The Importance of the Effect of Exogenous Interaction Factors on Endogenous Variables in Accounting Modeling

Most researchers in accounting use models having dummy independent variables to divide the research objects into two or more groups of objects or individuals. In other researches, they employ more than one variable in the model. However, the dummy variable(s) that stand alone may implicate improper correlations between the numerical independent variables with the corresponding dependent variable within the groups, namely the numerical independent variables have the same effects on the dependent variables within all groups of individuals. To overcome this problem, interaction factors between the dummies and the numerical variable(s) should be used as additional independent variables. So that the models considered would be heterogeneous regression models. It is found, only very few researchers employ the heterogeneous regression models, or the interaction models in general, in top accounting journal, recently. This paper attempts to show that the absence of interaction models may mislead the causal relationship between the independent and dependent variables, specifically between the numerical variables, of the models. The main objective of this paper is to present the importance of interaction models should be applied as the basic statistical models. It is expected that this paper can contribute for a better understanding to the students and less experience researchers in developing the statistical models, in various field of studies.