Veriable analysis using the hayes process method: An application in the fi̇eld of health

Çetinel, MS, Mediator Variable Analysis Using the Hayes Process Method: An Application in the Field of Health, Van Yuzuncu Yil University, Institute of Health Sciences, Faculty of Medicine, Department of Biostatistics, Master's Thesis, Van, 2025. This study examines the role of mediator variables in the relationships between independent and dependent variables in the field of health and analyses these relationships using the Hayes Process method. Traditional mediation analysis methods have limitations such as normal distribution assumption, low statistical power, and inability to directly test indirect effects. In contrast, the Hayes Process method can test indirect effects without requiring a normal distribution assumption and provides more reliable results thanks to the bootstrapping technique. In the study, simple and multiple mediation models were applied using open access data sets. In the study, how the effect of the age variable on biochemical markers (CREA, CAL, ALT, NAT, PCR, GLU, AST) is related to LDH levels was analyzed. The findings reveal that mediating variables play a critical role in the relationship between the independent variable and the dependent variable. The results obtained show that the Hayes Process method is a powerful analysis tool in health research in terms of determining complex relationships, testing indirect effects and making reliable inferences in statistical modelling. In addition, it has been observed that this method provides more flexible, valid and reliable results compared to traditional statistical approaches. In conclusion, this study aims to increase the effectiveness of statistical analyses in the field of health, to better understand the role of mediating variables and to provide important methodological and practical contributions to the development of health policies. Keywords: Mediation analysis, Hayes Process, bootstrapping, moderation analysis.