is an extension of analysis of variance in which quantitative variates, known as covariates, are fitted in addition to the usual treatment model terms (which involve factors and their interactions). Covariates provide additional numerical information about the experimental units in an analysis of variance, often measured prior to the experiment, and are fitted in order to improve the precision of the treatment estimates.
The analysis fits the treatment terms and then a linear regression for each covariate, allowing you to assess the usefulness of including them in the model. It also adjusts the treatment sums of squares and effects to take account of the way in which covariate values differ between the units that have received the various treatment combinations.