Tag | Value |
---|---|
file | Assumptions_vufsw-levene's_test-1050-en_vufsw-levene's_test-1050-en |
name | vufsw-levene's test-1050-en |
section | assumptions/homogeneity of variance/levene's test |
type | schoice |
solution | TRUE, FALSE, FALSE, FALSE, FALSE, FALSE |
Type | conceptual |
Program | NA |
Language | English |
Level | statistical literacy |
You run a MANOVA with 3 dependent variables and equal group sizes for the categories of the independent variable.
There are 3 Levene’s tests performed. If one of them is statistically [–I–], you can conclude that the assumption of ‘equality of covariance matrices’ is [–II–].
In MANOVA there is one Levene’s test performed for every dependent variable. Therefore, there are 3 Levene’s tests done in this case. If one of them is statistically significant, this means that the variances cannot be assumed equal across groups for one of the dependent variables. Therefore, the assumption of the homogeneity of the covariance matrices (which is another word for the equality of variances and covariances) is violated.