The point of departure of this study is that there is a need to successfully integrate and identify the relationship between employees’ perceptions of employment equity and black economic empowerment, the influence thereof on the various factors associated with employee commitment as well as the role of the mentor and union commitment. Thus, the development of a predictive model becomes an important tool to be used by organisations in future. Literature research The primary objective of the study is to provide research evidence for a predictive model of employee commitment. The secondary objectives focus on outlining the concept of commitment; describing comprehensively the antecedents of employee commitment; describing the relationship between employee commitment and union commitment; describing the perceptions of the Employment Equity Act; describing the perceptions of the Broad-based Black Economic Empowerment Act; and describing the concept of mentorship. Lastly, the role of a mentor as a moderator of organisation commitment, the perceptions of employment equity and the perceptions of black economic empowerment are described. A review of the relevant literature reveals that the construct of organisation commitment as characterised by the relevant literature includes a large number of explanatory and descriptive concepts without an attempt to integrate these concepts and therefore there is a lack of parsimony (Roodt, 1991, 1992). Organisation commitment has not developed in an evolutionary way with regard to meaning and relationships with other commitment concepts (Morrow, 1983). Knoop (1986); Morrow and McElroy (1986); O’Reilly and Chatman (1986); Rabinowitz and Hall (1977) and Roodt (2004) also support this idea. The relevant literature also confirms the lack of the integration of all the variables correlated with commitment and differences in opinion about dual commitment (organisation and union commitment). Employment equity and mentorship research are also fragmented and little research has been conducted on black economic empowerment. This emphasises the need for an integrated predictive model of employee commitment. Empirical research objective The primary objective of the study was to investigate key relationships between variables in the model in order to propose a predictive model of employee commitment. Other objectives were defined as follows: • Determine the interactive relationship between organisation-related commitment (dependent variable) and race, gender, age, tenure, marital status, level of education, home language and job levels (independent variables). • Determine the interactive relationship between union commitment (dependent variable) and race, gender, age, tenure, marital status, level of education, home language and job levels (independent variables). • Determine the interactive relationship between the perceptions of employment equity (dependent variable) and race, gender, age, tenure, marital status, level of education, home language and job levels (independent variables). • Determine the interactive relationship between the perceptions of black economic empowerment (dependent variable) and race, gender, age, tenure, marital status, level of education, home language and job levels (independent variables). • Determine the relationship between organisation-related commitment and union commitment. • Determine how the relationship between background variables and organisation-related commitment is mediated by the perceptions of employment equity. • Determine how the relationship between background variables and organisation-related commitment is mediated by the perceptions of black economic empowerment. • Determine whether the perceptions of the mentor’s role mediate the relationship between the perceptions of employment equity and organisation-related commitment. • Determine whether the perceptions of the mentor’s role mediate the relationship between the perceptions of black economic empowerment and organisation-related commitment. • Determine whether employees with positive perceptions of the mentor’s role are more committed to the organisation. • Determine how the relationship between background variables and union commitment is mediated by the perceptions of employment equity. • Determine how the relationship between background variables and union commitment is mediated by the perceptions of black economic empowerment. • Determine whether the perceptions of the mentor’s role mediate the relationship between the perceptions of employment equity and union commitment. • Determine whether the perceptions of the mentor’s role mediate the relationship between the perceptions of black economic empowerment and union commitment. • Determine whether employees with positive perceptions of the mentor’s role are more committed to the union. Participants A primary data set was used. The primary data was obtained from a sample drawn from a large transport organisation. A convenience sample was drawn from a sampling frame of 1 200 employees and yielded 637 completed questionnaires. A response rate of 53% was obtained. Only completed records were used for the data analyses. The respondents were predominantly Afrikaans-speaking, white males, 36 years and older, with 10 years of service or longer, who are married with a standard 10 or higher qualification. The participants were predominantly drawn from the junior officer group and belonged to a union. The majority did not have a mentor. The measuring instrument The Employee Commitment Questionnaire, the data-gathering tool of this study, consists of five questionnaires, namely Organisation-related Commitment, Union Commitment, Perceptions of Employment Equity, Perceptions of Black Economic Empowerment and Perceptions of the Mentor’s Role Questionnaires. The combined total items of the questionnaire consisted of 103 items. The research procedure The primary data set was obtained from a sample drawn from a transport organisation. In order to have an inclusive approach, the researcher used the intranet and sent hard copies of the survey to employees. Participation was voluntary, and confidentiality and anonymity were guaranteed. Statistical analysis The particular statistical procedures used were selected based on their suitability to test the research hypotheses of the study. These procedures include descriptive statistics, factor analyses, analyses of variance and covariance and a General Linear Modelling (GLM). In respect of the factor analyses, a procedure developed by Schepers (1992) was followed. This procedure includes first and second level factor analyses. A General Linear Modelling (GLM) was used to evaluate the predictive model of employee commitment. The Statistical Consultation Service of the Rand Afrikaans University conducted the analyses. All the calculations were done by means of the SPSS Windows program of SPSS International. Conclusions and recommendations An empirical predictive model of organisation-related commitment and union commitment was developed. This model indicates that organisation-related commitment and union commitment are mainly predicted by perceptions of employment equity and perceptions of black economic empowerment. These perceptions have a strong bearing on people’s beliefs, values and needs. This could be a possible reason for the large amount of variance that is explained. The theoretical model proposed initially was amended to reflect the findings. The study has little nomothetic value, as it was limited to a single organisation. Further research needs to be conducted across organisations in similar contexts to establish the external validity of the findings. The findings have some theoretical value as the perceptions of employment equity and the perceptions of black economic empowerment were for the first time included as predictors of employee commitment. Furthermore, a more sophisticated multi-variate General Linear Modelling (GLM) was used for the prediction of employee commitment. Isolating predictors that explain the variance in the criterion provided a parsimonious predictive model. The model also illustrates possible significant interaction effects between the different predictor variables. The model serves as a good point of departure for understanding and explaining employee commitment in a diverse workforce setting where the perceptions of employment equity and the perceptions of black economic empowerment are applied. / Prof. Gert Roodt