Single-factor independent groups design with covariate
This chapter deals with the application of adjustment to the dependent variable of a single-factor independent groups experiment. It shows how the covariate adjustment has an effect which appears as an additional term in the underlying model. Covariate adjustment makes use of one measurement, the covariate, to adjust another measurement, the dependent variable. Each subject’s score on the dependent variable is adjusted to what it would have been if all subjects had an identical value of the covariate. Scores are also obtained or are already available on a covariate. The weight of the rat in grams is the covariate. The bracketed expression consists of covar, the value of an individual’s covariate score with 394.61 subtracted from it. 394.61 is the mean covariate. The rationale for the covariate effect term is that it is a deviation from the score that would be expected from knowledge of only the overall mean and conditions effect.