Term of Award

Spring 2021

Degree Name

Doctor of Public Health in Biostatistics (Dr.P.H.)

Document Type and Release Option

Dissertation (restricted to Georgia Southern)

Copyright Statement / License for Reuse

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


College of Public Health

Committee Chair

Haresh Rochani

Committee Member 1

Hani Samawi

Committee Member 2

Jingjing Yin


A mediating variable is a variable that is intermediate in the causal path relating an independent variable to a dependent variable in statistical analysis. The mediation analysis of using a categorical predictor, mediator, and outcome variables was investigated in the literature. The equivalence of the mediated effect being equal to the difference between the total effect and the direct effect is not true when the dependent variable Y and the mediator variable M are categorical. Generally, standardized estimates are used in calculating the mediated effect and standard error in this scenario. Furthermore, it is extremely common to have missing data even after having a well-controlled study. It is also well known that missingness in a dataset has often been proven to produce biased results and can impact the power of a study. This dissertation focused on using the extended Baker, Rosenberger, and Dersimonian (BRD) model technique proposed by Rochani et al. (2017) to estimate the mediation effect under non-ignorable missing mechanisms. This dissertation also proposes four identifiable models to estimate the mediation effect for missingness in one categorical variable with two fully observed categorical variables. The relative bias and Mean Square Error (MSE) were used to compare the performance of these models to the Complete Case (CC) method and Multiple Imputation (MI) method in estimating the mediated effect () under the non-ignorable missing mechanism models for valid inference and conclusion. The application of these models in estimating the mediated effect was demonstrated using the Multiple Risk Factor Intervention Trial (MRFIT) dataset.

Research Data and Supplementary Material


Available for download on Sunday, March 29, 2026