TESTING FOR THE FACTORIAL VALIDITY, REPLICATION, AND INVARIANCE OF A MEASURING INSTRUMENT - A PARADIGMATIC APPLICATION BASED ON THE MASLACHBURNOUT INVENTORY

The intent of this article is twofold: (a) to report substantive findi ngs from a cross-validated study that tested for the factorial validit y and invariance of the Maslach Burnout Inventory (MBI) across gender for elementary (males n = 742; females n = 801) and secondary (males n = 659; females n = 721) teachers, and (b) for purposes of helping rea ders who may be new to the structural modeling methodology, to provide an example of how these procedures may be applied in testing for the construct validity of a measuring instrument. Unique to this article i s the use of multiple model-fitting criteria that address issues relat ed to nonnormality of the data, complexity and replicability of the mo del, and error of approximation and sampling error in the estimation o f fit; other ''experimental'' indices of fit are also included. Substa ntively, although results argue generally for the equivalency of the M BI across gender, calibration/validation groups, and teaching panels, they also suggest the need for a reexamination of content and retestin g of construct validity related to six of the 22 items.