Religious Identity and the Long-Term Effects of Religious Involvement, Orientation, and Coping in Prison

http://journals.sagepub.com/doi/full/10.1177/0093854818801410

 

Richard Stansfield

Rutgers University – Camden

 

Thomas O’ Connor

Transforming Corrections

 

Jeff Duncan

Oregon Department of Corrections

 

 Abstract

Criminological literature has largely ignored the specific religious and spiritual ways people in prison identify and make meaning in their life. Analysis of whether identification predicts recidivism above and beyond criminal risk could have significant implications for assessment and treatment of people in prison. Using data from a Spiritual Assessment of 571 people in prison in Oregon, this study compares recidivism behavior of individuals who identify as both religious and spiritual, with those who identify as only spiritual but not religious, religious but not spiritual, and neither spiritual nor religious. As expected, religious and spiritual participants were less likely to reoffend than spiritual but not religious inmates. Group differences were partially explained by intrinsic religious orientation and service attendance. The results highlight the importance of ensuring support for persons in prison in the process of making meaning, in addition to supporting the work of prison chaplains and religious volunteers in prison.

Keywords: religion, spirituality, religious orientation, coping, recidivism

A growing literature has considered the role of religiosity and spirituality in the prison context (Dye, Aday, Farney, & Raley, 2014; Hallett, Hays, Johnson, Jang, & Duwe, 2016; Johnson, 2011; Levitt & Loper, 2009). These studies suggest that religious beliefs and behaviors may provide people in prison support, hope, and a way to make sense of their situation. The perceived or actual social support supplied by community and social network members, including faith communities, has been considered one of the key theoretical principles in explaining involvement in crime (Cullen, 1994), and helping people cope with criminogenic strains and stresses (Agnew, 1992; Sampson & Laub, 1993) so that individuals who receive social support should, therefore, also desist from criminal offending (Colvin, Cullen, & Ven, 2002; Listwan, Colvin, Hanley, & Flannery, 2010). Reentry scholarship has highlighted the critical importance of social ties for helping ex-offenders cope with the many challenges they face in prison and after release (Bales & Mears, 2008; Berg & Huebner, 2011; Boman & Mowen, 2017; Cochran, 2014; Duwe & Johnson, 2016; Duwe & King, 2013; Visher & Travis, 2003). For many people in prison, especially those who face the greatest social isolation, faith-based services may offer a critical source of social support in prison. Furthermore, on-going connections with clergy, mentors, and volunteers may all provide a critical level of support otherwise missing during a person’s transition back to the community (Johnson, 2011).

In addition to increasing social support, faith-informed services or programs also offer men and women a way of fostering meaning making in their lives. Each person has choices for how they create meaning (Taylor, 2007). Humanists or secularists find meaning in relationship with humanity and life itself but tend to eschew a relationship with a divine being or spiritual source of meaning that is beyond the boundaries of earth. People who are spiritual do relate to a transcendent source of meaning, but they eschew any organized religious dimension. People who are religious prefer to derive meaning from within the context of one of a wide array of organized and evolving religious traditions, all of which relate meaning to a transcendent source, usually called God or the Divine. While most people who identify as religious are also spiritual (Zinnbauer et al., 1997), the United States has seen a growing trend in what Robert Fuller (2001) calls “unchurched traditions,” where individuals claim to be spiritual but not religious (Kosmin & Keysar, 2009; Mercadante, 2014; Zinnbauer et al., 1997). Recent research documents that subjective spirituality and religiosity (grounded in tradition) are highly independent constructs (Saucier & Skrzypińska, 2006; Wong & Vinsky, 2009). These distinctions also exist behind bars, with evidence that almost a quarter of people surveyed in Oregon prisons identify as spiritual but not religious, with fewer religious but not spiritual (O’Connor & Duncan, 2011).

One consideration that has yet to be addressed in the reentry literature is how the specific way an individual identifies affects the likelihood of recidivism, despite a growing delineation of religion and spirituality concepts in other literatures (Ammerman, 2013). A significant body of research has explored the relation between religion and spirituality in prison and recidivism, including numerous systematic reviews and meta-analyses on the topic (e.g., Clear et al., 1992; Dodson, Cabage, & Klenowski, 2011; Kelly, Polanin, Jang, & Johnson, 2015; Power, Ritchie, & Madill, 2014).While these studies are largely sanguine about the role of religion for ex-offenders after release, there is wide variation in the conceptualization and measurement of religion. Concepts commonly studied include religious attendance, religious salience (i.e. how important is religion in someone’s life), participation in faith-based programs, perceived support from religious participation, or a combination of these concepts.

Humanism and spirituality therefore are rarely studied as separate concepts from religion or religiousness in criminology. Jang and Franzen (2013) suggested this may be due in part to criminologists using large surveys to study the relation of meaning-making paths and crime, which typically only contain a small number of items capturing how often one attends church and how important religion is in their life (i.e., Pathways to Desistance, Pirutinsky, 2014; Stansfield, 2017; Serious Violent Offender Reentry Initiative, Bakken, DeCamp, & Visher, 2014; Stansfield, Mowen, O’Connor, & Boman, 2017; National Longitudinal Study of Adolescent Health, Jang & Franzen, 2013; Ulmer, Desmond, Jang, & Johnson, 2012; National Youth Survey, Chu, 2007). Because spiritual-only individuals tend to eschew involvement in organized religion, they may participate in faith-based programs at a lower rate and be less likely to draw support from religion while in prison.

Using data from a detailed spiritual assessment (SA) of persons incarcerated in Oregon, the current study aims to address this shortcoming by exploring the role of identification, religious orientation, coping styles, and service attendance. Specifically, we assess whether there are differences in recidivism depending on whether a person identifies as religious and spiritual, spiritual only, religious only, or neither. We then further examine the role of religious orientations and behaviors in the reentry process.

Religion, Crime, and Reentry

Several criminological and psychological perspectives suggest why adherence to humanist, spiritual, or religious ways of making meaning can reduce the risk of criminal involvement. From the perspectives of social control (Hirschi, 1969) and social support (Cullen,1994), recent reviews of the religion–crime relationship have suggested that religiosity provides a form of social support, connects individuals to prosocial peers, and the feelings of support a person receives act as a coping mechanism against strain (Adamczyk, Freilich, & Kim, 2017; Chu, 2007; Ulmer et al., 2012). There is also ample literature evidencing a positive relationship between religious behaviors and beliefs and inmate adjustment to prison (Dye et al., 2014; Levitt & Loper, 2009). A growing literature exploring the more specific relationship between religion and/or spirituality in prison and reentry success, often measured by reoffending and reincarceration (Duwe & King, 2013; Giordano, Longmore, Schroeder, & Seffrin, 2008; Hallett et al., 2016; Johnson, 2004, 2011; Pirutinsky, 2014; Power et al., 2014; Stansfield, Mowen, & O’Connor, 2018; Young, Gartner, O’Connor, Larson, & Wright, 1995), has also proliferated. Social control theories are commonly cited in these studies to explain the prosocial effect of religious identification and behavior, positing an important connection to a more prosocial network and social support, as well as offering more structured conventional activities, rendering individuals less likely to recidivate (Giordano et al., 2008; O’Connor & Perreyclear, 2002; Schroeder & Frana, 2009; Stansfield et al., 2017). As Stansfield and colleagues (2017) concluded, “religion can form the basis for an institutional support network through which recently released prisoners can build and repair relationships, find jobs, and establish social support” (p. 136).

Germaine to this literature are a set of important findings from a series of meta-analytic studies commissioned by the American Psychological Association (Norcross & Wampold, 2011). These studies examined the outcomes from adapting psychotherapy to eight different patient characteristics, one of which was the religious and spiritual outlooks or approaches of patients. The meta-analyses categorized these outlooks into four different types of spirituality based on the type of object a person feels a sense of closeness or connection to: (a) humanistic spirituality, (b) nature spirituality, (c) cosmic spirituality, and (d) religious spirituality (Worthington, Hook, Davis, & McDaniel, 2011). The four factors were demonstratively effective, and thus, the American Psychological Association has recommended that all psychologists and counselors integrate humanism, spirituality, and religion into their work in a way that matches each client’s particular way of establishing ultimate meaning in life or feeling connected to something that is vitally important in their lives (Norcross & Wampold, 2011; Worthington et al., 2011). To help further refine the assessment and treatment of people in prison, more work is needed to assess how a person’s particular way of making meaning is associated with long-term benefits.

Religious and Spiritual Identification, Orientation, and Coping Style

While the above research is instructive, criminologists have often used a very broad definition of religious identification and behavior, often called spirituality, regardless of one’s precise identification or way of making meaning (Emmons, 2005; Jang & Franzen, 2013).There are some notable exceptions, studies which have explored the implications for crime or recidivism of group differences in religious identification. An earlier study by Giordano and colleagues (2008) distinguished between religious participation and spirituality, although as Jang and Franzen (2013) pointed out, their definition of spirituality included perceived closeness to God, a concept with orthodox religious undertones, and thus not necessarily distinct from religious and spiritual. In one recent important study, Jang and Franzen (2013) explored the effect of identifying as spiritual but not religious, spiritual and religious, religious but not spiritual, and neither spiritual or religious, on the effect of crime during emerging adulthood. Using data from the third wave of the National Longitudinal Study of Adolescent Health, the authors used a Structural Equation Model approach to link identification with recent violent and property offending. Findings indicated that only spiritual individuals were more prone to violent crime than their religious andspiritual counterparts – a difference partly explained by lower religious involvement and self-control.

These findings mirrored recent work in the field of psychology and mental health that suggested individuals who profess spiritual beliefs, without the framework of organized religion, often have poorer mental health outcomes than individuals who are religiously affiliated, and individuals who are neither religious nor spiritual or humanist (King et al., 2013; Speck, Higginson, & Addington-Hall, 2004). As revealed by Jang and Franzen (2013), however, this important distinction is rarely considered by criminologists. Thus, while recent attention has been paid to the importance of religion among prisoners and ex-offenders, accurately capturing the ways individuals identify has remained a challenge, despite important theoretical reasons to expect differences.

Present Study

Although religion and spirituality share many characteristics, many forms of spirituality do not stipulate behavioral patterns (Hill et al., 2000), and people who identify as only spiritual are less likely to engage in traditional or communal faith-based programs (Zinnbauer et al., 1997). Both religion and spirituality can manifest in either healthy or unhealthy ways (Pargament, 1999), but with the absence of the religious framework and faith community, we hypothesize that spiritual-only people may be at a disadvantage after release from prison because of lower social support. By contrast, we hypothesize that identification as religious and spiritual will be associated with reduced recidivism, given that individuals can draw on the social support offered by the faith community and use their religion or spirituality as a coping mechanism. We did not hypothesize for the religious but not spiritual group because literature suggests the number of people in this category tends to be very small and different than the religious and spiritual group (Zinnbauer et al., 1997). These hypotheses guide the current study as we consider the impact of religious identification on the likelihood and rate of reoffending among formerly incarcerated individuals.

We also seek to explain these expected group differences in reoffending by measures of religious orientation and coping, which are likely to differ across group identification. Prior studies in religion have sought to determine whether individuals use their religion or spirituality in a more internal or intrinsic way to make meaning or in a more external manner to create social support and a social life. In a survey of Oregon inmates, O’Connor and Duncan (2011) found that prisoners were more likely to attend a religious or spiritual service for internal reasons, rather than to meet other inmates, volunteers, and to make friends. We explore the importance of religious orientation for recidivism.

Proponents of the value of faith in helping people desist from crime also highlight the value of religious leaders, prison chaplains, and the humanist, spiritual, and religious volunteers who assist prisoners in coping with the dehumanizing context of prison and the multitude of challenges accompanying the reentry process (Maruna, Wilson, & Curran, 2006; Pew Research Center, 2012; Schroeder & Frana, 2009; Stansfield et al., 2017). Because of this we assess attendance at the humanist, spiritual, and religious programs available to our study population. We also consider whether expected group differences in reoffending are explained by the way individuals use their religion or spirituality to cope with problems. For example, individuals vary greatly in the degree of responsibility assigned to the self and God for dealing with problems (Pargament et al., 1988; Schaefer & Gorsuch, 1993). We hypothesize that individuals who identify as religious and spiritual, will be more likely to turn to their faith and the faith community to deal with problems in a more collaborative manner, and this cooperation (whether perceived or actual) will explain the expected differences in reoffending among these groups.

 Data and Method

Data

Data for this study were provided by Correctional Services in the Oregon Department of Corrections (ODOC). The unit is responsible for overseeing intake and assessment, institutional services such as health, drug and alcohol, education, volunteer and Humanist, Spiritual, and Religious (HSR) services, in addition to reentry services across all prisons in Oregon. The current study relies on data from a Spiritual Assessment (SA) voluntarily completed by 807 persons admitted to the ODOC in 2004 and who served at least one full year in prison. Everyone admitted in 2004 was eligible to complete the SA, a paper and pencil survey handed out during intake throughout 2004. Almost a quarter of people admitted chose to complete the survey.Although the mean level of attendance at an HSR event was different between people who volunteered to complete a spiritual assessment and those who did not, comparisons of means revealed that persons who completed the assessment did not have significantly different demographics, including assessed risk for recidivism ( t= 1.264), compared to admits who did not complete the assessment.

The SA asked incoming prisoners about their religious identification and affiliation, religious history, religious orientation, and spiritual coping. To examine the long-term recidivistic behavior of individuals, we retained everyone who was released at least 8 years prior to follow up data collection, so that we could use a fixed follow-up period of 8 years. Data on arrests, including the date, charge, and outcome of new incidents were collected by the ODOC in 2017. Of the original 807 individuals who entered ODOC in 2004 and completed a spiritual assessment, 169 persons were only released recently or not released at all, and thus not included in the analysis. An additional 22 were dropped prior to analysis because they were released to an immigration detainer after identification by U.S. Immigration and Customs Enforcement, and deported. Another 45 were dropped prior to analysis due to incomplete data. This leaves an effective sample of 571 for analysis.

Setting

The prison chaplaincy system in Oregon differs in several ways to how prison chaplaincy is organized in many other US state and international prison systems. Many prison systems, for example, use unpaid volunteer chaplains, or have very few chaplains for the number of people in prison.  The ODOC, however, hires full-time professional chaplains who are relatively well paid, and hires a robust number of them to reach the entire inmate population. In Oregon, the chaplains answer to their own centrally-based Head Chaplain who sets the direction for the chaplaincy work in all 14 of the state prisons, and oversees the budget, goals, policies, practices, hiring, training and supervision of the unit staff. This releases the chaplains from having to focus primarily on maintaining the safety and security of the prisons they work in, and allows them to focus more on their unique roles of: 1) supporting both staff and inmates; 2) selecting, training, coaching and organizing volunteers; and 3) fostering a context that encourages rather than discourages the emergence and development of humanity, meaning, and new pro-social skills and ways of being in community.

The chaplains in Oregon meet the diverse and widespread HSR needs of the men and women in prison with the help of over 2,000 volunteers (O’Connor & Bogue, 2010).The majority of the services and events organized by the chaplains and volunteers are religious or spiritual in nature, however, the chaplains and volunteers are increasingly involved in facilitating services that take place in a humanist or secular context, without any religious or spiritual content, such as non-violent communication classes, social study groups, victim-offender dialogues, restorative justice programs, third level educational programs, parenting, and secular meditation. A recent survey of volunteers revealed that 47% primarily considered themselves to be “a religious volunteer”, 32% said they primarily considered themselves to be “a spiritual volunteer” and 22% said “a secular or humanist volunteer” (O’Connor & Duncan, 2011)

Sample

In some ways, the sample reflects the general population of Oregon, in that it is predominantly white (83%, n = 471) and Christian Protestant (58%, n = 331). In this regard, the percentage of inmates identifying as Protestant is like the general Oregon population (Pew, 2008), however, fewer inmates are religiously unaffiliated, with 14.6% (n = 83) of inmates responding as “no preference” or “don’t know” with regards to religious affiliation. This is compared to 27% of unaffiliated Oregonians in the general population. The sample also shows a moderate level of religious upbringing, with almost 28% (n = 160) of the sample claiming to be from a religious family. More details about the study sample are presented in Table 1.

Table 1

Descriptive Characteristics for Sample (N= 571)

Variable	Mean		Std. Dev.	Min	Max
Recidivism (Rearrest)	0.699		0.459	0.000	1.000
Religious and spiritual	0.607		0.489	0.000	1.000
Spiritual only	0.231		0.422	0.000	1.000
Religious only	0.088		0.283	0.000	1.000
Neither religious nor spiritual	0.075		0.263	0.000	1.000
Christian Protestant	0.581		0.494	0.000	1.000
Religious History	0.279		0.449	0.000	1.000
HSR Program Participation	2.365		3.471	0.000	28.916
Religious / Spiritual Coping
Cooperative Coping	3.051		1.084	1.000	5.000
Self-Coping	2.857		0.891	1.000	5.000
Religious Orientation
Intrinsic Score	33.301		6.693	9.000	45.000
Extrinsic Score	29.218		6.261	11.000	51.000
Assessed Risk	0.235		0.187	0.000	0.866
Minor Infractions	0.339		0.473	0.000	1.000
Major Infractions	0.638		0.480	0.000	1.000
Time Served (years)	3.054		2.489	1.002	7.803
Gang Member	0.116		0.321	0000	1.000
Sexual Offense	0.223		0.417	0.000	1.000
Male Gender (Female = 0, Male = 1)	0.847		0.359	0.000	1.000
White (Nonwhite=0, White=1)	0.831		0.374	0.000	1.000

Measures

Recidivism.There are numerous ways to measure recidivism, each with its own limitations and benefits. As an example, many studies of recidivism examine the likelihood of rearrest (Stansfield & Williams, 2014), which may include incidents for which a person is later found innocent or charges are later dropped. To address this issue, studies of recidivism (e.g. Hipp, Petersilia, & Turner, 2010; Mears, Wang, Hay, & Bales, 2008) have utilized reconviction as a more reliable and conservative estimate of post-release offending. Analyses were performed using both measures, with substantive results consistent across recidivism measure. In Oregon, however, parole violations are handled at the local county level and would not be counted in the DOC’s measure of recidivism. To ensure we are not undercounting reoffending, arrest is used as the dependent variable in results displayed (Goldstein, 2014). A little under 70% (n = 399) of our sample was rearrested for any new offense within 8 years after release from prison. Our dependent variable is a dichotomous measure where 0 indicates that the former inmate was not arrested for a new offense during our study period, and 1 indicates a person was arrested for a new offense.

Identification. Key independent variables were derived from a Spiritual Assessment administered to inmates in Oregon throughout 2004. The assessment first asked individuals to describe which of the following statements describes them best: “I am both spiritual and religious,” “I am spiritual but not religious,” “I am religious but spiritual,”, and “I am neither spiritual nor religious.”  Dummy variables for each category were created, with Spiritual and Religious as the reference category, and Spiritual Only, Religious Only, and Neither, as the displayed variables.

Religious History. We include two control variables capturing religious affiliation and religious history. Respondents were asked whether they grew up in a religious family setting. A dichotomous measure of religious family was created, where 1 indicated growing up in a religious family. Because our sample is predominantly Christian Protestant, we include a dichotomous variable to capture the effect of being Christian protestant (=1) on recidivism, in relation to all other religious affiliations (=0).

HSR Participation. We also captured the number of times an individual prisoner attended a humanist, spiritual, or religious program during their entire prison stay. While prior evaluation studies have captured program participation with a dichotomous measure, assessing the total attendances in HSR programs more accurately captures the ongoing commitment and extent to which an individual is exposed to program messages (Duwe & Clark, 2017). As the count of program attendances is highly skewed, we created a measure of the average monthly attendance at an HSR event throughout someone’s prison stay, by summing the total number of attendances and dividing by the number of months served during the prison sentence.

Religious Orientation. Our survey contained items comprising the age-universal Religious Orientation Scale (Gorsuch & Venable, 1983; Griffin, Gorsuch, & Davis, 1987) to measure two independent dimensions, intrinsic and extrinsic religious orientation. This includes 20 items scored on a 5-point Likert Scale, ranging from 1 (indicating strong disagreement) to 5 (indicating strong agreement). Nine of the items pertain to Intrinsic Orientation (alpha = .83), including items, “I enjoy reading about my religion,” “It is important to me to spend time in private thought and prayer,” and “I would rather join a Bible study group than a church social group.” Eleven items pertain to Extrinsic Orientation (alpha = .68) including, “I go to church because it helps me make friends,” “Although I am religious, I don’t let it affect my daily life,”, and “I go to church mainly because I enjoy seeing people I know there.” The score of each subscale is obtained by summing the scores of the 9 or 11 items, providing ranges of 11 to 55 for Extrinsic Orientation and 9 to 45 for Intrinsic Orientation.

Spiritual Coping. Respondents were also asked about the role that spirituality and religion play in the problem-solving and coping processes, using the short-form of the Religious Problem-Solving Scale (Pargament et al., 1988). Eighteen questions were asked of Oregon inmates comprising subscales designed to distinguish the degree of responsibility assigned to self or God in coping with problems. We created the Self-Directing (alpha = .91) and Collaborative (alpha = .93) subscales to measure religious coping styles. For each question, respondents were asked to indicate how often (from 1= never to 5 = always) a series of statements were true regarding several phases of the problem-solving process. Items indicative of individuals who act as co-partners with God included, “When I have a problem, I talk to God about it and together we decide what it means,” and “Together, God and I put my plans into action.” Items forming the self-directed subscale included, “When faced with trouble, I deal with my feelings without God’s help,” and “When deciding on a solution, I make a choice independent of God’s input.” These problem-solving styles have been found to be differentially associated with one’s level of religiousness, as well as mental and psychosocial health (Friedel & Pargament, 1995; Pargament et al., 1988; Schaefer & Gorsuch, 1993).

Recidivism Risk.The validated risk measure used to assess the risk of recidivism for an Oregon inmate is calculated automatically by the ODOC and is based on scores from 7 items: age, earned time, sentence length, prior revocation, number of prior incarcerations, prior theft convictions, and the current offense severity (Henning, Renauer, & Feyerherm, 2013). Based on these items, a percentage score between 0 and 1 is generated (ODOC, 2009). Assessed risk was included to determine whether the relation between religious and spiritual identification and recidivism was sustained even when the likelihood of persistent reoffending is considered, providing a more rigorous test of the impact of religion and spirituality.

We also control for several variables associated with recidivism risk, not captured by the assessed risk measure. Considering research that has indicated a relationship between types of prison misconduct and recidivism among adults (Caudill & Trulson, 2016; Cochran, Mears, Bales, & Stewart, 2014; Levitt & Loper, 2009), we capture whether a person had either a minor and / or major disciplinary infraction at any stage during their stay in the ODOC. The distinction between minor and major is determined by the ODOC’s misconduct severity level. Minor infractions may include disobedience of an order or being in an unauthorized area. Major infractions typically included violations for contraband, destruction of property and assault. Two dichotomous measures were created signaling whether a person had a minor (1 = yes) and a major (1 = yes) infraction. Additionally, we include a dichotomous measure indicating whether the prisoner was a known gang member. We also include the time served in prison (measured in years) because longer time incarcerated is correlated with lower risks of recidivism (Caudill & Trulson, 2016).

Perpetrator characteristics. We include other perpetrator characteristics shown to have implications for recidivism risk including, male gender (Female = 0, Male = 1), non-Hispanic white ethnicity (ethnic minority = 0; non-Hispanic White = 1), and, given the higher religious participation level among sexual offenders (O’Connor & Duncan, 2011), whether a person was convicted of a sexual offense. Approximately 22% of persons admitted through ODOC’s central admission in 2004 were sex offenders, a category of people known to have relatively lower rates of reoffending (Harris & Hanson, 2004; Soothill, 2010).

Analysis

The analysis proceeded in several steps. The initial step involved presenting descriptive statistics of all variables in the analysis (See Table 1), and exploring group differences in religious orientation, attendance at HSR programs, and religious / spiritual coping, to shed more light on how identification differences manifest themselves in prison. We also present a polychoric correlation matrix in an Appendix given that many of our key variables are dichotomous or ordinal.

To explore whether being spiritual only or neither religious or spiritual is associated with a higher likelihood of recidivism compared to religious and spiritual persons, we estimated a logistic regression model that includes dummy variables for religious and spiritual identification, perpetrator characteristics and assessed risk for recidivism. We next estimated a series of subsequent models introducing key explanatory variables (indicative of HSR attendance, religious orientation, and religious/ spiritual coping) one at a time to show the direct effect of these variables on recidivism, as well as observing changes in magnitude and significance of the identification coefficients. For each model, Odds Ratios (OR) are displayed to allow a relative comparison of the likelihood of reoffending between individuals who identify as spiritual only, religious only, and neither religious or spiritual, with the religious and spiritual group.

Finally, we sought to decompose direct and indirect effects of identification on rearrest. Although there is still debate, bootstrapping techniques have emerged as the leading method of assessing indirect effects (Preacher & Hayes, 2008). Bootstrapping is an intensive, non-parametric procedure that involves repeatedly resampling the data and estimating the indirect effect in each set of resampled data.We utilized the bootstrapping method of Preacher and Hayes (2008), in which this process of case resampling was repeated 5,000 times. All analyses were conducted using Stata 15

Results

 We began by exploring groups differences in religious history, religious orientation, spiritual coping, and participation in HSR programs in prison. Predictably, inmates who identified as both religious and spiritual had higher rates of participation in HSR programs during their first year in prison (M= 2.99, SD= 3.98), compared to spiritual only (M= 1.62, SD= 2.50), religious only (M= 1.48, SD= 2.05), and neither (M = 0.84, SD= 2.02). Intrinsic Orientation scores also varied between groups in the same pattern, with the highest scores for religious and spiritual persons, followed by spiritual only, religious only, and neither (M= 36.07, 30.64, 30.46. 23.81), (SD = 5.26, 6.25, 5.05, 7.28). Persons who identified as both religious and spiritual also had significantly higher Cooperative Coping scores (M= 3.46, SD= .92) and lower Self-Coping scores (M = 2.64, SD= .78) than all other groups. Individuals who were neither religious nor spiritual reported the highest scores on the scale of Self-Coping (M= 3.50, SD= 1.12).

An examination of polychoric correlations (see Appendix) also highlights that people assessed as higher risk are more likely to identify as spiritual only, whereas individuals assessed as lower risk are more likely to identify as both religious and spiritual.  Some prior work has suggested that higher risk persons may be less able to connect to and benefit from religion in prison (Stansfield, Mowen, & O’Connor, 2017). To test whether our measures of identification could explain variance in recidivism above and beyond assessed risk, risk and identification were entered into our models in a stepwise fashion, and an incremental f test confirmed that the effect of spiritual identification differed from zero (p = .01), increasing the R²value by .01.

To explore the presence of group differences in the likelihood of reoffending controlling for all covariates, we turn to Table 2. Beginning with our baseline model, model 1 displays the Odds Ratios of a logistic regression model of rearrest. Consistent with abundant existing literature, assessed risk, prisoner misconduct, and male gender were all positively and significantly associated with recidivism. A longer time served in prison was associated with lower odds of recidivism. Turning to our variables of interest, persons who identified as spiritual onlywere more than twice as likely to be rearrested than persons who identified as religious and spiritual(OR= 2.092, SE= .602, p = .009). Persons who were neither spiritual nor religious, however, did not have a higher likelihood of recidivism compared to religious-and-spiritual people in prison.

Table 2 Logistic Regression Estimating the Effect of Religious and Spiritual Identification on Recidivism

	Model 1		Model 2		Model 3		Model 4
variable	OR	SE	OR	SE	OR	SE	OR	SE
Religious & Spiritual	(ref)
Spiritual Only	2.092**	0.602	1.872*	0.505	1.971*	0.595	1.657	0.497
Religious Only	0.639	0.244	0.611	0.235	0.640	0.254	0.484	0.195
Neither	1.398	0.605	1.323	0.577	1.405	0.688	0.873	0.428
Christian Protestant	0.958	0.222	0.981	0.230	0.957	0.224	1.052	0.250
Religious History	0.828	0.208	0.844	0.213	0.828	0.210	0.785	0.199
Assessed Risk	1.033***	0.007	1.032***	0.007	1.033***	0.007	1.033***	0.007
Minor Infraction	1.766*	0.458	1.754*	0.455	1.766*	0.458	1.804*	0.470
Major Infraction	2.091**	0.487	2.080**	0.485	2.091**	0.487	2.100**	0.492
Time Served (Years)	0.794***	0.040	0.798***	0.040	0.794***	0.040	0.803***	0.041
Gang Member	1.388	0.575	1.336	0.555	1.388	0.575	1.348	0.558
Sexual Offense	0.944	0.279	0.978	0.293	0.944	0.280	0.943	0.281
Male	1.853*	0.488	1.811*	0.478	1.843*	0.486	1.820*	0.482
White	0.871	0.250	0.873	0.251	0.871	0.250	0.873	0.251
Monthly HSR in Prison			.936*	0.029
Coping Style (Cooperative)					1.003	0.128
Orientation (Intrinsic)							0.950**	0.018
Constant	2.915	1.708	2.945	1.726	2.885	2.185	4.139	4.322
Chi2	108.76		111.66		108.76		113.38
McFadden Pseudo R2	0.171		0.172		0.171		0.178

Notes. p < .05*, p < .01**,p < .001***; OR = Odds Ratio, SE = Standard Error. All Models also include dichotomous measures for participation in Job, Education, Cognitive, and Substance / Alcohol needs-based programs in prison. None were significantly related to recidivism.

Next, three intermediate models (Table 2, models 2 to 4) introduce measures of HSR attendance, spiritual coping and religious orientation. Model 2 reveals that a higher average monthly attendance at HSR services in prison was associated with lower odds of recidivism (OR= .936, SE= .029). Model 3 explores the relationship between spiritual coping and recidivism.  Self-Coping scores, indicating individuals who decide to cope with their problems on their own (without the help of religion, God, or the faith community) and Cooperative Coping scores are significantly and reverse correlated. Given that these two scales are seemingly capturing two ends of the same concept, in model 3 we just enter the Cooperative Coping score.

A higher Cooperative Coping score was not associated with recidivism. In Model 4, we assess the relation between an Intrinsic Orientation and recidivism. A higher Intrinsic Orientation score was significantly and negatively associated with recidivism (OR= .950, SE = .018). Furthermore, the addition of religious orientation significantly reduced the size of the spiritual only identification coefficient. This finding raises the possibility that religious orientation could explain observed differences in the odds of reoffending across identification groups. That is to say, spiritual only people had higher odds of reoffending partly because they have lower intrinsic religious orientations.

To further decompose the direct effect of identification on rearrest and indirect effect via HSR participation and religious orientation, supplementary tests were conducted in which we compared recidivism between religious and spiritual to spiritual only inmates. Bootstrap estimates suggested that indirect effects via HSR program participation (95% CI = [.005, .032]) and via religious orientation (95% CI = [.010, .045]) were both significant. This was also confirmed using Stata’s binary_mediation program (Ender, 2011) with bootstrapping to obtain standard errors and biased corrected 95% confidence intervals for the indirect effects and using the logistic decomposition function generated by Buis (2010). Specifically, using Stata’s ldecomp function to estimate a relationship between spiritual only identification and rearrest, the indirect effect via HSR program participation represented 19% of the total effect. The indirect effect via religious orientation represented 37% of the total effect.

Discussion

Religion in prison is an important responsivity factor (Worthington et al., 2011), with some estimates suggesting that over three-quarters of inmates will attend at least one humanist, spiritual, or religious event during their first year in prison (O’Connor & Duncan, 2011). While many studies have explored the implications for faith in prison and reoffending, few studies have considered differences by religious and spiritual identification. But much like the American public in general, almost a quarter of inmates identify as spiritual but deny that they are religious.  Building upon the work of Jang and Franzen (2013) and Giordano and colleagues (2008), the current study sought to examine whether there are higher odds of recidivism among persons who identified as spiritual only. Our results confirmed expectations, that people in prison claiming to be spiritual but not religious had higher odds of rearrest than persons claiming to be both spiritual and religious.

For those who identify in a religious or spiritual way, ensuring availability of services and allowing prisons to get chaplains and volunteers in, serves a constitutional right. Our study suggests that programs tailored to, and supportive of, the religious or spiritual ways one makes meaning may produce other long-term benefits (Worthington et al., 2011). The journey to desistance from crime also requires a great amount of individual willpower, determination, and social support. Ostensibly the work performed by chaplains and volunteers in Oregon meets this social support function. Our finding of a negative relationship between HSR attendance and rearrest suggests that the financial and institutional commitment made to HSR services may have some long-term benefit for people leaving Oregon’s prisons.

It is important to note that because of the scarcity of programming resources, many programs (such as cognitive, education, and substance abuse) are reserved for those with the highest risk to recidivate. Religious services, largely staffed by volunteers, are open to all inmates and so may include a disproportionate number of low risk individuals seeking guidance and support. Thus, for those who may not have received any other services, inmates can receive support from religious programs if they so choose. This may explain the significant negative association between HSR participation and recidivism. Furthermore, it may explain why individuals who identify as spiritual only tend to have poorer outcomes. Low risk ‘spiritual only’ inmates may similarly not receive any other services but lack the support that religious and spiritual inmates receive from religious programs, rendering them more isolated when dealing with problems.

The other major finding of the study is the negative relation between intrinsic religiosity and the likelihood of rearrest. While some commentators believe that religious belief and internal ways of making meaning are self-defeating  ways of thinking that could adversely affect mental health (Ellis, 1980), a growing body of psychological research has found that a stronger intrinsic religiosity provides individuals with a meaning and greater sense of purpose, in turn associated with with positive health outcomes including giving people strength (Graham, Furr, Flowers, & Burke, 2001), reduced anxiety (Shreve-Neiger & Edelstein, 2004), and mental health benefits generally (Hackney & Sanders, 2003; Hill & Pargament, 2008).

As such, correctional staff and programs must remain sensitive to the variety of ways individuals make meaning in their lives. This is related to the principle of specific responsivity in corrections, which is about adapting treatment to the individual. In a meta-analytic study of the benefits of adapting psychotherapy to a client’s religious or spiritual identification, Worthington and colleagues (2011) categorized identification into humanist spirituality, cosmic spirituality, nature spirituality, and religious spirituality. They concluded that however one defines their spirituality, psychologists and counselors should integrate each client’s particular way of establishing meaning into treatment. Data suggest that patients in specific religious or spiritual psychotherapies showed greater improvement in both psychological and spiritual outcomes (Norcross et al., 2010). Thus, for individuals who are spiritual but eschew faith-based programs or religious events, incorporating spirituality into other aspects of their treatment or daily life may bolster support.

There are some limitations to this study. Firstly, it is important to note that our findings are influenced by the makeup of the inmates in Oregon’s prison system. Our sample is dominated by males and individuals identifying as Protestant/Christian. Models estimated separately for males, for example, are almost identical to the results displayed here. Like many prior studies of religion and reoffending, however, we were unable to learn much from female-specific models because of reduced statistical power and less variation to explain (O’Connor, 2004). There are also other correlates of recidivism that we were not able to measure, including prison visitation (Bales & Mears, 2008), and post-incarceration changes in marriage and family support (generally Bersani & Doherty, 2013), peer support (Mowen & Boman, 2018) and employment (Uggen, 1999). Although these factors are not universally associated with positive reentry outceoms (e.g., Pleggenkuhle, Huebner, & Summers, 2018; Skardhamr & Savolainen, 2014), these variables would be important to measure in replications of this study. Although our study is unique in its ability to assess the long-term implications of group differences in religious and spiritual identification, the results do need to be replicated accounting for long term changes over time in these other important constructs. Without control for other structural changes, we cannot truly isolate the process of “religious conversion” in prison and in the community after release. As the use of detailed assessments of religion and spirituality have been slow coming in criminology (Jang & Franzen, 2013), studies linking religion, crime, and recidivism should consider adopting multi-wave surveys of spirituality and religiosity.

Finally, we were also unable to tease out the specific form of social support offered by HSR programs, and religion broadly, for persons who identify as both religious and spiritual. As detailed by Cullen (1994) and other scholars since (e.g., Taylor, 2016), support can take many forms including affectionate support and instrumental assistance. As noted by Stansfield and colleagues (2017), persons who attach themselves to religion in prison and the community are embedding themselves into a community of prosocial people already woven in to the fabric of society where they can act as a bridge to employers, housing, education and a variety of other prosocial networks. Future research may benefit from exploring the specific forms of social support provided to various groups in prison through religion and spirituality.

Conclusion

Criminology has been slow to examine the relationship between self-identification in terms of both spirituality and religiousness (Jang & Franzen, 2013), and this is the first study to use detailed data to be able to demonstrate long term implications of identification for reoffending. Furthermore, by using data from a large survey specifically designed to assess religious orientation and spiritual coping among offenders in prison, we could understand group differences in reoffending more fully. This enabled us to uncover not just the importance of attending faith-based programs in prison, for those who identify in religious or spiritual ways, but specifically the importance of facilitating an intrinsic religious orientation and participation in services to deal with the problems people in prison face. A 2012 survey of prison chaplains from all 50 states called HSR programs and the faith community critical for reentry success. We echo this conclusion, with the clarification that for individuals who make meaning in a variety of religious or spiritual ways, emotional coping is a critical part of adaption to life in prison, and upon release to the community (Schroder & Frana, 2009). Ensuring an adequate availability of programs and volunteers from the full array of traditions and ways people make meaning can provide crucial venues for the development of this support. And given that many of these programs are staffed by volunteers, such programs may be among the most cost-effective programs prisons can supply.

References

Adamczyk, A., Freilich, J. D., & Kim, C. (2017). Religion and crime: A systematic review and assessment of next steps. Sociology of Religion, 78(2), 192-232.

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

Ammerman, N. T. (2013). Spiritual but not religious? Beyond binary choices in the study of religion. Journal for the Scientific Study of Religion, 52(2), 258-278.

Bakken, N. W., DeCamp, W., & Visher, C. (2014). Spirituality and desistance from substance abuse among reentering offenders. International Journal of Offender Therapy and Comparative Criminology, 58, 1321-1339. Doi:10.1177/0306624X13494076

Bales, W. D., & Mears, D. P. (2008). Inmate social ties and the transition to society: Does visitation reduce recidivism? Journal of Research in Crime and Delinquency, 45(3), 287-321.

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.

Bersani, B. E., & Doherty, E. E. (2013). When the ties that bind unwind: Examining the enduring and situational processes of change behind the marriage effect. Criminology, 51(2), 399-433.

Boman, J. H., & Mowen, T. J. (2017). Building the ties that bind, breaking the ties that don’t. Criminology & Public Policy, 16(3), 753-774.

Buis, M. L. (2008). LDECOMP: Stata module decomposing effects in logistic regression into direct and indirect effects. The Stata Journal, 9, 1-16.

Caudill, J. W., & Trulson, C. R. (2016). The hazards of premature release: Recidivism outcomes of blended-sentenced juvenile homicide offenders. Journal of Criminal Justice, 46, 219-227.

Chu, D. C. (2007). Religiosity and desistance from drug use. Criminal Justice and Behavior, 34(5), 661-679.

Clear, T. R., Stout, B. D., Dammer, H., Kelly, L., Hardyman, P., & Shapiro, C. (1992). Prisoners, prisons, and religion: Final report. New Jersey: Newark: School of Criminal Justice, Rutgers University.

Cochran, J. C. (2014). Breaches in the wall: Imprisonment, social support, and recidivism. Journal of Research in Crime & Delinquency, 51, 200-229.

Cochran, J., Mears, D. P., Bales, W. D., & Stewart, E. A. (2014). Does inmate behavior affect post-release offending? Investigating the misconduct-recidivism relationship among youth and adults. Justice Quarterly, 31(6), 1044-1073.

Colvin, M., Cullen, F.T., & Vander Ven, T.. (2002). Coercion, social support, and crime: An emerging theoretical consensus. Criminology,40, 19-42.

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.

Dodson, K. D., Cabage, L. N., & Klenowski, P. M. (2011). An evidence-based assessment of faith-based programs: Do faith-based programs “work” to reduce recidivism?. Journal of Offender Rehabilitation, 50(6), 367-383.

Duwe, G., & Clark, V. (2017). The rehabilitative ideal versus the criminogenic reality: The consequences of warehousing prisoners. Corrections: Policy, Practice, and Research, 2, 41-69.

Duwe, G., & Johnson, B. R. (2016). The effects of prison visits from community volunteers on offender recidivism. The Prison Journal, 96, 279-303.

Duwe, G., & King, M. (2013). Can faith-based correctional programs work? An outcome evaluation of the InnerChange Freedom Initiative in Minnesota. International Journal of Offender Therapy and Comparative Criminology, 57(7), 813-841.

Dye, M. H., Aday, R. H., Farney, L., & Raley, J. (2014). “The rock I cling to”: Religious engagement in the lives of life-sentenced women. The Prison Journal,94, 388-408.

Emmons, R. A. (2005). Striving for the sacred. Personal goals, life meaning, and religion. Journal of Social Issues, 61, 731-745.

Ender P. D. (2011). Binary_mediation: Indirect effects with binary dv and/or mv. Retrieved from http://www.ats.ucla.edu/stat/stata/faq/binary_mediation.htm

Fuller, R. C. (2001). Spiritual, but not religious: Understanding unchurched America. New York, NY: Oxford University Press.

Giordano, P., Longmore, M., Schroeder, R., & Seffin, P. (2008). A life-course perspective on spirituality and desistance from crime. Criminology,46, 99-132.

Goldstein, D. (2014). The misleading math of recidivism. New York, NY: The Marshall Project

Gorsuch, R. L., & Venable, G. D. (1983). Development of an age universal I-E scale. Journal for the Scientific Study of Religion, 22, 181-187.

Griffin, G. A. E., Gorsuch, R. L., & Davis, A. (1987). A cross-cultural investigation of religious orientation, social norms, and prejudice. Journal for the Scientific Study of Religion, 26, 358-365.

Hackney, C. H., & Sanders, G. S. (2003). Religiosity and mental health: A meta-analysis of recent studies. Journal for the Scientific Study of Religion, 42, 43-55.

Hallett, M. A., Hays, J., Johnson, B. R., Jang, S. J., & Duwe, G. (2016). The Angola prison seminary: Effects of faith-based ministry on identity transformation, desistance, and rehabilitation. New York: Routledge.

Hallett, M., & McCoy, J. S. (2015). Religiously motivated desistance: An exploratory study. International Journal of Offender Therapy and Comparative Criminology, 59(8), 855-872.

Henning, K., Renauer, B., & Feyerherm, B. (2013). Risk assessment and the Public Safety Checklist. Portland, OR: Portland State University.

Hill, P. C., & Pargament, K. I. (2008). Advances in the conceptualization and measurement of religion and spirituality: Implications for physical and mental health research. Psychology of Religion and Spirituality, S(1), 3-17. Doi:10.1037/1941-1022.S.1.3

Hill, P. C., Pargament, K. I., Hood, R. W., McCullough, J. M. E., Swyers, J. P., Larson, D. B., & Zinnbauer, B. J. (2000). Conceptualizing religion and spirituality: Points of commonality, points of departure. Journal for the Theory of Social Behaviour, 30(1), 51-77.

Hipp, J. R., Petersilia, J., & Turner, S. (2010). Parolee recidivism in California: The effect of neighborhood context and social service agency characteristics. Criminology, 48(4), 947-979.

Hirschi, T. (1969).The Causes of Delinquency.Berkeley, CA: University of California Press

Jang, S. J., & Franzen (2013). Is being spiritual enough without being religious? A study of violent and property crimes among emerging adults. Criminology, 51(3), 595-623.

Jang, S. J., Johnson, B. R., Hays, J., Hallett, M., & Duwe, G. (2018). Religion and misconduct in “Angola” prison: Conversion, congregational participation, religiosity, and self-identities. Justice Quarterly, 35, 412-442. Doi:10.1080/07418825.2017.1309057

Johnson, B. (2004). Religious programs and recidivism among former inmates in prison fellowship programs: A long-term follow-up study. Justice Quarterly, 21(2), 329-354.

Johnson, B. (2011). More god, less crime: Why faith matters and how it could matter more. Conshohocken: Templeton Foundation Press.

Kelly, P. E., Polanin, J. R., Jang, S. J., & Johnson, B. R. (2015). Religion, delinquency, and drug use: A meta-analysis. Criminal Justice Review, 40(4), 505-523.

Kenny, D. A. (2008). Mediation with dichotomous outcomes. Retrieved April 23, 2010 from website: http://davidakenny.net/doc/dichmed.pdf .

King, M., Marston, L., McManus, S., Brugha, T., Meltzer, H., & Bebbington, P. (2013). Religion, spirituality and mental health: Results from a national study of English households. British Journal of Psychiatry, 202, 68-73.

Koenig, H. G. (2009). Research on religion, spirituality, and mental health: A review. The Canadian Journal of Psychiatry, 54(5), 283-291.

Kosmin, B. A., and Keysar, A (2009). American religious identification survey (ARIS 2008): Summary report. Hartford, CT, USA: Trinity College.

Levitt, L., & Loper, A. B. (2009). The influence of religious participation on the adjustment of female inmates. American Journal of Orthopsychiatry, 79, 1-7.

Listwan, S. J., Colvin, M., Hanley, D., & Flannery, D. (2010). Victimization, social support, and psychological well-being. A study of recently released prisoners. Criminal Justice & Behavior, 37, 1140-1159.

Maruna, S., Wilson, L., & Curran, K. (2006). Why God is often found behind bars: Prison conversions and the crisis of self-narrative. Research in Human Development, 3(2-3), 161-184.

Mears, D. P., Cochran, J. C., Siennick, S. E., & Bales, W. D. (2012). Prison visitation and recidivism. Justice Quarterly, 29, 888-918.

Mercadante, L. A. (2014). Belief without borders: Inside the minds of the spiritual but not religious. London: Oxford University Press.

Mowen, T. J., & Boman IV, J. H. (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, 1094-1116.

Norcross, J. C., & Wampold, B. E. (2011). What works for whom: Tailoring psychotherapy to the person. Journal of Clinical Psychology, 67(2), 127-132.

O’ Connor, T., & Bogue, B. (2010). Collaborating with the community, trained volunteers and faith traditions. In F. McNeill, P. Reyner, & C. Trotter (Eds.), Offender supervision: New directions in theory, research, and practice (pp. 301-322). Milton, UK: Willan Publishing.

O’Connor, T., & Duncan, J. (2011). The sociology of humanist, spiritual and religious practice in prison: Supporting responsivity and desistance from crime. Religions,2(4), 590-610.

O’Connor, T. P., & Perreyclear, M. (2002). Prison religion in action and its influence on offender rehabilitation. Journal of Offender Rehabilitation, 35(3-4), 11-33.

Oregon Department of Corrections. (2009). Automated Criminal Risk Score (ACRS). Issue Brief, 56. Salem, OR: Oregon Department of Corrections.

Pargament, K. I. (1999). The psychology of religion and spirituality? Yes and no. The International Journal for the Psychology of Religion, 9(1), 3-16.

Pargament, K. I., Kennell, J., Hathaway, W., Grevengoed, N., Newman, J., & Jones, W. (1988). Religion and the problem-solving process: Three styles of coping. Journal for the Scientific Study of Religion, 27, 90-104.

Pew Forum on Religion and Public Life. (2008). U.S. religious landscape survey: Diverse and dynamic. Washington, D.C.: Pew Research Center.

Pew Research Center (2012). Religion in prisons – A 50-state survey of prison chaplains. Washington, D. C.: Pew Trust.

Pirutinsky, S. (2014). Does religiousness increase self-control and reduce criminal behavior? A longitudinal analysis of adolescent offenders. Criminal Justice and Behavior, 41(11), 1290-1307.

Pleggenkuhle, B., Huebner, B. M., & Summers, M. (2018). Opting out: The role of identity, capital, and agency in prison visitation. Justice Quarterly, 35(4), 726-749.

Power, J., Ritchie, M., & Madill, D. (2014). Faith-based correctional interventions: a systematic review of the literature. Ottawa, CA: Correctional Service Canada.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, 36(4), 717-731.

Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

Saucier, G., & Skrzypińska, K. (2006). Spiritual but not religious? Evidence for two independent dispositions. Journal of Personality,74(5), 1257-1292.

Schaefer, C. A., & Gorsuch, R. L. (1993). Situational and personal variations in religious coping. Journal for the Scientific Study of Religion, 32, 136-147.

Schroeder, R. D., & Frana, J. F. (2009). Spirituality and religion, emotional coping, and criminal desistance: A qualitative study of men undergoing change. Sociological Spectrum, 29(6), 718-741.

Shreve-Neiger, A. K., & Edelstein, B. A. (2004). Religion and anxiety: A critical review of the literature. Clinical Psychology Review, 24, 379-397.

Skardhamar, T., & Savolainen, J. (2014). Changes in criminal offending around the time of job entry: A study of employment and desistance. Criminology, 52(2), 263-291.

Soothill, K. (2010). Sex offender recidivism. Crime & Justice, 39(1), 145-211.

Speck, P., Higginson, I., & Addington-Hall, J. (2004). Spiritual needs in health care: May be distinct from religious ones and are integral to palliative care. BMJ: British Medical Journal, 329(7458), 123-124.

Stansfield, R. (2017). Drawing on Religion in the Desistance Process: Paying Attention to Race and Ethnicity. Criminal Justice and Behavior, 44(7), 927-945.

Stansfield, R., Mowen, T. J., O’Connor, T., & Boman, J. H. (2017). The role of religious support in reentry: Evidence from the SVORI data. Journal of Research in Crime and Delinquency, 54(1), 111-145.

Stansfield, R., & Williams, K. R. (2014). Predicting family violence recidivism using the DVSI-R: Integrating survival analysis and perpetrator characteristics. Criminal Justice and Behavior, 41(2), 163-180.

Storey, J. E., Kropp, P. R., Hart, S. D., Belfrage, H., & Strand, S. (2014). Assessment and management of risk for intimate partner violence by police officers using the brief spousal assault form for the evaluation of risk. Criminal Justice and Behavior, 41(2), 256-271.

Taylor, C. (2007). A secular age. Cambridge, MA: Harvard University Press.

Taylor, C. J. (2016). The family’s role in the reintegration of formerly incarcerated individuals: The direct effects of emotional support. The Prison Journal,96, 331-354.

Uggen, C. (1999). Ex-offenders and the conformist alternative: A job quality model of work and crime: Social Problems, 46, 127-151.

Ulmer, J. T., Desmond, S. A., Jang, S. J., & Johnson, B. R. (2012). Religious involvement and dynamics of marijuana use: Initiation, persistence, and desistence. Deviant Behavior, 33(6), 448-468.

Visher, C. A., & Travis, J. (2003). Transitions from prison to community: Understanding individual pathways. Annual Review of Sociology, 29(1), 89-113.

Wong, Y. L. R., & Vinsky, J. (2009). Speaking from the margins: A critical reflection on the ‘spiritual-but-not-religious’ discourse in social work. British Journal of Social Work, 39, 1343-1359.

Worthington, E. L., Hook, J. N., Davis, D. E., & McDaniel, M. A. (2011). Religion and spirituality. Journal of Clinical Psychology, 67, 204-214.

Young, M. C., Gartner, J., O’Connor, T., Larson, D., & Wright, K. N. (1995). Long-term recidivism among federal inmates trained as volunteer prison ministers. Journal of Offender Rehabilitation, 22(1-2), 97-118.

Zinnbauer, B. J., Pargament, K. I., Cole, B., Rye, M. S., Butter, E. M., Belavich, T. G., & Kadar, J. L. (1997). Religion and spirituality: Unfuzzying the fuzzy. Journal for the Scientific Study of Religion, 36, 549-564.

Appendix

Polychoric Correlation Matrix of Variables Used in the Study

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1. Rearrest	1.00
2. Relig & Spiritual	-0.17	1.00
3. Spiritual Only	0.19	-1.00	1.00
4. Assessed Risk	0.39	-0.17	0.15	1.00
5. Gang	0.20	-0.11	0.09	0.30	1.00
6. Minor Infractions	0.12	0.06	-0.17	0.10	0.29	1.00
7. Major Infractions	0.24	-0.04	-0.05	0.23	0.65	0.39	1.00
8. Time Served	-0.31	0.09	-0.12	-0.16	0.18	0.23	0.20	1.00
9. Sexual Offense	-0.28	0.03	-0.07	-0.39	-0.39	-0.09	-0.08	0.37	1.00
10. Male	-0.11	0.10	-0.01	-0.14	-0.46	-0.04	-0.26	-0.19	-0.59	1.00
11. White	0.05	-0.12	0.04	0.04	-0.27	-0.04	-0.11	-0.07	-0.02	0.14	1.00
12. Religious History	-0.02	-0.28	0.23	0.05	-0.03	-0.01	0.02	0.00	-0.03	0.16	0.28	1.00
13. Christ Prot	-0.03	0.41	-0.36	-0.08	-0.08	0.01	-0.04	0.00	0.09	0.17	0.01	-0.08	1.00
14. HSR Attendance	-0.21	0.35	-0.22	-0.16	-0.45	-0.02	-0.10	0.13	0.18	0.13	0.04	0.04	0.26	1.00
15. Cooperative	-0.12	0.58	-0.30	-0.13	-0.06	0.06	0.00	0.14	0.13	0.06	-0.20	-0.27	0.26	0.30	1.00
16. Intrinsic	-0.17	0.59	-0.28	-0.13	-0.11	0.05	-0.03	0.12	0.10	0.10	-0.11	-0.19	0.35	0.36	0.69	1.00