D.Phil. / Identity is a cognitive concept that describes “who I am”, and an important part of an individual’s identity is derived from shared social entities. However, as work and the work setting began playing prominent roles in most people’s social lives, the social identity derived from the working environment became the focus of several studies. It was, however, recognised that the locus of identification stretched beyond the organisation to other social phenomena available in the work setting. This finding encouraged a change of focus, which asserts that facets of work are sources of identification. The construct, work identity (WI), captures this extended understanding of social identity derived from work. There were also several research agendas about ways to strengthen or manipulate the relationship between employees and their work and/or workplace. Concepts such as work involvement, work commitment, work engagement, work centrality, and person-organisation fit represent this category of enquiry. As identity or identification was often mentioned in the definitions of these concepts, the concepts are perceived as WI-associated concepts. As several researchers recognised concept redundancy or concept contamination among the WI-associated concepts and between the WI-associated concepts and WI, further clarification of WI was needed. The purpose of this research project was to generate a model of the antecedents and consequences of WI in multicultural work settings. Conducting the research in a multicultural setting did not only test the theoretical ideas in a challenging context, but it contributed to a better understanding of employees from the research setting. The study used a cross-sectional field survey in order to gather responses from a convenience sample of employees from workplaces in Dubai. Data from 644 respondents was subjected to regression analysis and structural equation modelling. One characteristic of the dataset was the positively skewed distribution within some scales and significant disparity in the mean calculations of different nationality groups. Applying multiple regression to analyse and explore bivariate relationships, the data supported a positive relationship between job resources and WI. As an unexpected weak positive relationship between job demands and WI was also found, further research into the behaviour of job demands is required. A strong predictive relationship between WI and work engagement was indicated and a negative relationship between WI and turnover intentions was supported. Structural equation modelling identified a parsimonious model of WI in multicultural work settings which contained the concepts of job resources and job demands as antecedents of WI, with work engagement and turnover intentions as consequences of WI. The strength of causal relationships within this model was significantly affected by three moderator variables, namely level of education, age and, most strongly, by nationality. Therefore, although a stable covariance model was accepted, different nationality groups still processed the relationships between variables within the model in unique ways. Although WI manifested itself as a single-component structure in previous research that operationalised WI in the same way as this study, three facets of WI emerged here: work centrality, person-organisation fit, and value congruence. The different WI facets did not consistently relate to the antecedents or the consequences of WI in the same way. In addition, in response to findings of redundancy and contamination in previous studies, WI was found to be distinct from work engagement. This finding is a catalyst for future research to explore facets of WI and for researchers to revisit work engagement a consequence of WI and work engagement in multicultural work settings.
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uj/uj:10440 |
Date | 25 October 2012 |
Creators | Bester, Francois |
Source Sets | South African National ETD Portal |
Detected Language | English |
Type | Thesis |
Page generated in 0.0023 seconds