Analysis of Variance and Covariance

Among the statistical procedures used in the analysis of data are the analysis of variance (ANOVA), analysis of covariance (ANCOVA), multivariate analysis of variance (MANOVA), multivariate analysis of covariance (MANCOVA) procedures which require that cases are randomly allocated to treatment and control groups or randomly selected from a population where the groups occur naturally.

These procedures seek to partition the total variance into its components by the analysis of sums of squared terms derived from the data. Where the effect of a variable confounds the analysis, the effect of the variable may be removed or controlled by a covariance adjustment.

However, alternative procedures of analysis, such as regression analysis, must be used where the variable, whose effects are being controlled in statistical analysis, might have influenced the treatment conditions under consideration. Where there is more than one criterion variable, the term multivariate analysis is commonly employed.

Compiled by John P. Keeves 1995