Psychosocial pathways to recidivism-risk among offenders in correctional centres in Eswatini : a mediation study

Abstract

Background: Recidivism risk has become one of the most serious social problems that have gained significant attention from social and behavioural scientists. Relapse into criminal behaviour among released inmates is a significant concern in developing countries due to lack of knowledge on how to prevent such behaviour. This study focuses on the influence of personality and social support on recidivism risk among offenders. In addition, it investigates the mediating roles of mental health, forgiveness, spirituality and religiosity on the associations of personality traits and social support with recidivism risk. Methodology: This study used a cross-sectional research design. To select the five correctional centres the MULTI-Staged sampling technique was used and purposive sampling technique was used to select 244 offenders {192 (78.69%) males; 52 (21.31%) females; mean age= 31.61, SD=8.27} in five correctional centres. Structured instruments were used to collect data, comprising demographic variables, Eysenck Personality Questionnaire, Berlin Social Support Scales, Forgiveness Scale, Spirituality Well-Being Scale-revised, 10-Item Hoge Intrinsic Religiosity scale, General Health Questionnaire and Brief Assessment for Recidivism-risk. Six hypotheses were tested using structural equation modelling (SEM). Results: Among personality factors, only neuroticism (β = -.13, p = .02) significantly predicted recidivism risk, while extraversion (β = .05, p = .33) and psychoticism (β = -.06, p = .27) did not. Instrumental support (β = .20, p = .03) significantly predicted recidivism risk while emotional support (β = .02, p = .85), need support (β = -.07, p = .33) and support seeking (β = -.05, p = .47) did not. The direct influence of neuroticism (β = -.10, p = .23), instrumental support (β = .11, p = .08) and mental health (β = -.08, p = .35) were not significant on recidivism risk. An increase in neuroticism predicted an increase in poor mental health while an increase in instrumental support predicted a decrease in poor mental health. Outcomes of mediation analysis indicate that mental health, forgiveness, spirituality and religiosity did not mediate the association of instrumental support and neuroticism with recidivism risk. All models met the acceptable criteria for model fit. Results of mediation analysis suggest that standardized total effect of neuroticism on recidivism risk was significant (β = -.14, p = .026). The vii standardized total effect of instrumental support on recidivism risk was not significant (β = .12, p = .067). The fit statistics met the acceptable criteria for model fit, χ2 (1) = 2.74, p = .10; CFI = .94; RMSEA = .08 [90% CI = (.00, .21)], SRMR = .03. Conclusion: It is established, that neuroticism and instrumental support are significantly associated with recidivism risk among offenders. However, mental health, forgiveness, spirituality and religiosity did not have mediating effects on these associations. It is important that psychologists, social workers and chaplains help the offenders in therapy to reduce the recidivism risk among offenders by considering offenders' personality and experience of social support.