The Nexus of Violent Victimization,Mental Health,and Employment: Findings From a Sample of Post-Incarcerated Individuals

Abstract

The goals of this study were to assess the prevalence of victimization among people who are transitioning from prison to the community, and to examine the nexus of violent victimization during reentry, mental health, and weekly work hours. The participants (n = 724; average age = 29.09 years) were interviewed before release, and in the 3rd, 9th, and 15th months into reentry. Longitudinal data about their mental health, work hours, family environment, and victimization were collected. We drew on Agnew’s general strain theory and employed multilevel longitudinal modeling to examine how victimization affected respondents’ work hours via mental health. Findings revealed that greater exposure to violent victimization deteriorated mental health among respondents. Furthermore, an indirect effect between victimization and reduced work capacity operating through poor mental health was observed at the between-person level. These results underscored the alarmingly high prevalence of victimization among reentering individuals and that maintaining stable employment, a critical step of reintegration after imprisonment, is difficult for reentering individuals when they become a victim of violence and suffer mental health deterioration. Implications for addressing victimization among people transitioning out of prison are discussed.

Keywords

reentry violent victimization mental health work capacity general strain theory

Introduction

In the current era of decarceration, more than 650,000 people reenter society each year (Alper et al., 2018; Durose et al., 2014). Unfortunately, a staggering 68% of returning citizens go back to prison within 3 years after release (Durose & Antenangeli, 2021). One of the most intimidating challenges during reentry and exiting from crime is maintaining employment (Lukies et al., 2011; Sampson & Laub, 2003). Without stable employment, it is hard to achieve financial independence (Rydberg, 2018; Sampson & Laub, 2003) and establish structured daily routines (Gwynne et al., 2020; Osgood & Anderson, 2004), which are proven catalysts to desistance (Uggen & Kruttschnitt, 1998; Williams et al., 2019). Several factors have been proposed to explain why returning citizens encounter difficulty securing stable employment, ranging from employers’ reluctance to hire job applicants with criminal records (Holzer et al., 2002; Western & Sirois, 2019) to returning individuals’ low levels of education, limited vocational skills, and poor work histories (Fahey et al., 2006; Pager & Western, 2009). However, there is another important barrier to successful employment that has been insufficiently examined in the context of reentry: violent victimization.

There is a long-standing consensus among researchers that violent victimization can lead to poor performance in work, family, and community roles (Moon et al., 2019; Okumu et al., 2020; Schwartz et al., 2005). The effects of victimization have been studied within different populations that are especially vulnerable to violence, including juveniles (Juvonen et al., 2000; Raskauskas et al., 2015), women (Banyard et al., 2011; Renner, 2009), and veterans (Iverson et al., 2017). However, even though victimization is deemed to be a serious issue, limited research has been devoted to investigating its effects among the particularly vulnerable social group of individuals who have recently been released from prison. Even less is known how this detrimental experience impacts their reentry and reintegration.

A commanding majority of released prisoners return to impoverished neighborhoods with high crime rates (Liu & Visher, 2019; Visher, 2004), and they are likely to be at a particularly high risk of violent victimization during this critical stage of reintegration. Reuniting with family members after release may trigger tension, distrust, and conflict, especially when some family members have endured traumatic experiences caused by returning citizens’ past crimes (Gideon, 2007; Laird, 2013). Those who become dependent on family members after release may become a victim of abuse and experience strained relationships (Arditti & Few, 2008; Mallik-Kane & Visher, 2008; Western, 2018). These experiences are likely to take a toll on returning citizens’ ability to succeed at work and thus thwart their efforts to reintegrate. However, without a proper understanding of the mechanisms underlying the association between victimization during reentry and employment outcomes, policymakers, researchers, and practitioners are unlikely to develop efficacious intervention initiatives to protect returning citizens from violent victimization and facilitate a successful transition from prison to the community.

Reintegration is a process featuring dynamic temporal changes in people’s relationships, mental health, and employment (Bushway, 2004; Liu et al., 2022; Myers & Olson, 2013; Travis, 2002; Western, 2018). The transition from prison to the community is a dynamic rather than static stage. During reentry, individuals tap into various forms of resources and encounter new challenges as reintegration progresses (Berg & Huebner, 2011; Harding et al., 2019). Weekly work hours of post-incarcerated persons are found to vary as they progress further into reentry (Harding et al., 2019; Western, 2018; Western & Sirois, 2019). There is also heterogeneity in mental health and employment situations across different returning citizens (Visher et al., 2008; Visher et al., 2011). Given these realities, the person-period as well as between-person variation in victimization, mental health, and work situations should be assessed.

To fill this gap in knowledge, this study embarks on exploring the occurrence of violent victimization and its impact on employment using a 15-month follow-up study of returning citizens who were transitioning from prison to the community. Specifically, the study investigates the mental health mechanisms through which victimization may undermine post-incarcerated people’s employment outcomes. Prior studies have identified mental health as a mechanism through which victimization undermines the ability to function productively (Fu et al., 2021; Price et al., 2019). Building on this line of research, we assess whether the effect of violent victimization on employment operates indirectly by undermining mental health. To address the nested nature of repeated measures of participants in the longitudinal data, we use longitudinal multilevel modeling to assess the temporal changes of predictors and outcomes while controlling for population heterogeneity.

Literature Review

The Link Between Victimization and Employment from a GST Perspective

Agnew’s (1992) general strain theory (GST) provides an applicable framework for understanding how violent victimization and subsequent mental health issues might affect employment. GST argues that a variety of strains and stressors can trigger deviant behavior. While strain theory did not originate with Agnew, he emphasized its social-psychological component (Froggio, 2007). He argued that there are three main types of strain: (1) those that prevent or threaten individuals from achieving their positively valued goals, (2) the removal of positively valued stimuli that one possesses, and (3) the presentation of noxious or negatively valued stimuli (Agnew, 1992; Froggio, 2007).

Each type of strain is theorized to increase the likelihood that an individual will experience negative emotions such as anger, frustration, depression, and anxiety (Agnew, 1992; Froggio, 2007). These emotions can accumulate, leading to adverse psychological states (Agnew, 1992). In turn, negative emotional states can contribute to delinquent and criminal behavior by providing a pressure to alleviate negative emotions through maladaptive coping strategies. Summarily, Agnew (1992, 2001) argues that the effect of strain on delinquency and crime should be indirect and operate through negative emotions. In his original iteration of GST, Agnew argues that any form of strain can negatively impact emotional states and eventually lead to criminal behavior (Agnew, 1992; Agnew, 1999; Agnew et al., 2002). While most studies that utilize Agnew’s theory apply it to criminality and antisocial behavior outcomes (Baron, 2006; Hay et al., 2010; Ostrowsky & Messner, 2005), others have utilized GST by applying it to, among other things, suicidal ideation and self-protective behavior (Archer, 2019; Hay & Meldrum, 2010). Herein, we contribute to this small body of research employing GST to examine outcomes other than delinquency and crime by focusing on employment and the manner in which strain and negative emotions can hamper the economic reintegration of post-incarcerated individuals.

Maintaining Employment to Achieve Economic Reintegration

According to Sampson and Laub (Sampson & Laub, 1995; Sampson & Laub, 2003), desistance is likely to occur when individuals experience turning points in life such as getting married and obtaining stable employment. The turning point of employment is a process that does not arise intact and full-grown but develops over time like a pension plan funded by regular installments (Laub et al., 1998). As “the investment” in employment grows, the incentive for avoiding crime increases and solidifies because the various forms of achievement and benefits coming along with stable employment are incompatible with crime. Thus, individuals who can maintain stable and gainful employment after prison release are more likely to desist and sever a criminal past.

However, there are various barriers faced by those returning to society from prison to maintain employment. Among these include mental health issues (Hamilton, 2016), substance use (Aresti et al., 2010), and criminal records (Solomon, 2012). Beyond these empirically established barriers, though, lies the possibility that when returning citizens experience violent victimization, mental health struggles may manifest and could jeopardize their ability to form the important adult social bond of stable and gainful employment.

Research on Violent Victimization, Mental Health, and Employment

Several studies show that violent victimization impairs one’s physical (Hager & Leadbeater, 2016; Herge et al., 2016) and mental (Biebl et al., 2011; Janssen et al., 2021) health. For the most frequently studied group (i.e., juveniles), there have been consistent findings. Depression is one of the most examined outcomes (Okumu et al., 2020; Raskauskas et al., 2015; Turner et al., 2010). Turner et al (2010), for example, used the Developmental Victimization Survey to examine various types of victimization and how they can lead to depressive symptoms in juveniles. Results showed that victimization had substantial direct effects on depressive symptoms. Another significant finding was that depressive symptoms were linked to each of the four victimization types measured (i.e., maltreatment, sexual, peer, adversity). In a more recent study, Okumu et al. (2020) analyzed similar measures with an additional outcome of school performance. They too found that victimization was positively associated with depression.

While much of the literature regarding juveniles is cross-sectional, some studies have been longitudinal and have linked adolescent victimization with adult outcomes. Turanovic (2019) examined how violent victimization in adolescence can affect the life course, finding that childhood victimization produced psychological problems later in life (Turanovic, 2019). Focusing on intimate partner violence (IPV), Iverson et al. (2017) found that both physical and sexual IPV were associated with severe mental health issues. Regarding the link between victimization and work capacity, Tolman and Wang (2005) found that the occurrence of domestic violence can impair one’s inability to work. Directly, abusive partners may interfere with any attempt to gain or maintain employment (Tolman & Wang, 2005). Indirectly, violence may lead to adverse mental health outcomes which can, in turn, affect employment.

Another study that explored the various outcomes of violent victimization was conducted by Banyard et al. (2011). They examined the impact of multiple types of victimization (sexual violence, physical IPV, psychological abuse, and stalking) on a range of work outcomes (job satisfaction, job benefits, job interference). Using data collected as part of a telephone survey of 1,079 women living in New Hampshire, the results indicated clear links between victimization experiences and negative work outcomes. Separately, Janssen et al. (2021) conducted a longitudinal study with a general population sample to examine the link between victimization and mental health. By employing a two-wave panel survey of almost 3,000 participants in Germany, they found that individuals who experienced violent victimization had lower levels of well-being compared to individuals who were not victimized (Janssen et al., 2021).

As strain theory suggests, violent victimization can contribute to poor emotional and mental health, and subsequent consequences can also follow. For example, Merikangas et al. (2007) found that mental health disorders, including those related to anxiety and depression, are positively associated with greater role disability, such as the inability to work. Likewise, Lerner and Henke (2008) found that depressive symptoms are strongly correlated with poorer job performance, operationalized as work hours. Summarily, while no single study has focused on longitudinal associations linking violent victimization to poor mental health and employment outcomes among individuals who recently were released from prison, prior theory and research focused on other vulnerable populations strongly suggest such associations exist.

Current Study

Much of the previous literature on violent victimization and its cascading effects focuses on vulnerable populations (e.g., juveniles and veterans). Yet, one noticeably absent vulnerable population from this area of research is the formerly incarcerated. Furthermore, although it is argued that mental health is a mechanism through which victimization undermines people’s ability to function productively (Janssen et al., 2021; McDougall & Vaillancourt, 2015; Turanovic, 2019), the majority of prior studies centered around how victimization leads to crime and delinquency. Returning citizens are often stereotyped as violent perpetrators (Owens, 2009; Sheppard & Ricciardelli, 2020), but we know little about their risk of violent victimization during the transition from prison to the community and how victimization experiences impair their ability to work, which is critical to their economic reintegration. In the current study, we use longitudinal data to examine the associations between violent victimization, mental health, and work capacity of returning citizens. The longitudinal nature of the data allows us to address the time-dynamic rather than static nature of reentry experiences. Our first hypothesis is that greater exposure to violent victimization will result in poorer mental health outcomes among returning citizens. In addition, the present study also assesses the extent to which mental health functions as a mediator between victimization and work capacity. In this regard, our second hypothesis is that the mental health mediates the effect of violent victimization on work hours.

Methods

Data and Sample

The current study used data from the evaluation of the Serious and Violent Offender Reentry Initiative (SVORI) program. SVORI was a federally funded program that assisted states in developing programs and policies to improve the transition from prison to the community for returning individuals (Lattimore & Steffey, 2009). Participants were selected from 12 states receiving SVORI funding between July 2004 and November 2005. While this data set has limitations regarding gender diversity—all respondents were justice-involved individuals in male prisons who participated in the study,¹ there was good racial diversity. Furthermore, respondents were from 12 states (Indiana, Iowa, Kansas, Maine, Maryland, Missouri, Nevada, Ohio, Oklahoma, Pennsylvania, South Carolina, and Washington), strengthening the geographic diversity of the sample. We are mindful of the limited generalizability of the findings to females and youth, a point to which we return in the discussion section.

The SVORI evaluation included participants who received in-prison SVORI programming and others who were selected as comparison subjects. Three post-release in-person interviews were conducted after 3, 9, and 15 months into reentry (hereafter T1, T2, and T3 data, respectively). These interviews collected information about participants’ current reentry experiences with a focus on employment, mental health, housing, and other aspects of reintegration. For the three post-release interviews, the same set of questions were administered, which allowed for an examination of the longitudinal change in various aspects of the reentry experience (Lattimore & Visher, 2009). Given that a major focus of this study is how victimization affects the weekly work hours of participants, the sample for the present study was comprised of SVORI men who were employed during their post-release interviews (n = 790).

Missing Data

Among these 790 employed SVORI men, 66 of them (8%) only participated in one of all three post-release interviews, making it impossible to examine how the temporal change of their mental health struggles affects their time-variant work capacity. Following the past practices of studies using panel data in which respondents did not participate in all waves of data collection (Allison, 2019; Thompson & Pickett, 2021), we included employed SVORI respondents who participated in at least two of the three post-release interviews. Sensitivity tests were conducted to examine whether the 8% attrition is at random. Results show no noticeable differences in age, race, SVORI participation, and education level between individuals included and not included in the study (complete results are available upon request).

Measures

Dependent variable

T1, T2, and T3 works hours are represented by the number of work hours respondents reported to have each week. They are numeric variables created based on a question asking “On average, how many hours a week do you usually work for your current job?” at T1, T2, and T3 interviews.

Mediating variable

To measure poor mental health, the hypothesized mediator, at each of the three assessment periods, we employ multi-item indicators of depressive symptoms, hostility, and anxiety. As we describe in greater detail in a later section, these three indicators are used to construct a higher-order latent measure of poor mental health based on the weighted scores. This approach is supported by prior work establishing strong associations between these three indicators (Suls, 2018; Temple et al., 2016), suggesting that a global measure of poor mental health is preferred over the estimation of separate mental health indicators.²

T1, T2, and T3 depressive symptoms are assessed based on the Center for Epidemiologic Studies Depression Index (Radloff, 1977). Respondents were asked how often in the past 7 days they were bothered by (1) an inability to shake off the blues even with help from family and friends, (2) feeling no interest in things, (3) feeling hopeless about the future, (4) feeling your mind going blank, and (5) feeling inferior to others. The responses for each item range from (1) “not at all” to (5) “extremely” (α = 0.76–0.79 across all waves). T1, T2, and T3 feelings of hostility are measured by the Brief Symptom Inventory (BSI; Derogatis, 1975), which assessed how often over the past 7 days participants (1) had temper outbursts that they could not control, (2) had an urge to beat, injure, or harm someone, (3) had an urge to break or smash things, and (4) got into frequent arguments. Reponses range from (1) “not at all” to (5) “extremely” (α = 0.77–0.80 across all waves). T1, T2, and T3 anxiety are also assessed by the BSI. Five questions asked the frequency respondents experienced such symptoms over the past 7 days: (1) feeling suddenly scared, (2) feeling fearful, (3) feeling tense or keyed up, (4) having spells of terror or panic, and (5) feeling so restless they could not sit still. Reponses range from (1) “not at all” to (5) “extremely” (α = 0.76–0.78 across all waves).

Independent variables

T1, T2, and T3 violent victimization, the focal predictor variable of interest, was a latent variable based on weighted scores. Five items were drawn from Conflict Tactics Scales (Straus, 1990). These questions asked participants since release/the prior interview, whether they had any of the following experiences: (1) having someone threaten to hit them, (2) having something thrown at them, (3) being pushed/grabbed/shoved, (4) being slapped, kicked, bit, or hit, and (5) being threatened with a weapon by someone. Responses for each of the items ranged from 0 (never) to 7 (several times a week). The items exhibit good internal reliability (α = .85–.87 across waves).

Control variables

Because familial and community environments are salient predictors of reentry outcomes (Mowen & Visher, 2015; Visher, 2007), especially in predicting reentering individuals’ mental health and job opportunities (Bakken & Visher, 2018; Western & Sirois, 2019), we test the mediation model with family and neighborhood environmental proxies adjusted. During the first post-release interview, respondents reported their family members’ involvement in drug use and crime, as well as criticism toward them. We used five variables to capture risky familial environments. The first captures excessive criticism of participants by family members and is based on a question asking to what extent participants agree that “my family criticizes me a lot.” Responses range from (1) strongly disagree, to (4) strongly agree. The other four family-related variables are binary items (Yes = 1, No = 0) indicating the presence of the following situations in respondents’ families: (1) family members have ever been incarcerated, (2) family members currently use drugs, (3) family members currently engage in illicit activities, and (4) family members use alcohol in respondents’ presence.

To capture risky neighborhood environments, the analysis includes a time-variant measure of T1, T2, and T3 concerns over neighborhood disorder, which is measured as a latent factor at each wave. Drawing on insights from Skogan’s (1986) discussion of neighborhood disorder, these latent variables are constructed based on three items asking to what extent participants agree with the statements: (1) it is hard to stay out of trouble in this neighborhood, (2) drug selling is a major problem in this neighborhood, and (3) this neighborhood makes it hard to stay out of prison. Responses range from (1) strongly disagree to (4) strongly agree. The items exhibit good internal reliability (α = .76–.79 across waves). Another time-variant variable pertaining to the neighborhood environment is T1, T2, and T3 perceived job opportunities in neighborhood, measured based on an item asking participants to what extent they agree with the statement: “your neighborhood is a good place to find a job.” Responses range from (1) strongly disagree to (4) strongly agree.

Several additional control variables are included in the analysis. Specifically, we control for the effect of correctional programs in the model. SVORI participant is a binary variable indicating if the participant was in the SVORI program group or the comparison group. Lastly, we control for participants’ age (in years), race (White [reference group], Black, Other race), and education level (in years).

Analytic Strategies

Second-order latent measure of poor mental health

To measure the global indicator for poor mental health, we followed operations of past studies on mental health well-being and employed Partial Least Squares Path Modeling (PLSPM) to construct a second-order latent variable poor mental health based on the three first-order latent variables reflecting depression, hostility, and anxiety (Ciavolino, 2012). To accomplish this, the three lower-order latent variables of depression, hostility, and anxiety were linked to their respective raw items, and the higher-order latent variable has paths toward each of the three lower-order latent variables. Rather than using the unweighted scores, weighted scores that reflected the path coefficients were used to construct both the first- and second-order latent variables. We tested the construct T1, T2, and T3 poor mental health. For all three waves of data, PLSPM results indicated that the three lower-order latent variables loaded highly onto the high-order latent factor of poor mental health. For example, the goodness-of-fit indices for this model using T1 data were satisfactory (χ² = 361.75, p < .001, comparative fit index = 0.94, root mean square error of approximation = .06, root mean square residual = .04). The model fit indices of T2 and T3 models were similarly acceptable to those of T1 model.

Assessing both person-period and between-person effects

This study employed longitudinal multilevel modeling, an analytical approach widely used in the social sciences to model panel data in which both person-period and between-person variation are measured by several waves of data collection (Berg & Loeber, 2011; Bersani et al., 2009; Janssen et al., 2021; Raudenbush & Bryk, 2002; Ward & Forney, 2020; Widdowson et al., 2021). In this modeling framework, each respondent is a cluster (a level-two unit) and repeated measures of this person are level-one units. In the current study, as a level 2 unit (a cluster), each respondent has his own intercept. Furthermore, the model answers how a respondent’s reentry experience differs from others by estimating whether someone who experiences higher-than-average levels of a risk factor also experiences a higher-than-average severity of an adverse outcome. The model also disentangles the temporal dynamic by estimating whether a temporal change in a risk factor is associated with a temporal change in the outcomes. Due to space considerations, the basic model in this study is expressed as follows, using the predictor of victimization and outcome of mental health as an example (where i indicates respondent and ti indicates respondent periods):

\begin{array}{l} y_{t i} = γ_{0 i} + γ_{1 i} (p e r s o n - s p e c i f i c - m e a n - c e n t e r e d v i c t i m i z a t i o n_{t i}) \\ + π_{01} (g r a n d - m e a n - c e n t e r e d v i c t i m i z a t i o n_{t i}) + β_{01} r a c e_{i} \\ \begin{array}{l} + β_{02} a g e_{i} + β_{03} e d u c a t i o n_{i} + β_{04} f a m i l y m e m b e r s' d r u g u s e_{i} \\ + β_{05} f a m i l y m e m b e r s' e n g a g e m e n t i n o f f e n d i n g_{i} \end{array} \\ \begin{array}{l} + β_{06} f a m i l y m e m b e r s' e x c e s s i v e c r i t i c i s m o n r e s p o n d e n t s_{i} \\ + β_{07} S V O R I p a r t i c i p a n t_{i} + ε_{t i} \end{array} \end{array}

where y_ti represents the severity of poor mental health of individual i at time t. γ_1i is the level-1 coefficient of the time-variant variable victimization, which is dependent on both i and t. It estimates how a temporal change in victimization is associated with a temporal change in the severity of mental health issues. The person-specific-mean-centered victimization is calculated as: person-specific-mean-centered victimization_ti = victimization_ti − $(1 / T) \sum_{i = T}^{T} X_{i t}$ . π₀₁ is the coefficient of grand-mean-centered victimization, a level-2 coefficient estimating how victimization against individual i differs from an average level of victimization experiences of all respondents. It is calculated as: grand-mean-centered victimization_t_i ⁼ $(1 / T) \sum_{i = T}^{T} X_{i t} - \underline{X_{..}}$ , where $\underline{X_{..}}$ is the average level of victimization. β₀₁–β₀₇ are coefficients of time-invariant variables such as race and education, which are only dependent on individual i. γ_0i is the random intercept, a function of the grand mean intercept (γ₀₀) and a random effect component (u_0i). ε_ti is the residual term.

A three-part analytical strategy was implemented to fulfill our research purposes. First, we conduct descriptive analysis and generate a series of visualizations to illustrate post-incarcerated individuals’ trajectories of victimization, mental health, and work hours over time. Second, using a longitudinal multilevel model, we estimate whether victimization is associated with mental health struggles, and, if so, via person-period or between-person effect. Lastly, we test whether poor mental health mediates the effect of violent victimization on work hours. We follow past practices of three-step modeling to test mediation effects (Dudley et al., 2004; Kenny, 2008). The three models test whether victimization (the predictor) is significantly associated with work hours (the outcome), whether mental health (the mediator) is significantly associated with work hours (the outcome), and whether the coefficient of predictor will change after the mediator is added to the model. In each multilevel model, all time-variant and time-invariant confounders of family and neighborhood environments are controlled. The sampling distribution of a mediation effect is complicated because the mediation effect is quantified by a product of at least two parameters, which might affect the robustness of results from the three-step modeling (Preacher & Selig, 2012). To confirm the results, we employ the SAS RBMLM procedure to conduct nonparametric bootstrapping with a total number of 1,000 resamples with the identical multilevel structure of our data, and these 1,000 samples are used to analyze the bootstrapped confidence intervals of coefficients in the third model (Version 9.4, SAS Institute Inc., Cary, NC, USA). If the 95% confidence interval of the predictor contains zero, the predictor’s effect is completely mediated by the mediator (e.g., Valente et al., 2020).

Results

Descriptive Statistics and Binary Analysis

Table 1 reports the descriptive statistics for all variables used in this analysis. On average, participants were nearly 29 years old when released from prison. The largest proportion of participants are Black (53%), followed by White (34%) and Other race (13%). Approximately half of the sample consists of SVORI program participants. On average, respondents have 12 years of education, which is equivalent to a high school diploma. Using the original items measuring victimization, we examined its prevalence by calculating the percentage of respondents who experienced violent victimization at least once over the 15 months regardless of the type of victimization. We observed that at T1, 26% of respondents reported any type of victimization. This percentage increased to 37% and 39% at T2 and T3, respectively. In other words, one in four returning citizens experienced violent victimization between release and the third month into reentry, and this figure increased to nearly 40% in the 15th month into reentry.

Table 1.

Descriptive Statistics (N = 724).

Variables	Mean or Percentage	Std. Dev.	Min.	Max.
Dependent variables
Poor mental health	2.10	0.80	1.48	7.29
Work hours	40.47	12.31	1.00	80.00
Independent variables
PP violent victimization	0.00	0.44	−3.35	3.98
BP violent victimization	0.00	0.75	−0.46	5.44
PP^a concerns over neighborhood disorder	0.00	0.91	−2.35	2.34
BP^b concerns over neighborhood disorder	0.00	0.71	−1.64	1.91
PP rating of job opportunities in neighborhood	0.00	0.50	−2.00	2.00
BP rating of job opportunities in neighborhood	0.00	0.73	−1.37	1.63
Family members have ever been imprisoned (yes)	19.48%
Family members currently use drugs	5.72%
Family members currently engage in illicit acts	2.06%
Family members use alcohol in respondent’s presence	21.05%
Being criticized excessively by family members	1.95	0.77	1.00	4.00
SVORI participant	50.85%
Age	29.09	7.31	18.00	73.00
Education (in years)	11.83	2.15	1.00	18.00
White	34.12%
Black	53.36%
Other race	12.52%

PP: person-period. For the detailed calculation of person-period parameters, please read the section of Analytic Strategies.

BP: between-person. For the detailed calculation of between-person parameters, please read the section of Analytic Strategies.

Figure 1 illustrates the temporal changes of violent victimization, poor mental health, and weekly work hours among respondents over the 15 months after release. Violent victimization and poor mental health were latent factors; the value on the Y-axis for the two variables reflects uncentered factor scores calculated based on item weights from factor analyses. As shown in the figure, respondents experienced escalated victimization as they progressed further into reentry. There is also noticeable temporal change in respondents’ mental health, with poor mental health peaking fifteen months after release. Lastly, we observed relatively minor fluctuations in their weekly work hours, with a slight increase in the sample average over time. While informative, these descriptive analyses tell little about the interrelationships between each of these three variables. Therefore, we employ multilevel longitudinal modeling to assess the nexus between them. Specifically, whether the effect of violent victimization on work hours operates through impacting mental health.

Figure 1.

Temporal changes in violent victimization, mental health, and work hours.

Multilevel Longitudinal Model Predicting Poor Mental Health

Following past practices (Hausmann et al., 2007), the person-period and between-person predictors were centered around zero prior to the regression analyses (Table 1). A positive coefficient of a person-period variable in a model would show the longitudinal effect of the variable when it is increases above the average level of a respondent, and a positive coefficient for a between-person variable would show the effect of the variable when a respondent has a value higher than the average of his peers (Hausmann et al., 2007).

Model 1 of Table 2 illustrates how victimization and the covariates are predictive of poor mental health. As shown, violent victimization is positively associated with poor mental health, and it operates at both at the person-period and between-person level. For participants who experienced more victimization incidents, they fared worse on mental health (b = 0.36, p < .001). Furthermore, one’s temporal fluctuation in mental health struggles is a function of temporal fluctuation in violent victimization experiences. At a time when respondents experienced a higher-than-usual number of victimization incidents, they reported a higher-than-usual level of mental health struggles (b = 0.17, p < .001).

Table 2.

Multilevel Longitudinal Model Results.

	Model 1; Poor Mental Health	Model 2; Work Hours	Model 3; Work Hours	Model 4; Work Hours
Parameters	b (SE)	b (SE)	b (SE)	b (SE)
PP violent victimization	0.17*** (0.03)	−1.03 (0.58)	—	−1.00 (0.59)
BP violent victimization	0.36*** (0.04)	−1.56* (0.65)	—	−1.02 (0.70)
PP poor mental health	—	—	−0.43 (0.68)	−0.24 (0.69)
BP poor mental health	—	—	−1.74*** (0.61)	−1.34* (0.67)
Age	0.00 (0.00)	0.01 (0.05)	0.02 (0.05)	0.01 (0.05)
Black	−-0.02 (0.05)	−2.78*** (0.82)	−2.71*** (0.82)	−2.80*** (0.82)
Other Race	0.05 (0.07)	0.44 (1.20)	0.50 (1.20)	0.49 (1.20)
Education level	–0.01 (0.01)	0.53*** (0.19)	0.54*** (0.19)	0.52** (0.19)
SVORI participant	0.00 (0.04)	−0.93 (0.74)	–0.95 (0.74)	–0.91 (0.74)
PP concerns over neighborhood disorder	0.06*** (0.02)	−0.06 (0.34)	–0.07 (0.35)	–0.04 (0.35)
BP concerns over neighborhood disorder	0.10* (0.04)	0.24 (0.74)	0.29 (0.74)	0.39 (0.74)
PP rating of job opportunities in neighborhood	–0.04 (0.02)	0.64 (0.48)	0.66 (0.48)	0.63 (0.48)
BP rating of job opportunities in neighborhood	–0.08** (0.03)	0.04 (0.53)	–0.04 (0.53)	–0.09 (0.53)
Family members have ever been imprisoned	0.07 (0.06)	−1.62 (1.03)	–1.55 (1.03)	–1.54 (1.03)
Family members currently use drugs	–0.01 (0.12)	2.33 (2.02)	1.86 (2.00)	2.34 (2.02)
Family members currently engage in illicit acts	0.14 (0.18)	0.81 (3.16)	0.89 (3.16)	0.86 (3.16)
Family members use alcohol in respondent’s presence	0.04 (0.06)	0.17 (0.94)	0.17 (0.94)	0.24 (0.94)
Being criticized excessively by family members	0.17*** (0.03)	−0.82 (0.50)	–0.64 (0.51)	−0.62 (0.51)
Random effects
Person-period intercept	0.22	50.01	49.61	49.53
Between-person intercept	0.20	90.81	91.12	90.92

Note. PP = person-period; BP = between-person.

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

Respondents’ concerns about neighborhood disorder and job opportunities were also associated with poorer mental health, with effects operating both at the between-person and person-period level. Participants who had more concerns about neighborhood safety were found to experience poorer mental health than their peers (b = 0.10, p < .05). Meanwhile, at a time when they report higher concerns over neighborhood disorder, they also report more severe mental health struggles (b = .06, p < .001). For the predictor of neighborhood job opportunities, we found that those who reported sufficient job opportunities in their neighborhoods were found to fare better on mental health (b = −.08, p < .01). Family environment also affected returning citizens’ mental health. Those who were excessively criticized by family members fared worse on mental health (b = 0.17, p < .001).

Multilevel Longitudinal Models Testing the Mediation Hypothesis

Model 2 in Table 2 tests whether violent victimization is significantly associated with work hours, without mental health—the mediator—in the model. Results show that victimization is a significant predictor of work hours among returning individuals. For respondents who experienced more victimization incidents, the work hours they report was significantly lower than other respondents. A one unit increase in victimization is associated with a 1.56-hour decrease in weekly work hours (b = −1.56, p < .05). Model 3 presents the results on the link between mental health and work hours, without the predictor of violent victimization. Results show that poor mental health emerged as a significant predictor of weekly work hours. For respondents with more severe mental health issues, their work hours are significantly fewer than their peers. Each one unit increase in mental health struggles is associated with a nearly 2-hour drop in work hours per week (b = −1.74, p < .001). Meanwhile, mental health does not affect work performance at the person-period level. Put differently, the fluctuation in one’s mental health over time does not affect one’s variation of work hours over time.

Moving on to the last hierarchical model, it assesses whether poor mental health mediates the effect of violent victimization on work hours (Model 4 of Table 2). The model results supported our hypothesis: poor mental health completely mediated the significant between-person effect of victimization on work hours. In this model, poor mental health exerted a significant effect on work hours at the between-person level: respondents with more severe mental struggles are found to work fewer hours per week (b = −1.34, p < .05). Meanwhile, the coefficient for violent victimization, which was statistically significant in model 2, is no longer significant in this model (b = −1.02, p > .05). Bootstrap tests of mediation showed that the indirect effect exerted by violent victimization on work hours via the mediator of poor mental health was −0.56 (95% CI: −1.08 to −.08; bootstrapped standard error: 0.26; p = .02), suggesting a statistically significant mediation effect. These results indicate that respondents who suffered more victimization incidents had more severe mental health struggles, and that they reported fewer weekly work hours. Regarding the effects of control variables, we found that respondents with more education (b = 0.52, p < .01) report more work hours than less educated respondents. Lastly, Black respondents reports fewer work hours than White respondents (b = −2.80, p < .001).

Discussion

Although successful economic reintegration after incarceration has been predicted by returning citizens’ education (Cnaan et al., 2008; Uggen et al., 2005), family and friend networks (Lattimore et al., 2010; Western & Sirois, 2019), and neighborhood job opportunities (Chamberlain & Wallace, 2016; Stansfield, 2016), limited research has been dedicated to examining whether violent victimization is detrimental to returning citizens’ mental health and capacity to work. This study drew on insights from Agnew’s (1992) GST and examined how victimization undermines reentry by deteriorating mental health and work capacity. In particular, given that few studies have used panel data to examine the time-variant nature of reentry experiences, this study employed longitudinal multilevel modeling to reveal the time dynamic effect of victimization. Several major findings emerged from the study.

First, results show that violent victimization against returning citizens is a significant predictor of their poor mental health. Victimization undermines returning citizens’ mental health via two mechanisms. On one hand, returning citizens who experienced greater victimization incidents were found to have poorer mental health. On the other hand, at a time when a returning citizen experienced an increase in victimization incidents, this person experienced greater deterioration in mental health. These significant effects were observed when adjusting for a wide range of covariates. The link between violent victimization and mental health deterioration found in this study echoes previous findings using samples of youth, elderly, and general populations, confirming the detrimental effect of violent victimization (Lynch et al., 2017; Turner et al., 2013; Tyler et al., 2014).

Individuals who had contact with the criminal justice system are more likely to come from marginal groups with past traumatic experiences and mental health issues (O’keefe & Schnell, 2007; Underwood & Washington, 2016; Wilper et al., 2009). During incarceration, their mental health needs are usually not adequately addressed (Brooker et al., 2002; Rutherford & Duggan, 2009). To exacerbate matters, upon release, most of them navigate lives in resource depleted, disadvantaged, and violence-ridden neighborhoods (Harding et al., 2013; Kubrin & Stewart, 2006). They may find it extremely challenging to quickly adapt to an environment completely different from prison and full of unknown risks. Without knowing the places not to go and people not to speak with, they may easily become victims of violence. Our data show that victimization triggers mental health issues for returning citizens. They might feel jumpy, hostile, develop low trust toward others after being victimized, and experience unrelenting stress. Clearly, victimization during reentry takes a heavy toll on justice-involved persons’ already vulnerable mental health.

Second, results show that poor mental health mediates the effect of violent victimization on work hours. Extant tests of GST showed that victimization undermines people’s functional outcomes by inducing negative emotions and deviant coping strategies such as drug use (Lo et al., 2008; Tyler et al., 2014). The present study examines work hours, an under-studied proxy of functional outcomes, and revealed that victimization impaired work capacity via poor mental health. This finding provides a significant piece of empirical support for the mediation proposition of GST. Within the longitudinal multilevel modeling framework, we found that victimization deteriorates work capacity at the between-person level: the variation of victimization experiences among respondents explains the variation in the work hours they reported, and the underpinning mechanism of this link lies in poor mental health. Victimization deteriorates mental health, which, in turn, undermines work capacity. Therefore, mental health not only mediates the link between negative stimuli and crime—as revealed in several prior studies (Slocum, 2010; Tittle et al., 2008)—but also the link between negative stimuli and work capacity.

Given that maintaining stable employment is one of the most critical tasks involved for a successful reintegration after imprisonment, our findings provide significant policy implications. First, initiatives are needed to address victimization during reentry. Currently, compared to the programs provided to women and elderly victims, victim-support programs for citizens returning from prison are relatively fewer. Social service agencies, correction agencies, and community organizations should provide programs to address victimization against this social group and provide shelter and other forms of support. Community supervision officers should also be alerted to possible victimization and refer individuals to victim support services.

Second, mental health services with a focus on victimization counseling should be tailored to address the specific life context of returning citizens. Training programs should be provided to mental health professionals such as clinical psychologists to equip them with a strong understanding of the reentry life context. Victimization, concerns over drug selling and crime in the neighborhood, and tensions with family members are salient triggers of mental health struggles among returning citizens (Harding et al., 2019; Mowen & Boman, 2019). Clinical psychologists and counselors should understand these challenges to effectively address returning citizens’ mental health issues and trauma. Due to limited education and job skills, the majority of returning citizens enter a low-wage labor market. They may be exploited and abused by employers and receive little to no sick leave to address mental health struggles. Reentry organizations should consider providing legal protections and counseling services to formerly incarcerated individuals who have experienced unfair treatment and exploitation by employers.

Beyond these practical implications, it is important to address several limitations of the study, which direct attention to opportunities for future research. First, the sample consisted of only male returning citizens, limiting the generalizability of findings. It is unclear whether similar results would emerge among female returning citizens. Compared to male returning citizens, a high proportion of female returning citizens are single parents who need to fulfill both the breadwinner and nurturing roles for children, which may place them in the middle of a distinct set of challenges (Brown & Bloom, 2018; Holtfreter & Wattanaporn, 2014).

Another limitation of the present study is that we were unable to provide definitive evidence of causality. Although the longitudinal data enabled us to conduct more rigorous analysis, the SVORI researchers could not randomize the environment in which post-incarcerated individuals reentered or their victimization experiences. Furthermore, the length of the study period from wave one to wave three was only 1 year, which might have resulted in limited within-person variation in work hours. This could potentially explain the null effects observed at the within-person level of analysis.

Furthermore, all respondents in this study had been incarcerated in state prisons, whose convictions for the current incarceration were marked as serious and violent. It is beyond the scope of this study to assess the impact of victimization on mental health and work capacity among people both with and without criminal records. Lastly, controls for other potential factors contributing to lower work capacity, such as self-control, past delinquency, and work skills were not included in the analyses. Future research should collect data on these and other potential confounders to assess the nexus of victimization, mental health, and work capacity during reentry.

Conclusion

Researchers have widely documented that maintaining stable employment is a salient catalyst to a successful reintegration (Rydberg, 2018; Sampson & Laub, 2003; Williams et al., 2019). The present longitudinal study investigated the nexus of post-release violent victimization, mental health, and work hours. Results show that returning citizens are at a high risk of violent victimization, and their mental health covaries with the temporal change of victimization experiences. Violent victimization also impairs returning citizens’ work capacity via its deleterious effect on mental health. Findings yielded significant implications for policies that address returning citizens’ needs and facilitate their economic reintegration. Given the costs of reincarceration, both fiscal and social, such policy efforts should have widespread appeal and therefore should be aggressively pursued.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Lin Liu

Notes

Author Biographies

Lin Liu, PhD is an Assistant Professor in the Department of Criminology and Criminal Justice at Florida International University. Her research interests revolve race and policing, sentencing, former prisoners’ reentry, and life-course offending trajectories. Recent publications appear in Crime & Delinquency, Justice Quarterly, Criminal Justice Policy Review, Women & Criminal Justice, The Prison Journal, Journal of Offender Rehabilitation, Social Science Research, Marquette Law Review, Contemporary Justice Review, International Journal of Offender Therapy and Comparative Criminology, and Sociological Inquiry.

Gabriella Lazazzara is currently a PhD student in the International Crime and Justice Program at Florida International University. Her research interests include juvenile crime, restorative justice, and prosecutor biases in the criminal justice system.

Ryan Charles Meldrum is professor and director of Research and Communications within the Department of Criminology and Criminal Justice at Florida International University. His research interests include testing and refining theories of delinquency and crime. His recent research has appeared in journals such as Criminology, Justice Quarterly, and Journal of Research in Crime and Delinquency.

References

Agnew

(1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30(1), 47–88.

Agnew

(1999). A general strain theory of community differences in crime rates. Journal of Research in Crime and Delinquency, 36(2), 123–155.

Agnew

(2001). Building on the foundation of general strain theory: Specifying the types of strain most likely to lead to crime and delinquency. Journal of Research in Crime and Delinquency, 38(4), 319–361.

Agnew

Brezina

Wright

J. P.

Cullen

F. T.

(2002). Strain, personality traits, and delinquency: Extending general strain theory. Criminology, 40(1), 43–72.

Allison

P. D.

(2019). Asymmetric fixed-effects models for panel data. Socius Sociological Research for a Dynamic World. Advance online publication. https://doi.org/10.1177/2378023119826441

Alper

Durose

M. R.

Markman

(2018). 2018 Update on prisoner recidivism: A 9-year follow-up period (2005—2014). US Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Archer

R. J. L.

(2019). Sexual assault victimization, fear of sexual assault, and self-protective behaviors: A test of general strain theory. Victims & Offenders, 14(4), 387–407.

Arditti

Few

(2008). Maternal distress and women's reentry into family and community life. Family Process, 47(3), 303–321.

Aresti

Eatough

Brooks-Gordon

(2010). Doing time after time: An interpretative phenomenological analysis of reformed ex-prisoners’ experiences of self-change, identity and career opportunities. Psychology, Crime & Law, 16(3), 169–190.

10.

Bakken

N. W.

Visher

C. A.

(2018). Successful reintegration and mental health: An examination of gender differences among reentering offenders. Criminal Justice and Behavior, 45(8), 1121–1135.

11.

Banyard

Potter

Turner

(2011). The impact of interpersonal violence in adulthood on women’s job satisfaction and productivity: The mediating roles of mental and physical health. Psychology of Violence, 1(1), 16.

12.

Baron

S. W.

(2006). Street youth, strain theory, and crime. Journal of Criminal Justice, 34(2), 209–223.

13.

Berg

M. T.

Huebner

B. M.

(2011). Reentry and the ties that bind: An examination of social ties, employment, and recidivism. Justice Quarterly, 28(2), 382–410.

14.

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.

15.

Bersani

B. E.

Laub

J. H.

Nieuwbeerta

(2009). Marriage and desistance from crime in the Netherlands: Do gender and socio-historical context matter?. Journal of Quantitative Criminology, 25(1), 3–24.

16.

Biebl

S. J.

DiLalla

L. F.

Davis

E. K.

Lynch

K. A.

Shinn

S. O.

(2011). Longitudinal associations among peer victimization and physical and mental health problems. Journal of Pediatric Psychology, 36(8), 868–877.

17.

Brooker

Repper

Beverley

Ferriter

Brewer

(2002). Mental health services and prisoners: A review. Mental Health Task Force.

18.

Brown

Bloom

B. E.

(2018). Women’s desistance from crime: A review of theory and the role higher education can play. Sociology Compass, 12(5), e12580.

19.

Bushway

S. D.

(2004). Labor market effects of permitting employer access to criminal history records. Journal of Contemporary Criminal Justice, 20(3), 276–291.

20.

Chamberlain

A. W.

Wallace

(2016). Mass reentry, neighborhood context and recidivism: Examining how the distribution of parolees within and across neighborhoods impacts recidivism. Justice Quarterly, 33(5), 912–941.

21.

Ciavolino

(2012). General distress as second order latent variable estimated through PLS-PM approach. Electronic Journal of Applied Statistical Analysis, 5(3), 458–464.

22.

Cnaan

R. A.

Draine

Frazier

Sinha

J. W.

(2008). Ex-prisoners’ re-entry: An emerging frontier and a social work challenge. Journal of Policy Practice, 7(2–3), 178–198.

23.

Derogatis

L. R.

(1975). Brief Symptom Inventory. Clinical Psychometric Research.

24.

Dudley

W. N.

Benuzillo

J. G.

Carrico

M. S.

(2004). SPSS and SAS programming for the testing of mediation models. Nursing Research, 53(1), 59–62.

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 (Vol. 28). US Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

27.

Fahey

Roberts

Engel

(2006). Employment of ex-offenders: Employer perspectives. Crime and Justice Institute, 55(4), 297–305.

28.

Froggio

(2007). Strain and juvenile delinquency: A critical review of Agnew’s general strain theory. Journal of Loss and Trauma, 12(4), 383–418.

29.

Waasdorp

T. E.

Randolph

J. A.

Bradshaw

C. P.

(2021). Peer victimization and mental health problems: Racial-ethnic differences in the buffering role of academic performance. Journal of Youth and Adolescence, 50(9), 1839–1855.

30.

Gideon

(2007). Family role in the reintegration process of recovering drug addicts: A qualitative review of Israeli offenders. International Journal of Offender Therapy and Comparative Criminology, 51(2), 212–226.

31.

Gwynne

J. L.

Yesberg

J. A.

Polaschek

D. L.

(2020). Life on parole: The quality of experiences soon after release contributes to a conviction-free re-entry. Criminal Behaviour and Mental Health, 30(6), 290–302.

32.

Hager

A. D.

Leadbeater

B. J.

(2016). The longitudinal effects of peer victimization on physical health from adolescence to young adulthood. Journal of Adolescent Health, 58(3), 330–336.

33.

Hamilton

I. S.

(2016). Employment of ex-prisoners with mental health problems: A review. Journal of Criminological Research, Policy and Practice, 2(1), 40–53.

34.

Harding

D. J.

Morenoff

J. D.

Herbert

C. W.

(2013). Home is hard to find: Neighborhoods, institutions, and the residential trajectories of returning prisoners. The Annals of the American Academy of Political and Social Science, 647(1), 214–236.

35.

Harding

D. J.

Morenoff

J. D.

Wyse

J. J.

(2019). On the outside: Prisoner reentry and reintegration. University of Chicago Press.

36.

Hausmann

L. R. M.

Schofield

J. W.

Woods

R. L.

(2007). Sense of belonging as a predictor of intentions to persist among African American and White first-year college students. Research in Higher Education, 48(7), 803–839.

37.

Hay

Meldrum

(2010). Bullying victimization and adolescent self-harm: Testing hypotheses from general strain theory. Journal of Youth and Adolescence, 39(5), 446–459.

38.

Hay

Meldrum

Mann

(2010). Traditional bullying, cyber bullying, and deviance: A general strain theory approach. Journal of Contemporary Criminal Justice, 26(2), 130–147.

39.

Herge

W. M.

La Greca

A. M.

Chan

S. F.

(2016). Adolescent peer victimization and physical health problems. Journal of Pediatric Psychology, 41(1), 15–27.

40.

Holtfreter

Wattanaporn

K. A.

(2014). The transition from prison to community initiative: An examination of gender responsiveness for female offender reentry. Criminal Justice and Behavior, 41(1), 41–57.

41.

Holzer

H. J.

Raphael

Stoll

M. A.

(2002). Can employers play a more positive role in prisoner reentry?. Urban Institute.

42.

Iverson

K. M.

Vogt

Maskin

R. M.

Smith

B. N.

(2017). Intimate partner violence victimization and associated implications for health and functioning among male and female post-9/11 veterans. Medical Care, 55, S78–S84.

43.

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.

44.

Juvonen

Nishina

Graham

(2000). Peer harassment, psychological adjustment, and school functioning in early adolescence. Journal of Educational Psychology, 92(2), 349.

45.

Kenny

D. A.

(2008). Reflections on mediation. Organizational Research Methods, 11(2), 353–358.

46.

Kubrin

C. E.

Stewart

E. A.

(2006). Predicting who reoffends: The neglected role of neighborhood context in recidivism studies. Criminology: An Interdisciplinary Journal, 44(1), 165–197.

47.

Laird

(2013). Doing time extended: Ex-offenders face tens of thousands of legal restrictions, bias and limits on their rights. ABAJ, 99, 50.

48.

Lattimore

P. K.

Steffey

D. M.

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

49.

Lattimore

P. K.

Steffey

D. M.

Visher

C. A.

(2010). Prisoner reentry in the first decade of the twenty-first century. Victims and Offenders, 5(3), 253–267.

50.

Lattimore

P. K.

Visher

C. A.

(2009). The multi-site evaluation of SVORI: Summary and synthesis. RTI International.

51.

Laub

J. H.

Nagin

D. S.

Sampson

R. J.

(1998). Trajectories of change in criminal offending: Good marriages and the desistance process. American Sociological Review, 63(2), 225–238.

52.

Lerner

Henke

R. M.

(2008). What does research tell us about depression, job performance, and work productivity?. Journal of Occupational and Environmental Medicine, 50(4), 401–410.

53.

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.

54.

Liu

Mowen

T. J.

Visher

C. A.

Sun

(2022). Violent Victimization During Reentry: Prevalence, Triggers, and Impact on Mental Health. Justice Quarterly. https://doi.org/10.1080/07418825.2022.2104747

55.

C. C.

Kim

Y. S.

Church

W. T.

(2008). The effects of victimization on drug use: A multilevel analysis. Substance Use & Misuse, 43(10), 1340–1361.

56.

Lukies

Graffam

Shinkfield

A. J.

(2011). The effect of organisational context variables on employer attitudes toward employability of ex-offenders. International Journal of Offender Therapy and Comparative Criminology, 55(3), 460–475.

57.

Lynch

S. M.

DeHart

D. D.

Belknap

Green

B. L.

Dass-Brailsford

Johnson

K. M.

Wong

M. M.

(2017). An examination of the associations among victimization, mental health, and offending in women. Criminal Justice and Behavior, 44(6), 796–814.

58.

Mallik-Kane

Visher

C. A.

(2008). Health and prisoner reentry: How physical, mental, and substance abuse conditions shape the process of reintegration (p. 82). Urban Institute Justice Policy Center.

59.

McDougall

Vaillancourt

. (2015). Long-term adult outcomes of peer victimization in childhood and adolescence: Pathways to adjustment and maladjustment. American Psychologist, 70(4), 300.

60.

Merikangas

K. R.

Ames

Cui

Stang

P. E.

Ustun

T. B.

Von Korff

Kessler

R. C.

(2007). The impact of comorbidity of mental and physical conditions on role disability in the US adult household population. Archives of General Psychiatry, 64(10), 1180–1188.

61.

Moon

McCluskey

Morash

(2019). Aggression against middle and high school teachers: Duration of victimization and its negative impacts. Aggressive Behavior, 45(5), 517–526.

62.

Mowen

T. J.

Boman

J. H.

IV . (2019). Do we have it all wrong? The protective roles of peers and criminogenic risks from family during prison reentry. Crime & Delinquency, 65(5), 681–704.

63.

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.

64.

Myers

D. L.

Olson

(2013). Offender reentry and reintegration: Policy and research. Criminal Justice Policy Review, 24(1), 3–8.

65.

O’keefe

M. L.

Schnell

M. J.

(2007). Offenders with mental illness in the correctional system. Journal of Offender Rehabilitation, 45(1–2), 81–104.

66.

Okumu

Kim

Y. K.

Sanders

J. E.

Makubuya

Small

Hong

J. S.

(2020). Gender-specific pathways between face-to-face and cyber bullying victimization, depressive symptoms, and academic performance among US adolescents. Child Indicators Research, 13(6), 2205–2223.

67.

Osgood

D. W.

Anderson

A. L.

(2004). Unstructured socializing and rates of delinquency. Criminology, 42(3), 519–550.

68.

Ostrowsky

M. K.

Messner

S. F.

(2005). Explaining crime for a young adult population: An application of general strain theory. Journal of Criminal Justice, 33(5), 463–476.

69.

Owens

C. D.

Jr. (2009). Social symbols, stigma, and the labor market experiences of former prisoners. Journal of Correctional Education, 60(4), 316–342.

70.

Pager

Western

(2009). Investigating prisoner reentry: The impact of conviction status on the employment prospects of young men. National Institute of Justice.

71.

Preacher

K. J.

Selig

J. P.

(2012). Advantages of Monte Carlo confidence intervals for indirect effects. Communication Methods and Measures, 6(2), 77–98.

72.

Price

Polk

Hill

N. E.

Liang

Perella

(2019). The intersectionality of identity-based victimization in adolescence: A person-centered examination of mental health and academic achievement in a US high school. Journal of Adolescence, 76, 185–196.

73.

Radloff

L. S.

(1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401.

74.

Raskauskas

Rubiano

Offen

Wayland

A. K.

(2015). Do social self-efficacy and self-esteem moderate the relationship between peer victimization and academic performance?. Social Psychology of Education, 18(2), 297–314.

75.

Raudenbush

S. W.

Bryk

A. S.

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

76.

Renner

L. M.

(2009). Intimate partner violence victimization and parenting stress: Assessing the mediating role of depressive symptoms. Violence Against Women, 15(11), 1380–1401.

77.

Rutherford

Duggan

(2009). Meeting complex health needs in prisons. Public Health, 123(6), 415–418.

78.

Rydberg

(2018). Employment and housing challenges experienced by sex offenders during reentry on parole. Corrections, 3(1), 15–37.

79.

Sampson

R. J.

Laub

J. H.

(1995). Crime in the making: Pathways and turning points through life. Harvard University Press.

80.

Sampson

R. J.

Laub

J. H.

(2003). Life-course desisters? Trajectories of crime among delinquent boys followed to age 70. Criminology, 41(3), 555–592.

81.

Schwartz

Gorman

A. H.

Nakamoto

Toblin

R. L.

(2005). Victimization in the peer group and children’s academic functioning. Journal of Educational Psychology, 97(3), 425.

82.

Sheppard

Ricciardelli

(2020). Employment after prison: Navigating conditions of precarity and stigma. European Journal of Probation, 12(1), 34–52.

83.

Skogan

(1986). Fear of crime and neighborhood change. Crime and Justice, 8, 203–229.

84.

Slocum

L. A.

(2010). General strain theory and continuity in offending over time: Assessing and extending GST explanations of persistence. Journal of Contemporary Criminal Justice, 26(2), 204–223.

85.

Solomon

A. L.

(2012). In search of a job: Criminal records as barriers to employment. NIJ Journal, 270(1), 43–51.

86.

Stansfield

(2016). Juvenile desistance and community disadvantage: The role of appropriate accommodations and engagements. Justice Quarterly, 33(4), 708–728.

87.

Straus

M. A.

(1990). The conflict tactics scales and its critics: An evaluation and new data on validity and reliability. In Straus

M. A.

Gelles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (pp. 49–73). Transaction.

88.

Suls

(2018). Toxic affect: Are anger, anxiety, and depression independent risk factors for cardiovascular disease?. Emotion Review, 10(1), 6–17.

89.

Temple

J. R.

Choi

H. J.

Elmquist

Hecht

Miller-Day

Stuart

G. L.

Brem

Wolford-Clevenger

(2016). Psychological abuse, mental health, and acceptance of dating violence among adolescents. Journal of Adolescent Health, 59(2), 197–202.

90.

Thompson

A. J.

Pickett

J. T.

(2021). Asymmetry in process-based model relationships: A longitudinal study of adjudicated adolescents. Criminology, 59(4), 585–609.

91.

Tittle

C. R.

Broidy

L. M.

Gertz

M. G.

(2008). Strain, crime, and contingencies. Justice Quarterly, 25(2), 283–312.

92.

Tolman

R. M.

Wang

H. C.

(2005). Domestic violence and women’s employment: Fixed effects models of three waves of women’s employment study data. American Journal of Community Psychology, 36(1), 147–158.

93.

Travis

(2002). Beyond the prison gates: The state of parole in America. Justice Policy Center.

94.

Turanovic

J. J.

(2019). Heterogeneous effects of adolescent violent victimization on problematic outcomes in early adulthood. Criminology, 57(1), 105–135.

95.

Turner

H. A.

Finkelhor

Ormrod

(2010). The effects of adolescent victimization on self-concept and depressive symptoms. Child Maltreatment, 15(1), 76–90.

96.

Turner

H. A.

Shattuck

Hamby

Finkelhor

(2013). Community disorder, victimization exposure, and mental health in a national sample of youth. Journal of Health and Social Behavior, 54(2), 258–275.

97.

Tyler

K. A.

Kort-Butler

L. A.

Swendener

(2014). The effect of victimization, mental health, and protective factors on crime and illicit drug use among homeless young adults. Violence and Victims, 29(2), 348–362.

98.

Uggen

Kruttschnitt

(1998). Crime in the breaking: Gender differences in desistance. Law & Society Review, 32(2), 339–366.

99.

Uggen

Wakefield

Western

(2005). Work and family perspectives on reentry. In Travis

Visher

(Eds.), Prisoner reentry and crime in America (pp. 209–243). Cambridge University.

100.

Underwood

L. A.

Washington

(2016). Mental illness and juvenile offenders. International Journal of Environmental Research and Public Health, 13(2), 228.

101.

Valente

M. J.

Rijnhart

J. J.

Smyth

H. L.

Muniz

F. B.

MacKinnon

D. P.

(2020). Causal mediation programs in R, M plus, SAS, SPSS, and Stata. Structural Equation Modeling: A Multidisciplinary Journal, 27(6), 975–984.

102.

Visher

(2004). Returning home: Understanding the challenges of prisoner reentry: Maryland pilot study: Findings from Baltimore. Urban Institute.

103.

Visher

(2007). Returning home: Emerging findings and policy lessons about prisoner reentry. Federal Sentencing Reporter, 20(1), 93–102.

104.

Visher

Debus

Yahner

(2008). Employment after prison: A longitudinal study of releasees in three states. The Urban Institute.

105.

Visher

Debus-Sherrill

S. A.

Yahner

(2011). Employment after prison: A longitudinal study of former prisoners. Justice Quarterly, 28(5), 698–718.

106.

Ward

J. T.

Forney

(2020). Unpacking within-and between-person effects of unstructured socializing and differential association on solo-and co-offending. Journal of Criminal Justice, 70, 101720.

107.

Western

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

108.

Western

Sirois

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

109.

Widdowson

A. O.

Hay

Siennick

S. E.

(2021). Romantic partners and young adult offending: Considering the role of partner’s socioeconomic characteristics. Criminology, 59(1), 158–190.

110.

Williams

J. M.

Wilson

S. K.

Bergeson

(2019). “It’s hard out here if you’re a Black felon”: A critical examination of Black male reentry. The Prison Journal, 99(4), 437–458.

111.

Wilper

A. P.

Woolhandler

Boyd

J. W.

Lasser

K. E.

McCormick

Bor

D. H.

Himmelstein

D. U.

(2009). The health and health care of US prisoners: results of a nationwide survey. American Journal of Public Health, 99(4), 666–672.