Indicator Variable Regression Analysis Of Covariance And Its Application Of Medical Research

In medical research, there maybe some troubles in the analysis statistical results because of some uncontrollable factors. How to eliminate the influence of these uncontrollable factors and enable us to make correct comment on various solutions is the problem we need to solve in analysis of covariance. Analysis of covariance is a combination of regression analysis and variance analysis. It eliminates the influence of independent variable by the use of regressional relationships and then it makes the variance analysis. On the basis of studying variance analysis, this paper studies the set-up of indicator variable and the way of definition. It demonstrates the set-up of various indicator variables and the disadvantages. On the basis of understanding the indicator variables, we can put it into the analysis of covariance. By using multiple regression equation, we turn the analysis of regression lines and adjusted means to the analysis of regression coefficient. By adjusting the indicator variable, the paper makes the comparison between pairs of adjusted means. It not only makes the understanding of the issue easier, but also simplify the process of calculation. There is no report so far in our country about the use of indicator variables in covariance analysis.This paper uses the computer programming - SAS and take practical examples, and it makes covariance analysis by using indicator variables. The result is completely consistent with the one that is made by traditional analysis of covariance. It provides theoretical and practical basis for the spread of indicator variable in covariance analysis.