Multiple Correspondence Analysis

Multiple Correspondence Analysis (MCA) is a method to analyze qualitative data and it is an extension of Correspondence analysis (CA). MCA can be used to detect patterns or structure in the data as well as in dimension reduction.

This exercise is part of the course

Helsinki Open Data Science

View Course

Exercise instructions

Do multiple correspondence analysis with the function MCA(). Give tea_time as the functions first argument. Note that the MCA() function visualizes the analysis by default, and the plots can be turned off with the argument graph = FALSE.
Look at the summary of the model.
Plot the variables of the model. You can either plot the variables or the individuals or both. You can change which one to plot with the invisible argument.
Adjust the code: add argument habillage = "quali" (how french!) to the plot. Do you notice what changes?

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# tea_time is available

# multiple correspondence analysis
mca <- MCA("change me!", graph = FALSE)

# summary of the model


# visualize MCA
plot("change me!", invisible=c("ind"))

Edit and Run Code