Work stress has a substantial impact on employees, organisations and economies; especially in the fragile economic environment since the ‘Great Recession’ of 2008; which has seen employment levels drop and employees willing to endure more stress at work to avoid retrenchment. These impacts include serious health and financial consequences. Attempts should therefore be made to effectively manage and address work stress to lessen these dire consequences. Many models have been developed and theorised to assist in explaining work stress, the pinnacle of these being the job demands-resources (JD-R) model. In JD-R theory, the dual process explains that work-related well-being follows the following processes: An energetic, also called the health impairment process, in which job demands leads to ill health outcomes through burnout; and then a motivational process which presents that job resources leads to positive organisational outcomes, e.g. organisational commitment, through engagement. The main objectives of this research were 1) to investigate a JD-R model in a large South African sample with a categorical estimator; 2) to investigate the reversed causal hypotheses of burnout and engagement in job demands-resources theory over time; 3) to investigate the likelihood of reporting treatment for health conditions based on burnout and engagement, and 4) to investigate the link between burnout and objective financial outcomes, i.e. by medical aid provider expenditure. To achieve the first objective a cross-sectional design was used (n = 15 633) covering numerous sectors in South Africa. A dual process model was specified with job demands (work overload) leading to ill health through burnout, and job resources (colleague and supervisor support, communication, growth opportunities and role clarity) leading to organisational commitment through engagement. Results of structural equation modelling indicated that the proposed JD-R model was a good fit to the sample. Furthermore, burnout was found to mediate the relationship between job demands and ill health with a medium effect. Engagement was found to mediate the relationship between job resources and organisational commitment with a large effect. The second objective, concerning reversed causality, was achieved with a longitudinal design (n = 593). The hypothesized model included burnout and engagement at time one, and at time two work overload as indicator of job demands, and colleague and supervisor support, communication, growth opportunities and role clarity as indicators of job resources. Results indicated that burnout had a significant negative reversed causal effect to supervisor support and colleague support. Engagement showed only one significant result, i.e. a small negative reversed causal relationship with supervisor support. To achieve the third objective, a cross-sectional design was used (n = 7 895). Results for logistic regression analyses showed that an increase in burnout was associated with a significant increase in the estimated odds for reporting an affirmative answer for receiving treatment for any of the health conditions, i.e. cardiovascular conditions, cholesterol, depression, diabetes, hypertension and irritable bowel syndrome. In contrast, an increase in engagement was associated with a decrease in affirmative reporting for cardiovascular conditions, cholesterol and depression; but not for diabetes, hypertension or irritable bowel syndrome. Addressing the link between burnout and financial outcomes was the fourth objective; and met with a cross-sectional design (n = 3 182). Participants were divided into a high and low burnout group based on the comorbidity of exhaustion and cynicism Analysis of covariance (ANCOVA) was implemented, controlling for age and gender, to investigate the difference in medical aid provider expenditure of the two groups. Results revealed that expenditure in the high burnout group was consistently more in all cases, compared to the low burnout group. By way of conclusion, the implications of the research were discussed and recommendations for managers and for future research were made. / Thesis (PhD (Industrial Psychology))--North-West University, Potchefstroom Campus, 2013

Work stress has a substantial impact on employees, organisations and economies; especially in the fragile economic environment since the ‘Great Recession’ of 2008; which has seen employment levels drop and employees willing to endure more stress at work to avoid retrenchment. These impacts include serious health and financial consequences. Attempts should therefore be made to effectively manage and address work stress to lessen these dire consequences. Many models have been developed and theorised to assist in explaining work stress, the pinnacle of these being the job demands-resources (JD-R) model. In JD-R theory, the dual process explains that work-related well-being follows the following processes: An energetic, also called the health impairment process, in which job demands leads to ill health outcomes through burnout; and then a motivational process which presents that job resources leads to positive organisational outcomes, e.g. organisational commitment, through engagement. The main objectives of this research were 1) to investigate a JD-R model in a large South African sample with a categorical estimator; 2) to investigate the reversed causal hypotheses of burnout and engagement in job demands-resources theory over time; 3) to investigate the likelihood of reporting treatment for health conditions based on burnout and engagement, and 4) to investigate the link between burnout and objective financial outcomes, i.e. by medical aid provider expenditure. To achieve the first objective a cross-sectional design was used (n = 15 633) covering numerous sectors in South Africa. A dual process model was specified with job demands (work overload) leading to ill health through burnout, and job resources (colleague and supervisor support, communication, growth opportunities and role clarity) leading to organisational commitment through engagement. Results of structural equation modelling indicated that the proposed JD-R model was a good fit to the sample. Furthermore, burnout was found to mediate the relationship between job demands and ill health with a medium effect. Engagement was found to mediate the relationship between job resources and organisational commitment with a large effect. The second objective, concerning reversed causality, was achieved with a longitudinal design (n = 593). The hypothesized model included burnout and engagement at time one, and at time two work overload as indicator of job demands, and colleague and supervisor support, communication, growth opportunities and role clarity as indicators of job resources. Results indicated that burnout had a significant negative reversed causal effect to supervisor support and colleague support. Engagement showed only one significant result, i.e. a small negative reversed causal relationship with supervisor support. To achieve the third objective, a cross-sectional design was used (n = 7 895). Results for logistic regression analyses showed that an increase in burnout was associated with a significant increase in the estimated odds for reporting an affirmative answer for receiving treatment for any of the health conditions, i.e. cardiovascular conditions, cholesterol, depression, diabetes, hypertension and irritable bowel syndrome. In contrast, an increase in engagement was associated with a decrease in affirmative reporting for cardiovascular conditions, cholesterol and depression; but not for diabetes, hypertension or irritable bowel syndrome. Addressing the link between burnout and financial outcomes was the fourth objective; and met with a cross-sectional design (n = 3 182). Participants were divided into a high and low burnout group based on the comorbidity of exhaustion and cynicism Analysis of covariance (ANCOVA) was implemented, controlling for age and gender, to investigate the difference in medical aid provider expenditure of the two groups. Results revealed that expenditure in the high burnout group was consistently more in all cases, compared to the low burnout group. By way of conclusion, the implications of the research were discussed and recommendations for managers and for future research were made. / Thesis (PhD (Industrial Psychology))--North-West University, Potchefstroom Campus, 2013