STATISTICAL ANALYSIS ABOUT DETECTING CERTAIN ELEMENTS THAT AFFECT THE STROKE

We investigate and identify several factors that have a major influence on strokes. The study involves a sample of 5,110 individuals from Spain. Participants ranged in age from 0 to 82, comprising 41.4% males and 58.6% females. Among the sample, 4.9% have experienced a stroke. The dataset provides detailed insights into several variables, including gender, age, hypertension, heart disease, marital status, work types, residence types, smoking habits, glucose levels, and body mass index. By employing a Binary Logistic Regression Model, the analysis revealed a significant statistical relationship among the independent variables - age, hypertension, heart disease, and glucose level - concerning strokes, showing a chi-square value with a p-value of 0.000 at a 5% significance level. The model demonstrated a high accuracy rate of 94.8% in predicting cases, with a Cox and Snell R-square of 0.073 and a Nagelkerke R-square of 0.217. The model’s effectiveness was supported by its accuracy in predicting many cases based on the observed groups and predicted probabilities. The findings indicate that factors such as gender, marital status, work types, residence type, and smoking status do not show statistically significant relationships and do not affect the occurrence of stroke cases.