ANOVA is a commonly used method to study the significance of influencing factors,and it requires the data to meet normal distribution and have homogeneity of variance.However,these conditions cannot often be met in the research of many practical problems,so non-parametric statistical method is a good choice.At present,Kruskal-Wallis single-factor ANOVA is commonly used in practical problems,which often needs the consideration of the influence of multiple factors and their interaction effects.While single-factor nonparametric ANOVA is no longer applicable,so the study discusses the theory of two-factor nonparametric ANOVA and its application in practice.Based on Kruskal-Wallis single-factor ANOVA,the study derives a theoretical model of two-factor nonparametric ANOVA,selects the incidence data of solid cancer,evaluates the effects of ion radiation dose and radiation exposure duration on the incidence of solid cancer with the model,and verifies two-factor nonparametric ANOVA by logistic regression model.The results show that the results of two-factor non-parametric ANOVA and logistic regression analysis are consistent,indicating that there is no statistically significant difference of both ion radiation dose and radiation exposure duration on solid cancer risk,and the interaction effect between ion radiation dose and radiation exposure duration is not significant.This study provides a theoretical basis for the application of two-factor nonparametric ANOVA in cancer risk assessment,and demonstrates its practicability in practice with practical cases.
关键词
方差分析/非参数检验/Kruskal-Wallis检验/logistic回归模型
Key words
ANOVA/Nonparametric test/Kruskal-Wallis test/Logistic regression model
维普
万方数据