Analysis of covariance (ANCOVA)

It is initially used for an expansion of the analysis of variance which permits to the possible effects of continuous concomitant variables (such as covariates) on the response variable, additionally to the effects of the factor or the treatment variables. Generally supposed that covariates are unaffected by the treatments and that their relationship to the response is linear in nature. If such a relationship exists then the inclusion of covariates in this means decreases the error mean square and thus increases the sensitivity of the F-tests used in assessing the treatment differences. The term now seems to also be more generally used for almost any of the analysis seeking to assess the relationship among response variable and a number of the explanatory variables