Confidence ellipses around categories
The plotellipses() function draws confidence ellipses around the categories of several categorical variables.
The objective is to see whether the categories of a categorical variable are significantly different from each other.
We are going to use the data set "tea" and draw confidence ellipses for four categorical variables: "where", "price", "sex" and "age_Q".
First load the package and the data set by typing:
Then perform an MCA:
res.mca = MCA(tea, quanti.sup=19, quali.sup=c(20:36))
Then launch the plotellipses() function:
#res.mca: an object of class MCA
#keepvar: a boolean or numeric vector of indexes of variables or a character vector of names of variables
Categories of "sex" and "where" are significantly different from each other.
For "age_Q" variable, the category "15-24" is significantly different from the others and other categories are not significantly different from each other.
For "price" variable, "p_upscale" and "p_variable" are significantly different from the others and from each other whereas other categories are not significantly different from each other.