ABSTRACT
This chapter presents a method called the analysis of covariance. This method incorporates the covariates into the model along with the controlled factors in order to reduce the experimental error. It focuses on the situation involving one covariate, a completely randomized design, and a linear relationship between the covariate and the response variable. The chapter discusses procedures for comparing the treatments when the assumption of equal slopes is reasonable. A regression model can be used to analyze data obtained in an analysis of covariance problem. The recommendation for an analysis of covariance problem is to first test for equal slopes. The major purpose of an analysis of covariance is to adjust for the effect of the covariate and as a consequence reduce the experimental error. Analysis of covariance is an alternative to blocking. When blocking, we group the experimental units into blocks usually based upon the "value" of the blocking variable.
