ABSTRACT
Structural equations modelling (SEM) is a family of powerful yet relatively underutilized statistical techniques in the accounting literature. SEM allows a set of relations between one or more independent variables, and one or more dependent variables, to be examined simultaneously. In this chapter, we explain why SEM is preferred over conventional ANOVA and multiple regression analysis. We then outline different SEM techniques, and explain when each should be used. To provide examples of the scope and diversity of SEM techniques, we subsequently review examples of accounting research from the financial accounting, management accounting, audit and tax sub-disciplines. We also discuss deficiencies in SEM reporting practices in the accounting literature, and present “best practice” reporting guidelines for use by behavioural accounting researchers. Finally, we suggest future directions for behavioural accounting researchers to consider when using SEM.
