Logistic regression modeling of risk factor on leukemia

Description

Abstract

Logistic regression modeling of risk factor on leukemia. The main focus of this study is to demonstrate the application of binary logistic regression methods in modelling the risk factors of leukemia. The data sourced were analyzed using the binary logistic regression model in SPSS 22.0. The study shows that frequent exposure to radiation is insignificant in experiencing leukemia having a 0.999 level of significance. The odd of frequent radiation has not effectively superseded others despite its high level (11796776426/098). Age reported a p-value of 0.562 which is insignificant and having an odd of 0.098. The result of the study also shows that cancer syndrome and smoking habits have no significant impact on the incidence of leukemia among individuals. The result of the study revealed that the family history of individuals has a significant effect in influencing the event of having leukemia in this study since it has a p-value of less than 0.05. The study also pointed that the odds of having a leukemia level is 53.786 times the other factors presented. The evaluation of the model adequacy shows that the Cox and Snell R Square statistic explained that the 52.4% probability of an event of having leukemia is explained by the logistic regression model.