A detailed explanation and graphical representation of the Blinder-Oaxaca decomposition method with its application in health inequalities

This paper introduces the Blinder-Oaxaca decomposition method to be applied in explaining inequality in health outcome across any two groups. In order to understand every aspect of the inequality, multiple regression model can be used in a way to decompose the inequality into contributing factors. The method can therefore be indicated to what extent of the difference in mean predicted outcome between two groups is due to differences in the levels of observable characteristics (acceptable and fair). Assuming the identical characteristics in the two groups, the remaining inequality can be due to differential effects of the characteristics, maybe discrimination, and unobserved factors that not included in the model. Thus, using the decomposition methods can identify the contribution of each particular factor in moderating the current inequality. Accordingly, more detailed information can be provided for policy-makers, especially concerning modifiable factors. The method is theoretically described in detail and schematically presented. In the following, some criticisms of the model are reviewed, and several statistical commands are represented for performing the method, as well. Furthermore, the application of it in the health inequality with an applied example is presented.