Multivariate Analysis of Variance - Correlation of Dependent Variables

Correlation of Dependent Variables

MANOVA is most effective when dependent variables are moderately correlated (.4 - .7). If dependent variables are too highly correlated it could be assumed that they may be measuring the same variable. It is also recommended to not include dependent variables that are supposedly measuring the same construct in MANOVA (Tabachnick & Fidell, 2007).

Read more about this topic:  Multivariate Analysis Of Variance

Famous quotes containing the words dependent and/or variables:

    One of the baffling things about life is that the purposes of institutions may be ideal, while their administration, dependent upon the faults and weaknesses of human beings, may be bad.
    Mary Barnett Gilson (1877–?)

    The variables are surprisingly few.... One can whip or be whipped; one can eat excrement or quaff urine; mouth and private part can be meet in this or that commerce. After which there is the gray of morning and the sour knowledge that things have remained fairly generally the same since man first met goat and woman.
    George Steiner (b. 1929)