STATISTICAL ANALYSIS STUDYING THE FACTORS AFFECTING HEMOGLOBIN

Abstract

This is a descriptive, cross-sectional study to analyze the effect of alcohol and smoking in the hemoglobin present in blood and determining the other factors that affect it. The data was obtained from the national health insurance service in Korea. The multiple linear regression model was performed on the sample size of 65535 individuals, which contain adults aged between 20 to 85 years of both males and females in Korea. This sample covers people who smoke and drink during their lifetime. There is a statistically significant effect of the explanatory variables (Sex, Age, Height, Weight, Smoking state, Drinking state) on the dependent variable (Hemoglobin), with F-stat (10325.983) and P-value (0.000) at 5% level of significant. The variance inflation factor (VIF) ranged between (1.280 to 3.327); is less than 5; which means that there is no collinearity. Also, the R squared (0.486) is less than Durbin Watson statistic (2.006) which means this model is not spurious suggesting that there is no autocorrelation, or partial correlation in the data. The explanatory variables explain 48.6% of the total variation in hemoglobin levels in the blood. Received: September 7, 2023 Accepted: November 2, 2023