The literature on depression is dominated by theories which focus on individualistic variables, including biological differences, personality, and individual cognition. Whilst the importance of social variables for depression risk and recovery has also been recognised, there has been a notable absence of a unifying theory explaining, how, when, and why they might impact on depression. In recent years, the Social Identity Approach (SIA) - a theoretical framework with roots in social psychology - has been used to provide a new understanding of the role of group processes in depression. The aim of this thesis was to add to a growing body of evidence in support of the SIA to depression by replicating previous findings using a sample, who on average, scored high on a measure of depression symptomology, and by identifying additional mediators of the relationship between social identity processes and depression. Specifically, it was theorised that optimism - a personality variable associated with depression, but traditionally conceptualised as a fixed trait - would vary along with group memberships and mediate the effect of social identity processes on depression. Participants who had experienced depression (N = 288) completed an online survey. It was found that in support of previous research, an increase in group involvement predicted lower depression scores. However, this relationship was mediated by increased optimism. Similarly, identifying more strongly with a specific group predicted depression indirectly through an increase in perceived social support, and increased optimism. For a sub-sample of participants with experience of psychological therapy (N = 135), the negative association between a good therapeutic alliance and depression was serially mediated by increased identification with the therapist, internalisation of the therapist identity between sessions, and increased optimism. The extent to which the therapist was perceived as prototypical of therapists in general also indirectly reduced depression via increased identification with the therapist category, and increased optimism. These findings are discussed in relation to the further development of the social identity approach to depression, with consideration of their implications for onset, maintenance and recovery.