| dc.description.abstract | Workaholism is recognised as a problem, for both the individual and the organisation. The phenomenon is associated with several negative outcomes (i.e. lower levels of work engagement, less commitment to the organisation and also heightened levels of individual burnout), which adversely impact the well-being of individuals as well as the productivity and efficiency of organisations. The modern economic environment is also driving individuals to work harder than ever before, ultimately reinforcing workaholic behaviour. In spite of the destructive nature of workaholism, there is no validated measuring instrument available for South African organisations. The study therefore intended to validate the Dutch Work Addiction Scale (DUWAS-10) within the South African context.
Specifically, the research aimed at investigating whether the DUWAS-10 is a valid and reliable measuring instrument, by exploring its factor structure, convergent validity, discriminant validity and predictive validity. To this end, a quantitative research design was used, i.e. a cross-sectional research approach was implemented. Participants from the banking industry (N = 345) were chosen based on their convenient availability and proximity to the researcher. The reliability of the DUWAS-10 was explored by considering Cronbach‟s alpha coefficients, which should be 0.70 and larger to be considered acceptable. Furthermore, the convergent and discriminant validity was examined; convergent validity was established by determining the degree to which workaholism is similar to other theoretical constructs with which it should be comparable (e.g. work overload and work hours), whereas discriminant validity was established by exploring whether workaholism differs from constructs from which it should differ theoretically (e.g. work engagement). Lastly, predictive validity was established by investigating the regressions between workaholism and applicable organisational outcomes (e.g. work engagement, organisational commitment and burnout).
The results showed that the DUWAS-10 should be operationalised as a one-factor structure, as the two-factor structure (i.e. working excessively and working compulsively) was rejected due to problematic discriminant validity between those two components. Furthermore, the scale showed acceptable reliability (α = 0.78) as well as convergent and discriminant validity by meeting the specified criteria. It was also found that the scale provided valid relationship directions with pre-determined organisational outcomes (e.g. work engagement, organisational commitment and burnout). It can therefore be concluded that workaholism has a negative relationship with work engagement and organisational commitment, while it has a positive relationship with burnout. Furthermore, the evidence suggests that the workaholism scale can be used to assess workaholism within the South African context, specifically in the banking industry | en_US |
