Analysis of covariance is actually the examination of distinctions of the mean in the dependent variable that is related with the controlled independent variables. This is also used to compare the one variable in two or more groups in order to consider the values for the variability of variables which is known as covariates. Covariance analysis is also known to be the representative of the set of theory for the co relational data used to represent various systems of equations. Analysis of covariance probably comprises of one way variables or two way variables with the linear regression that is the general linear model. In the analysis one variable is the dependent variable and the other used is the covariate variable.
For the analysis of covariate, following inputs are required to accomplish the analysis.
• Dependent variable: the continuous variable is entered with the name like VarY.
• Factors: the one variable is used for the variance category might be for one way analysis of covariance or for two way analysis of covariance like factor A or factor B.
• Covariates: one or more covariate is used.
• Select: it is the optional input which involves the cases of sub groups.
Regarding General Linear Model
Analysis of covariance involves the implementation procedure as stated for the general linear model. The following points are included in the procedure;
• As in the analysis of variance one way variable when no specifically identified covariate and only one factor.
• For the two way analysis of variance does specify the two factors but not the covariate.
• Accessing on multiple regression when there is no covariate.
Assumptions for the Analysis of Covariance
In the understanding of analysis of covariance, following assumptions should be taken in a view to avoid errors.
• The relationship of dependent variable in case of the independent variable should always be linear.
• The relationship for the covariate variables and the dependent variables is taken as the identical across each independent variable.
• For instance if there appears more than one independent variables then the analysis of covariance assumes that the variance is homogeneous in reality in case of the controlled independent variables.
Analysis of covariance is most probably involves the some of the procedure of the analysis of variance and the regression. Other than that the covariate which is considered to be the continuous variable is introduced in this analysis. The continuous variable that is covariate used in the analysis does affect the dependent variable. The covariate is used for the known significance in the dependent variable. Most of the time values for a variable are selected before certain experiments which are the pretest values and some are taken after the experiment are post test. The analysis of covariance made certain adjustments in order to have the appropriate the analyses that are firstly reducing the approximate errors and secondly adjustments fro the treatment effects with respect to the continuous variable that is covariate. In the analysis justification for the covariate is properly made.
• Huitema, B.E., (1980). The analysis of covariance and alternatives. Wiley-Interscience
• Wildt, A.R., and Ahtola, O.T., (1978) Analysis of covariance: Sage Publications.