The Effect of Religion on Emotional Well-Being Among Offenders in Correctional Centers of South Africa: Explanations and Gender Differences

Abstract

We examined (1) whether the relationship between religiosity and negative emotions (anger, frustration, depression, and anxiety) among prisoners is attributable to inmates’ sense of meaning and purpose in life and personal virtues and (2) whether religiosity has a larger positive relationship with a search for and a presence of meaning in life as well as the virtues of forgiveness, gratitude, and self-control among female than male inmates. To examine these relationships, we analyzed survey data from a sample of offenders in South African correctional centers. Findings showed that more religious inmates reported lower levels of negative emotions to the extent that their religiosity enhanced a sense of meaning and purpose in life and levels of self-control than their less or non-religious peers. We also found the salutary effect of religiosity to be applicable equally to male and female inmates. Substantive and practical implications of our findings are discussed.

Keywords:

• Religion
• meaning and purpose in life
• virtue
• mental health
• prison

Acknowledgement

The authors are grateful to the editor and three anonymous reviewers for helpful comments on earlier drafts of the paper. They are also greatly indebted to the South Africa Department of Correctional Services and the following individuals including volunteers (listed in no order) for data collection: Dr. Menzi Mkhathini, Chaplain Fortein, Chaplain Rev. Oscar Madlala, Chaplain Rev. Nkuna, Chaplain Mamiki, Mrs. Mohapi, Ms. Dee Lombard, Ms. Porsha Moore, Ms. Cornelia (Connie) Wehrmann, Mr. Bryan Kraynauw, Rev. Andrew Walker, and correctional staff members and officers whose name could not be all mentioned here.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We use the terms “offenders” and “correctional centers” (which is spelled centres in South Africa, though we use American English spelling in this paper) instead of “prisoners” and “prisons” when they are related to South Africa as required by the South Africa Department of Correctional Services (Department of Correctional Services, 2005), which approved data collection for the present study.

2 In this paper the terms “mental health” and “emotional well-being” are not used interchangeably as the former concept is broader in scope than the latter, which we focus on in our study described below.

3 The positive association’s reverse causality (i.e., self-control affecting religiosity) has been proposed (Ellis, 1987), while previous studies providing empirical support for the proposition were based on cross-sectional data (Cochran, Wood, & Arneklev, 1994; Kerley, Copes, Tewksbury, & Dabney, 2011). However, testing for causal direction requires panel data, and one longitudinal study found religion to predict self-control but failed to find evidence for the reverse causality (Pirutinsky, 2014). More importantly, the question of causal direction is primarily a theoretical issue, and we believe religiosity is more likely to increase self-control than the other way around because there are more theoretical reasons for the former (Smith, 2003b) than the latter (Ellis, 1987).

4 Although the official data included other information, such as the inmate’s criminal history and prison records, we could not use them because they were plagued by too many missing cases to be included in analysis.

5 This is plausible given that an unusually high percentage (97.2%) of the survey participants said they had religion.

6 The four items of private religiosity were loaded on a single factor and had high loadings (ranging from .598 to .723) and good inter-item reliability (α = .736).

7 An additional analysis, where we kept the four negative emotional states separate in the model, revealed that the unexpected relationship applied only to anxiety (see Supplemental Table 4), though it is difficult to explain why being grateful was positively associated with anxiety among female inmates. However, the overall results showed gender similarities more than gender dissimilarities. Furthermore, when we estimated the public vs. private religiosity model separately for males and females, private religiosity was still more likely to have existential and virtuous effects and significant mediation than public religiosity, whether we used the omnibus or separate measures of negative emotional states (see Supplemental Tables 5 and 6). We also found no significant interactions involving public and private religiosity in the total sample (see Supplementary material Tables 7 and 8), whereas a few exceptions were observed in the gender subsamples (see Supplementary material, Tables 9 and 10).

8 Following an anonymous reviewer’s suggestion, we explored whether the relationships between religiosity and negative emotional states were spurious due to a third variable, self-control. Zero-order correlation between religiosity and negative emotional states was significant in the female sample (r = ‒.135) but not in the total (r = ‒.038) and male samples (r = .044). Partial correlation, controlling for self-control, was found not significant in the female (r = ‒.106) as well as total sample (r = .053) and significant but positive in the male sample (r = .209). First, the correlation becoming non-significant when self-control was held constant in the female sample implies its spuriousness as well as indirect relationship between religiosity and negative emotions via self-control. While we examined the indirect relationship where religiosity affects self-control (see footnote 3), the spuriousness suggests the obverse (Cochran et al., 1994; Ellis, 1987). Thus, the relationship between religiosity and self-control is likely to be reciprocal although we could not examine it as such due to our analysis of cross-sectional data. Second, the non-significant zero-order correlation between religiosity and negative emotional states in the total and male samples revealed that whether religion enhanced emotional well-being or not depended on the extent to which inmates’ religious involvement and practice provided them with a sense of meaning and purpose in life and promoted their personal virtues as our multivariate analysis showed rather than whether they simply had religious affiliation or even engaged in religious activities and practices, felt close to God, and believed religion to be important to them.

Additional information

Funding

Funding for this research was supported, in part, by Prison Fellowship International (PFI). Opinions expressed in this publication are those of the authors and do not necessarily reflect the views of PFI.

Notes on contributors

Sung Joon Jang

Sung Joon Jang is Research Professor of Criminology and Co-director of Program on Prosocial Behavior at the Institute for Studies of Religion at Baylor University. His research focuses on the effects of strain and religion on criminal offending and desistance and has appeared in various journals of criminology and criminal justice. Jang is currently conducting a series of studies on the rehabilitative effects of faith-based programs on prisoners in Colombia and South Africa as well as in the United States.

Byron R. Johnson

Byron R. Johnson is Distinguished Professor of the Social Sciences and Founding Director of Baylor's Institute for Studies of Religion. He is recognized as a leading authority on the scientific study of religion, the efficacy of faith-based organizations, and criminal justice. He is currently working with the Gallup Organization to design empirical studies exploring religion and spirituality in the world.

Matthew L. Anderson

Matthew Lee Anderson is a Postdoctoral Research Fellow at the Institute for Studies of Religion at Baylor University. His research is primarily in ethics and focuses on reproductive ethics and the ethics of incarceration. He is currently working on a project on the effects of incarceration on family members of offenders, and the state's duties to resist or mitigate the spread of secondary stigmas.

Karen Booyens

Karen Booyens is a senior lecturer in the Department of Social Work and Criminology, University of Pretoria. She has published numerous articles in peer reviewed journals as well as chapters in books and has participated in national and international conferences. She presents life skills programmes to male sentenced offenders and developed a programme to specifically address sexual assault and rape in male correctional centres. Her current research interest is the elderly offender in South African corrections.