Investigating the mediating role of psychological states and the moderating role of angry rumination in the relationship between supervisors' history of family aggression and subordinates' perceptions of abusive supervision

Abstract

Drawing upon the Social Learning Theory (Bandura, 1973, 1977) and the General Aggression Model (GAM; Anderson & Bushman, 2002), this research program endeavours to provide a deeper understanding of the causes of abusive supervision. In particular, it examines the mediating role of two psychological states (i.e., hostile cognitions and hostile affect) and the moderating role of angry rumination in the relationship between history of family aggression and subordinates' perceptions of abusive supervision. The proposed relationships were tested in a programmatic series of five studies each providing constructive replication and empirical extension of the study findings.

The primary objective of Study 1 was to test the proposed positive relationship between history of family aggression and hostile cognitions and hostile affect. Results supported the hypothesised relationship in a sample of 255 student-parent dyads. Study 2 builds on Study 1 by examining the predicted relationships using a sample of 154 supervisor-subordinate dyads and extending the theoretical model to include abusive supervision. Results supported the proposed mediating role of hostile cognitions and hostile affect.

Study 3 aimed to constructively replicate findings in Study 2 by using a different sample of 191 supervisor-subordinate dyads and operationalisation of hostile cognitions. In particular, the word completion task (Anderson et al., 2004) was used which taps into implicit social cognitions. Results once again supported the mediating role of hostile cognitions and hostile affect in the relationship between history of family aggression and abusive supervision. Alternative model tests also revealed that history of family aggression exerts its influence via a dual-activation process. Study 4 has two main objectives. First, it aimed to rule out alternative explanations by controlling for previously established antecedents of abusive supervision. Second, it examined the moderating role angry rumination in the proposed mediated relationship between history of family aggression and subordinates' perceptions of abusive supervision.

Once again, results supported the main effects and mediation hypotheses. In addition, the conditional indirect effect of history of family aggression in predicting subordinates' perceptions of abusive supervision via hostile affect was stronger for supervisors' with high as opposed to low levels of angry rumination. However, the conditional indirect effect involving hostile cognitions was not supported. Finally, Study 5 aimed to replicate the findings from Study 4 and address issues associated with the use of retrospective data (e.g., history of family aggression) by obtaining parent ratings of history of family aggression. Once again, results supported the main effects and mediation hypotheses. Similar to Study 4, only the conditional indirect effect involving hostile affect was supported.

This research program makes several contributions to the understanding of abusive supervision. First, it presented a different theoretical approach in explaining the occurrence of abusive supervision by implicating the role of social learning experiences, psychological states, and personality characteristics. Second, it addressed methodological issues associated with extant abusive supervision research through the use of multi-source data, constructive replication, and implicit measures of social cognition. Theoretical implications with reference to abusive supervision research and practical implications for reducing and managing its occurrence are discussed. -- provided by Candidate.