The factorial validity and measurement invariance of the Maslach Burnout Inventory for human services

This paper first examines the factorial validity of the Maslach Burnout Inventory for human services professions (MBI-HSS). The authors test the original 22-item version and a shortened 20item version. Model fit is evaluated with confirmatory factor analysis starting with data collected from a sample of hospital nurses (n = 2515) and a sample of nurses and assistants working in residential welfare institutions (n = 1639). Only the fit of the 20-item MBI-HSS was good in both samples. Next the hypothesis of measurement invariance of the 20-item MBI-HSS across the samples is examined, but this hypothesis does not hold. Partial invariance at the level of the subscales also could not be confirmed. The authors conclude that for hospital nurses the MBI-HSS factors mean something other than what they mean for professionals working in welfare institutions, and suggest that future research could focus upon this difference. Copyright (c) 2006 John Wiley & Sons, Ltd.