A cross-level analysis of burnout in supervisor-subordinate dyads

Cody Logan Chullen, Purdue University


Employee burnout has an extensive history in applied psychology, with decades of research and hundreds of studies examining both its antecedents and consequences (Cordes & Dougherty, 1993; Shirom, 1989; Taris, 2006). Despite these advancements, literature continues to largely disregard important temporal dynamics associated with the onset and duration of burnout, lags between causal effects, and pace of progression (Ashforth & Lee, 1997). However, without a thorough understanding of burnout and its relationship to time organizations may struggle to properly develop intervention strategies aimed at reducing and eliminating burnout within the workforce. In this paper, I advance burnout theory by integrating five theoretical frameworks - emotional contagion theory (Hatfield, Cacioppo, & Rapson, 1994), social exchange theory (Blau, 1964), role theory (Rizzo, House, & Lirtzman, 1970), conservation of resources (COR) theory (Hobfoll, 1989), and the job demands-resources model (JD-R) (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) - to explain burnout change from a dyadic perspective. Specifically, I articulate a model of dyadic burnout change wherein I contend that supervisor burnout leads to change in subordinate burnout both directly and indirectly. I tested my model using a sample of 59 supervisors and 157 subordinates drawn from two organizations (one within the field of manufacturing and one within the field of mental health services). Using Hierarchical Linear Modeling (HLM), results did not indicate that supervisor burnout leads to change in subordinate burnout either directly or indirectly, highlighting that once burnout progresses to a certain stage it remains relatively stable across time. Thus, if organizations are interested in reducing and eliminating burnout's prevalence it is very important they identify early signs of developing burnout and that they consider monitoring employees at higher risk such as organizational newcomers. However, in a series of follow-up exploratory analyses (which disregarded considerations of time) results did indicate that supervisor emotional exhaustion in particular was positively associated with subordinate emotional exhaustion both directly and indirectly. Although causality and change cannot be inferred, these findings and their implications for future research are discussed.^




Benjamin B. Dunford, Purdue University.

Subject Area

Business Administration, Management|Psychology, Industrial

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server