ANCOVA (analysis of covariance) can be used in observational designs, in experimental designs, and in regression models. 

• In observational designs, ANCOVA is used to remove the effects of variables that modify the relationship between the interval dependent and the categorical independents. 
• In experimental designs, ANCOVA can be used to control factors that can be measured on an interval scale and not on factors which cannot be randomized. 
• In regression models, ANCOVA can be used to fit regressions where both categorical and interval independents are present.