Burnout, characterised by feelings of emotional exhaustion, depersonalisation and reduced personal accomplishment can prove detrimental to both the individual employee and the organisation. These negative effects can significantly affect the service culture of client service organisations, since research has shown that burnout amongst front-line service employees can result in these employees displaying negative feelings and behaviour towards their clients and co-workers (Yagil, 2006: 259). Research into antecedents of burnout has primarily focused on organisational and job variables, such as role conflict, role ambiguity, work overload and lack of social support. The present study departed from this tradition by focusing on the relationship between role identities (subjective perceptions) and burnout amongst 100 client service employees in three client service organisations in South Africa. The research was informed by previous studies that suggest that client service employees who feel subordinate to the client and powerless in their interactions with the client may display higher levels of burnout than those who feel in control of the service relationship (Buunk, Peiro, Rodriguez&Bravo, 2007; Vanheule&Verhaeghe, 2004). By applying a sequential mixed-methods approach consisting of a quantitative and a qualitative phase, the research explored the differences in role identities of client service employees who measure higher on burnout with the role identities of client service employees who measure lower on burnout. In the quantitative phase, a survey questionnaire incorporating the Maslach Burnout Inventory – Human Services Survey (Maslach&Jackson, 1996) and a modified version of the Burke-Tully role-identity measurement (Burke&Tully, 1977) was used. The quantitative phase was followed by a qualitative phase consisting of semi-structured interviews with eight higher burnout and nine lower burnout employees. The quantitative data were analysed by means of Maximum Likelihood Factor Analysis (MLFA) with Direct Quartimin rotation, analysis of variance (ANOVA) and Pearson and Spearman correlation analysis. The analysis of qualitative data proceeded through a process of open, axial and selective coding as suggested by Miles and Huberman (1994). Both the quantitative and qualitative data are interpreted within the conceptual framework developed, and a number of findings are presented. Analysis of the quantitative data shows that the Maslach Burnout Inventory (MBI) items load on two, instead of the three factors as conceptualised by Maslach and Jackson (1986). One of the two factors corresponds to the reduced personal accomplishment subscale. The other factor comprises items from both the emotional exhaustion and depersonalisation subscales. The two subscales derived from the factor analysis were then correlated with client service employees‟ descriptions of self in role, counter-role and self in relation to the client descriptions on the bipolar adjective scales. This analysis revealed a number of significant correlations − suggesting a difference in the role identities of client service employees who measure higher on burnout when compared with client service employees who measure lower on burnout. For instance, higher levels of burnout are associated with feeling weak, powerless, unhelpful, inconsiderate, not respected and unimportant. The more rigid, impatient and inconsiderate the client is perceived to be, the higher the levels of experienced burnout. The qualitative data reveal that the role identities of higher burnout employees differ from the role identities of lower burnout employees. While higher burnout employees regard themselves as subordinate to and powerless against the client, lower burnout respondents define themselves as superior to and more knowledgeable than the client. Lower burnout employees are able to exert a level of control and power over the client, while higher burnout employees feel controlled by the client. The qualitative research also illustrates how role identities inform behaviour which may contribute to the development of burnout. The role identities of lower burnout employees also enable self-verification, while the role identities of higher burnout client service employees inhibit self-verification. The study introduces the concept of role identity as an important variable to consider in the development of burnout and links the development of client service role identities to organisational client discourse. In so doing, the study has provided organisational theorists and practitioners with a further point of intervention with which to reduce burnout in client service settings. The study has also developed a conceptual framework, derived from the literature and supported by both qualitative and quantitative findings, that shows how role identity can contribute to role-related attitudes and behaviours that could lead to or inhibit the development of burnout. The study is therefore not merely descriptive in nature, but provides a tentative explanatory framework linking burnout and role identity and exploring the mechanisms by virtue of which this relationship exists. The dissertation concludes with recommendations as to how organisational client discourse may be framed so as to facilitate the creation of role identities which empower the employee in relation to the client. By facilitating the development of empowered client service employees, organisations could greatly reduce levels of experienced burnout. As a result, organisational performance will improve, since lower levels of burnout are associated with reduced absenteeism, increased job satisfaction and commitment to the organisation and improved relationships with clients.