010). Reentry and the ties that bind: An examination of social ties, employment, and recidivism. Justice Quarterly, 28, 382–410.

Berg

M. T.

Loeber

(2011). Examining the neighborhood context of the violent offending-victimization relationship: A prospective investigation. Journal of Quantitative Criminology, 27(4), 427–451.

Berghuis

. (2018). Reentry programs for adult male offender recidivism and reintegration: A systematic review and meta-analysis. International Journal of Offender Therapy and Comparative Criminology, 62(14), 4655–4676.

Blasko

B. L.

Friedmann

P. D.

Rhodes

A. G.

Taxman

F. S.

(2015). The parolee–parole officer relationship as a mediator of criminal justice outcomes. Criminal Justice and Behavior, 42(7), 722–740.

Boman

Mowen

T. J.

(2018). The role of turning points in establishing baseline differences between people in developmental and life course criminology. Criminology, 56(1), 191–224.

Bronson

Carson

E. A.

(2019). Prisoners in 2017. Bureau of Justice Statistics, U.S. Department of Justice.

Buck

(2020). Peer mentoring in criminal justice. Routledge.

10.

Bullis

Yovanoff

(2002). Those who do not return: Correlates of the work and school engagement of formerly incarcerated youth who remain in the community. Journal of Emotional and Behavioral Disorders, 10(2), 66–78.

11.

Bushway

Uggen

(2021). Fostering desistance. Contexts, 20(4), 34–39.

12.

Camp

S. D.

Daggett

D. M.

Kwon

Klein-Saffran

(2008). The effect of faith program participation on prison misconduct: The life connections program. Journal of Criminal Justice, 36(5), 389–395.

13.

Carson

E. A.

(2014). Prisoners in 2013. Bureau of Justice Statistics, U.S. Department of Justice.

14.

Carson

E. A.

Golinelli

(2012). Prisoners in 2012: Trends in admissions and releases, 1991–2012. Bureau of Justice Statistics, U.S. Department of Justice.

15.

Cartier

Farabee

Prendergast

M. L.

(2006). Methamphetamine use, self-reported violent crime, and recidivism among offenders in California who abuse substances. Journal of Interpersonal Violence, 21(4), 435–445.

16.

Chamberlain

A. W.

Gricius

Wallace

D. M.

Borjas

Ware

V. M.

(2018). Parolee–parole officer rapport: Does it impact recidivism? International Journal of Offender Therapy and Comparative Criminology, 62(11), 3581–3602.

17.

Clear

T. R.

Hardyman

P. L.

Stout

Lucken

Dammer

H. R.

(2000). The value of religion in prison: An inmate perspective. Journal of Contemporary Criminal Justice, 16(1), 53–74.

18.

Cloud

D. H.

Bassett

M. T.

Graves

Fullilove

R. E.

Brinkley-Rubinstein

(2020). Documenting and addressing the health impacts of carceral systems. American Journal of Public Health, 110(S1), S5–S5.

19.

Coleman

J. S.

(1990). Foundations of social theory. Harvard University Press.

20.

Cooke

B. J.

Farrington

D. P.

(2016). The effectiveness of dog-training programs in prison: A systematic review and meta-analysis of the literature. The Prison Journal, 96(6), 854–876.

21.

Cornwell

(2003). The dynamic properties of social support: Decay, growth, and staticity, and their effects on adolescent depression. Social Forces, 81(3), 953–978.

22.

Cullen

F. T.

(1994). Social support as an organizing concept for criminology: Presidential address to the Academy of Criminal Justice Sciences. Justice Quarterly, 11, 527–559.

23.

Cusick

G. R.

Goerge

R. M.

Bell

K. C.

(2009). From corrections to community: The juvenile reentry experience as characterized by multiple systems involvement. Chapin Hall at the University of Chicago.

24.

DeMichele

Payne

(2018). Taking officer time seriously: A study of the daily activities of probation officers. Probation Journal, 65(1), 39–60.

25.

Durose

M. R.

Antenangeli

(2021). Recidivism of prisoners released in 34 states in 2012: A 5-year follow-up period (2012–2017). Bureau of Justice Statistics.

26.

Durose

M. R.

Cooper

A. D.

Snyder

H. N.

(2014). Recidivism of prisoners released in 30 states in 2005: Patterns from 2005 to 2010 (NCJ Publication No. 244205). Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice.

27.

Farabee

Zhang

S. X.

Wright

(2014). An experimental evaluation of a nationally recognized employment-focused offender reentry program. Journal of Experimental Criminology, 10(3), 309–322.

28.

Fries

Fedock

Kubiak

S. P.

(2014). Role of gender, substance use, and serious mental illness in anticipated post jail homelessness. Social Work Research, 38(2), 107–116.

29.

Garcia

(2016). The importance of the mentor–mentee relationship in women’s desistance from destructive behaviors. International Journal of Offender Therapy and Comparative Criminology, 60(7), 808–827.

30.

Gariepy

Honkaniemi

Quesnel-Vallee

(2016). Social support and protection from depression: Systematic review of current findings in western countries. The British Journal of Psychiatry, 209(4), 284–293.

31.

Harding

D. J.

Morenoff

J. D.

Wyse

J. J.

(2019). On the outside. University of Chicago Press.

32.

Hsieh

M. L.

Hafoka

Woo

van Wormer

Stohr

M. K.

Hemmens

(2015). Probation officer roles: A statutory analysis. Federal Probation, 79, 20–37.

33.

Hucklesby

Wincup

(2014). Assistance, support and monitoring? The paradoxes of mentoring adults in the criminal justice system. Journal of Social Policy, 43(2), 373–390.

34.

Huebner

Shannon

(2022). Private probation costs, compliance, and the proportionality of punishment: Evidence from Georgia and Missouri. RSF: The Russell Sage Foundation Journal of the Social Sciences, 8, 179–199. https://doi.org/10.7758/RSF.2022.8.1.08

35.

Jacobson

D. E.

(1986). Types and timing of social support. Journal of Health and Social Behavior, 27, 250–264.

36.

Janssen

H. J.

Oberwittler

Koeber

(2021). Victimization and its consequences for well-being: A between-and within-person analysis. Journal of Quantitative Criminology, 37(1), 101–140.

37.

Johnson

B. R.

Larson

D. B.

Pitts

T. C.

(1997). Religious programs, institutional adjustment, and recidivism among former inmates in prison fellowship programs. Justice Quarterly, 14(1), 145–166.

38.

Johnson Listwan

Colvin

Hanley

Flannery

. (2010). Victimization, social support, and psychological well-being: A study of recently released prisoners. Criminal Justice and Behavior, 37(10), 1140–1159.

39.

Kail

B. L.

Carr

D. C.

(2020). Structural social support and changes in depression during the retirement transition: “I get by with a little help from my friends.” The Journals of Gerontology: Series B, 75(9), 2040–2049.

40.

Kenemore

T. K.

B. S.

(2020). Relationship matters with citizens returning from prison: Qualitative findings from a mentoring program. Journal of Offender Rehabilitation, 59(6), 315–333.

41.

LaCourse

Listwan

S. J.

Reid

Hartman

J. L.

(2019). Recidivism and reentry: The role of individual coping styles. Crime & Delinquency, 65(1), 46–68.

42.

Lattimore

P. K.

Steffey

D. M

. (2009). The multi-site evaluation of SVORI: Methodology and analytic approach. RTI International.

43.

Lattimore

P. K.

Steffey

D. M.

(2010). The multi-site evaluation of SVORI: Methodology and analytic approach. RTI International.

44.

Liu

(2020). Family, parochial, and public levels of social control and recidivism: An extension of the systemic model of social disorganization. Crime & Delinquency, 66(6–7), 864–886.

45.

Liu

Miller

S. L.

Visher

C. A.

(2021). The strain of procedural injustice in parole among former prisoners: A test with a mixed-gender sample. Justice Quarterly, 38(4), 653–677.

46.

Liu

Mowen

T. J.

Visher

C. A.

Sun

(2022). Violent victimization during reentry: Prevalence, triggers, and impact on mental health. Justice Quarterly. Advance online publication. https://doi.org/10.1080/07418825.2022.2104747

47.

Liu

Visher

C. A.

(2019). The crossover of negative emotions between former prisoners and their family members during reunion: A test of general strain theory. Journal of Offender Rehabilitation, 58(7), 567–591.

48.

Liu

Visher

C. A.

(2021). Decomposition of the role of family in reentry: Family support, tension, gender, and reentry outcomes. Crime & Delinquency, 67(6–7), 970–996.

49.

Liu

Visher

C. A.

O’Connell

D. J.

(2020). The strain from procedural injustice on parolees: Bridging procedural justice theory and general strain theory. Crime & Delinquency, 66(2), 250–276.

50.

Longmate

Lowder

Givens

Van Deinse

T. B.

Ghezzi

Burgin

Cuddeback

G. S.

(2021). Social support among people with mental illnesses on probation. Psychiatric Rehabilitation Journal, 44(1), 70–76.

51.

Lopez-Humphreys

Teater

(2018). Peer mentoring justice-involved youth: A training model to promote secondary desistance and restorative justice among mentors. The International Journal of Restorative Justice, 1, 187–207.

52.

Lopez-Humphreys

Teater

. (2019). Transformations of the self: Learning from the experiences of returned citizens participating in peer mentor support training. Urban Social Work, 3(2), 136–155.

53.

Loughran

T. A.

Mulvey

E. P.

Schubert

C. A.

Fagan

Piquero

A. R.

Losoya

S. H.

(2009). Estimating a dose-response relationship between length of stay and future recidivism in serious juvenile offenders. Criminology, 47(3), 699–740.

54.

Lovins

B. K.

Cullen

F. T.

Latessa

E. J.

Jonson

C. L.

(2018). Probation officer as a coach: Building a new professional identity. Federal Probation, 82, 13–31.

55.

Marlow

Grajeda

Lee

Young

Williams

Hill

. (2015). Peer mentoring for male parolees: A CBPR pilot study. Progress in Community Health Partnerships: Research, Education, and Action, 9(1), 91–100. https://doi.org/10.1353/cpr.2015.0013

56.

Martínez

R. S.

Aricak

O. T.

Graves

M. N.

Peters-Myszak

Nellis

(2011). Changes in perceived social support and socioemotional adjustment across the elementary to junior high school transition. Journal of Youth and Adolescence, 40(5), 519–530.

57.

Maruna

(2015). Foreword: Narrative criminology as the new mainstream. In Presser

Sandberg

(Eds.), Narrative criminology: Understanding stories of crime (pp. vii–x). New York University Press.

58.

Matsuda

Thornberry

T. P.

Loughran

T. A.

Krohn

M. D.

(2022). Are late bloomers real? Identification and comparison of late-onset offending patterns from ages 14–40. Journal of Developmental and Life-Course Criminology, 8(1), 124–150.

59.

Matthews

(2021). Peer-focused prison reentry programs: Which peer characteristics matter most? Incarceration, 2(2), 1–19.

60.

McKay

Comfort

Lindquist

Bir

(2016). If family matters: Supporting family relationships during incarceration and reentry. Criminology & Public Policy, 15, 529–611.

61.

Miller

(2015). Contemporary modes of probation officer supervision: The triumph of the “synthetic” officer? Justice Quarterly, 32(2), 314–336.

62.

Minton

T. D.

Zeng

Maruschak

L. M.

(2021). Impact of COVID-19 on the local jail population, January-June 2020. Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice.

63.

Moschion

Johnson

(2019). Homelessness and incarceration: A reciprocal relationship? Journal of Quantitative Criminology, 35(4), 855–887.

64.

Mowen

T. J.

Boman

J. H.

IV . (2018). The duality of the peer effect: The interplay between peer support and peer criminality on offending and substance use during reentry. Crime & Delinquency, 64(8), 1094–1116.

65.

Mowen

T. J.

Boman

J. H.

IV . (2019). The criminogenic influence of family on substance use during reentry: A life course perspective on between individual differences and within individual changes. Justice Quarterly, 36, 841–869.

66.

Mowen

T. J.

Stansfield

Boman

J. H.

IV . (2019). Family matters: Moving beyond “if” family support matters to “why” family support matters during reentry from prison. Journal of Research in Crime and Delinquency, 56(4), 483–523.

67.

Mowen

T. J.

Visher

C. A.

(2015). Drug use and crime after incarceration: The role of family support and family conflict. Justice Quarterly, 32(2), 337–359.

68.

Mowen

T. J.

Wodahl

Brent

J. J.

Garland

. (2018). The role of sanctions and incentives in promoting successful reentry: Evidence from the SVORI data. Criminal Justice and Behavior, 45(8), 1288–1307.

69.

Nixon

(2020). ‘Giving back and getting on with my life’: Peer mentoring, desistance and recovery of ex-offenders. Probation Journal, 67(1), 47–64. https://doi.org/10.1177/0264550519900249

70.

O’Connell

Visher

Liu

(2019). Violent offending, desistance, and recidivism. Marquette Law Review, 103, 983–1002.

71.

O’Connor

T. P.

(2005). What works, religion as a correctional intervention: Part II. Journal of Community Corrections, 14(2), 4–26.

72.

Paternoster

Bachman

Kerrison

O’connell

Smith

(2016). Desistance from crime and identity: An empirical test with survival time. Criminal Justice and Behavior, 43(9), 1204–1224.

73.

Pratt

T. C.

Cullen

F. T.

(2005). Assessing macro-level predictors and theories of crime: A meta-analysis. Crime and Justice, 32, 373–450.

74.

Raudenbush

S. W.

Bryk

A. S.

(2002). Hierarchical linear models: Applications and data analysis methods (Vol. 1). SAGE.

75.

Robinson

J. A.

(2005). The relationship between parole and recidivism in the criminal justice system. Mcnair Scholars Journal, 9(1), 12–32.

76.

Sampson

R. J.

Laub

J. H.

(1993). Crime in the making: Pathways and turning points through life. Harvard University Press.

77.

Sarason

I. G.

Sarason

B. R.

Shearin

E. N.

Pierce

G. R.

(1987). A brief measure of social support: Practical and theoretical implications. Journal of Social and Personality Relationships, 4, 497–510.

78.

Shollenberger

T. L.

(2009). When relatives return: Interviews with family members of returning prisoners in Houston, Texas. Urban Institute Justice Policy Center.

79.

Skeem

Eno Louden

Manchak

Vidal

Haddad

(2009). Social networks and social control of probationers with co-occurring mental and substance abuse problems. Law and Human Behavior, 33(2), 122–135.

80.

Stansfield

Mowen

T. J.

O’Connor

(2018). Religious and spiritual support, reentry, and risk. Justice Quarterly, 35(2), 254–279.

81.

Sugie

N. F.

Augustine

(2020). Social support in daily life at reentry. In Leverentz

A. M.

Chen

E. Y.

Christian

(Eds.), Beyond recidivism: New approaches to research on prisoner reentry and reintegration (pp. 197–218). New York University Press.

82.

Tsai

Harpaz-Rotem

Pietrzak

R. H.

Southwick

S. M.

(2012). The role of coping, resilience, and social support in mediating the relation between PTSD and social functioning in veterans returning from Iraq and Afghanistan. Psychiatry: Interpersonal & Biological Processes, 75(2), 135–149.

83.

Tyler

(2017). Procedural justice and policing: A rush to judgment? Annual Review of Law and Social Science, 13, 29–53.

84.

Vaux

(1988). Social support: Theory, research, and intervention. Praeger.

85.

Victor

Sightes

Watson

D. P.

Ray

Bailey

Robision

Fears

Edwards

Salyers

(2021). Designing and implementing an intervention for returning citizens living with substance use disorder: Discovering the benefits of peer recovery coach involvement in pilot clinical trial decision-making. Journal of Offender Rehabilitation, 60(2), 138–158.

86.

Vidal

Oudekerk

B. A.

Reppucci

N. D.

Woolard

(2015). Examining the link between perceptions of relationship quality with parole officers and recidivism among female youth parolees. Youth Violence and Juvenile Justice, 13(1), 60–76.

87.

Wallace

Fahmy

Cotton

Jimmons

McKay

Stoffer

Syed

(2016). Examining the role of familial support during prison and after release on post-incarceration mental health. International Journal of Offender Therapy and Comparative Criminology, 60(1), 3–20.

88.

Western

(2018). Homeward: Life in the year after prison. Russell Sage Foundation.

89.

Western

Braga

A. A.

Davis

Sirois

(2015). Stress and hardship after prison. American Journal of Sociology, 120(5), 1512–1547.

90.

Western

Sirois

(2019). Racialized re-entry: Labor market inequality after incarceration. Social Forces, 97(4), 1517–1542.

91.

Wodahl

E. J.

Mowen

T. J.

Garland

B. E.

(2021). The effect of individual characteristics and supervision experiences on the perceived quality of the supervision relationship. Criminal Justice Policy Review, 32(5), 523–545.

92.

Wolff

Draine

(2004). Dynamics of social capital of prisoners and community reentry: Ties that bind? Journal of Correctional Health Care, 10(3), 457–490.

93.

Wright

J. P.

Cullen

F. T.

(2006). Employment, peers, and life-course transitions. Justice Quarterly, 21(1), 183–205.

94.

Wyse

J. J.

Harding

D. J.

Morenoff

J. D.

(2014). Romantic relationships and criminal desistance: Pathways and processes. Sociological Forum, 29(2), 365–385.

95.

Yahner

Visher

Solomon

(2008). Returning home on parole: Former prisoners’ experiences in Illinois, Ohio, and Texas. The Urban Institute.