The relationship between family dynamics, personality and mental health of young offenders in North West province, South Africa

Abstract

The aim of the study is to compare family dynamics, personality and mental health of young offenders in correctional institutions with a control group of non-offenders in the North-West Province, South Africa. The study was anchored on three hypotheses and thereby (1) compared offenders and non-offenders on family dynamics (parenting styles, emotional distance) and personality (2) compared mental health reports of offenders and non- offenders and (3) determined whether parenting styles, emotional distance and personality would predict mental health. The study used a questionnaire with four sections- A, B, C and D. Section A contained demographic items, Section B contained family dynamics that constituted parenting style scale and emotional distance, Section C contained personality measuring, using the EPQ with three subscales, and section D contained the General Health Questionnaire 28 scale used to measure mental health with four subscales- somatic complaints, anxiety and insomnia, social dysfunction and depression. Psychometric properties of all the scales used are valid and reliable. One hundred and forty seven (14 7) participants were randomly selected using a table of random numbers of 'yes' or 'no' from two (2) youth care centres and a secondary school in the North-West Province, South Africa. Non-offenders were (male= 33, female= 55) and offenders were (male= 52, female= 7). Age of participants ranged from 12-18 years, with mean age (X-bar= 15.9 years) (SD= 1.8). The first and second hypotheses were tested with a t- test and the third hypothesis with a hierarchical multiple regression. Results for hypothesis one and two showed that there was a strong significance for (GHQ Anxiety), t (145) = -1.22, p< .004 with non-offenders scoring higher than offenders (X-bar 13.0 versus X-bar 12.25). GHQT, t (145) = -1.23, p< .003 with offenders reporting poorer mental health than non-offenders (X-bar 46.16 versus X-bar 44.46). Flexible, t (145) = .93, p< .002 with offenders scoring higher than non-offenders (X-bar 30.62 versus X-bar 29.77). GHQ-Depression, t (145) = 3.88, p< .001 with offenders scoring higher than non-offenders (X-bar 10.83 versus X-bar 7.77). Emotional distance from mother, t (145) = -4.62, p< .000 with non-offenders scoring higher than offenders (X-bar 6.70 versus X-bar 5.25). However, the third hypothesis expected that family dynamics and personality would predict poor mental health. The results for hypothesis three showed that permissive did not independently predict poor mental health (β= .076). While extraversion β= -.212) and neuroticism (β= -. 167) independently and significantly predicted mental health. The variables (permissive, extraversion and neuroticism) explained 10% of the total variance on poor mental health. The three variables jointly influenced and predicted mental health in the model, (R= .31, R2= .10, P< .05). In addition, the Durbin-Watson result (2 .003) showed that the assumption of independent error is met for this model. It was noted in conclusion that the study contributed to the body of knowledge by showing that there is a relationship between family dynamics, personality and mental health. It was also noted that family dynamics and personality can better predict poor mental health on offenders than the general population. A cognitively based intervention must be developed and offered to young offenders with a variety of mental health issues in different correctional homes. Other preventive recommendations were made in line with the findings of the study.