Nowadays, the complexity of organisational settings increases the demands workers need to cope with. An example of this complexity is showed by Arlington and Baker (2000) who introduced the concept of multiple accountabilities (MA) which refers to having to report to several individuals at the same or/and at higher levels of the hierarchy. MA might generate conflicting expectations between supervisors, being a stressor for people involved and a prejudicial factor for workers’ well-being. When such stressor is experienced during a long time, it might be a source of burnout.
Moreover, drawing from the Social Power Theory (French & Raven, 1959), MA might be a source of Workplace Incivility (WI; Andersson and Pearson, 1999), which refers to low-intensity deviant behaviors characterized by perpetrators’ ambiguous intent to harm the victim. WI might act as an instrument to reaffirm power (Cortina, 2001) and people who have MA might be more exposed to such negative dynamics, with a detrimental effect for their well-being. However, and according to the Job Demands-Resources model (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001), some resources, such as role clarity (RC), might moderate the negative effect of MA. Clarify roles might empower workers helping them to manage with multiple reports.
The goal of this research is to analyse the effect of MA on burnout through WI. Moreover, we are going to assess if RC moderates the effect of MA on both WI and burnout.
A total of 249 people employed at a multinational organisation completed an online questionnaire. 52.6% of the sample were women; the age of 44.2% of workers ranges between 40 and 49 years old. Controlling for age, sex and education, the macro PROCESS, model 8, was used to test the moderated mediation of MA on burnout mediated by WI, with the paths between MA and WI, and MA and burnout moderated by RC.
Results showed a significant interaction between MA and RC (b= -.25, t(242)= -2.87, p<.01) predicting WI. The relationship between MA and WI is significant at all levels of the moderator but its magnitude diminishes at higher levels of RC. Moreover, a significant interaction between MA and RC exists predicting burnout (b= .18, t(241)= 2.52, p<.05). Results show that higher levels of MA always predict higher levels of burnout, but when MA is low, lower levels of RC enhance burnout. Data for the index of moderated mediation demonstrated that the bootstrapped 95% confidence interval for burnout ranges from -.109 to -.016, indicating a conditional significant indirect effect. The confidence interval for mediation through WI excluded zero at any level of the moderator.
This study has several implications. At a theoretical level, we extend the current knowledge about MA; moreover, we demonstrate that MA is an antecedent of WI and burnout. We also demonstrated the positive effect of RC. At practical level, organisations should be aware of RC positive effect when demands, such as MA, are present. Organisations should improve such resources, clarifying expectations and fostering open communication between people involved.