ABSTRACT
Despite prior research explaining the benefits of using structural equation modeling (SEM) for analyzing accounting behavioral data, SEM remains underutilized in accounting behavioral research relative to related and reference domains such as psychology, information systems, and management. Prior research posits the frequency with which accounting behavioral data violate SEM assumptions as one probable reason for this underutilization. Accounting behavioral researchers may be unfamiliar with the techniques and approaches available to develop and estimate structural models when data violate SEM assumptions. Given this unfamiliarity, researches may opt to use less informative techniques. The purpose of this paper is to provide guidance on the testing, judgment, and decision-making processes that influence SEM estimation, analysis, and reporting with accounting behavioral data. A structural model is developed, tested, and evaluated using accounting behavioral data that violate, to varying degrees, the assumptions of SEM.
