Environmental antecedents of workplace bullying: A multi-design approach

The overall aim of this thesis is to contribute to an increased understanding of why bullying occurs at workplaces. Workplace bullying may be understood as the process of repeated and prolonged mistreatment and abuse of an individual at work who perceive to have little opportunity or recourses to retaliate in kind. Although the reasons for why bullying develops are likely to be both complex and interwoven, and may be related to both characteristics of the target and characteristics of the perpetrator, prevailing explanations all emphasise the role of the employing organisation in this process. In line with Leymann’s (1996) work environment hypothesis, stressful working environments have for long been assumed to create conditions that, directly or indirectly, may lead to the development of workplace bullying. Although an increasing amount of research have investigated the role of the work environment in relation to workplace bullying during the last couple of decades, many empirical as well as theoretical questions regarding why bullying occurs at workplaces are still left unanswered. This thesis comprises four empirical studies that all examine work environment factors as antecedents of workplace bullying by means of different research approaches and analytical designs. Paper 1 reports on data from a cross-sectional representative sample of the Norwegian workforce. The aim of Paper 1 was to simultaneously investigate a broad spectrum of job characteristics as antecedents of exposure to workplace bullying, in order to examine the relative importance of various factors in relation to bullying. The findings showed role conflict, interpersonal conflicts, and tyrannical and laissez-faire leadership behaviour to be strong predictors of exposure to workplace bullying. However, the strength of associations differed for various measures of bullying, with associations between the predictor variables and bullying measured by a behavioural experience measure found to be generally stronger as compared to a self-labelling measure of bullying. Support was also found for an interactive relationship in explaining bullying, between both decision authority and role conflict for different levels of laissez-faire leadership. The findings also showed that not only targets, but also observers of bullying assessed their work environment more negatively than did noninvolved employees. Paper 2 reports on data from the same cross-sectional sample as employed in Paper 1. The aim of Paper 2 was to investigate antecedent factors for engaging in bullying of others at work. Drawing on previous findings from workplace aggression research and the stressoremotion model of counterproductive work behaviour (Spector & Fox, 2005), Paper 2 investigated the effects of both individual and situational factors as predictors of being a perpetrator of bullying. Results from logistic regression analysis showed that being exposed to workplace bullying, regardless of the frequency, and being male strongly predicted involvement in bullying of others. Although supervisors are often argued to be perpetrators more often than subordinates, the results of Paper 2 identified no such differences as regards perpetrator status. Among the situational factors investigated, role conflict and interpersonal conflicts significantly predicted being a perpetrator of bullying. The findings of Paper 2 highlight the importance of also considering actor-oriented approaches when aiming to explain the occurrence of workplace bullying. Paper 3 reports on data from a longitudinal two-wave sample of the sample constituting the sample for Paper 1 and Paper 2. Although an increasing amount of crosssectional studies have shown perceived job characteristics to be related to exposure to workplace bullying, no studies have so far investigated the causal direction of this relationship by means of longitudinal data. The aim of Paper 3 was to investigate causal relationships according to a proposed individual level model, in which perceived role stress is regarded as antecedents of exposure to workplace bullying (Bowling & Beehr, 2006). However, contrary to hypothesised causal relationships and interpretations of prior crosssectional findings on perceived role stress as antecedents of exposure to workplace bullying, the findings failed to identify any significant relationships in support of such an explanation. Rather, exposure to workplace bullying accounted for subsequent variation in role ambiguity, role conflict, and role overload, alike. The findings of Paper 3 question the usefulness of regarding factors such as perceived role stress as antecedent conditions of exposure to workplace bullying at an individual level of analysis. Paper 4 reports on data from a large cross-sectional sample of employees nested within departments. The aim of Paper 4 was to investigate group-level environmental characteristics as predictors of workplace bullying. In line with the work environment hypothesis (Leymann, 1996), the predictive effects of role conflict, role ambiguity and nonsupportive leadership practises were investigated applying group-level and multilevel analysis. The findings show the presence of role conflict and non-supportive leadership practises to predict the overall level of bullying within departments, and further, that being exposed to such environmental characteristics are associated with an increased likelihood of being exposed to bullying. The findings from Paper 4 highlight the importance of taking into account work-group characteristics in explaining bullying, and point attention towards areas in which preventive measures against bullying may be successfully implemented. In conclusion, the findings of this thesis render general support to Leymann’s (1996) work environment hypothesis in that prevailing organisational conditions to a large extent may account for the overall incidence of workplace bullying, influencing on the prevalence of both perpetrators and targets of bullying within organisations. Leadership practises, interpersonal conflicts, and role conflict appear to be decisive factors in this process. At the same time, the present findings highlight challenges in investigating factors that may contribute to the development of bullying at workplaces. The overall findings question the general assumption that individual level perceptions of environmental factors may be regarded as actual antecedents of exposure to workplace bullying. On the other hand, such individual level perceptions may account for why individuals engage in bullying of others, thereby indirectly influencing also on exposure to workplace bullying. Thus, a sound and thorough integration of actor-oriented and target-oriented approaches in relation to workplace bullying is likely to bring valuable knowledge to the field in attempting to bring an end to the problem. Future research aimed at explaining why bullying occurs at workplaces is also likely to benefit from adopting multilevel approaches by simultaneously investigating both individual-level and higher-order factors in relation to workplace bullying. Being able to establish such relationships in relation to workplace bullying will undoubtedly bring about valuable knowledge in terms of counteracting the problem, but will also yield valuable knowledge to researchers in terms of developing sound and explicit theoretical frameworks that can be rightfully empirically tested