Graphs for Correspondence Analysis

Readable, complete and pretty graphs for multiple correspondence analysis, correspondence analysis and principal component analysis made with 'FactoMineR'. They can be rendered as interactive 'HTML' plots, showing useful information at mouse hover. The interest is not mainly visual but statistical. It helps the reader to keep in mind the data contained in the cross-table or Burt table while reading the correspondence analysis, thus preventing over-interpretation. Most graphs are made with 'ggplot2', which means that you can use the + syntax to manually add as many graphical pieces you want, or change theme elements. 3D graphs are made with 'plotly'.

ggfacto

ggfacto draws readable, interactive graphs of the principal component, correspondence and multiple correspondence analyses of FactoMineR. Hover over a point, and the graph shows the crosstables behind it, each deviation from the mean coloured: the geometry is always read against the data it summarises. The graphs are ggplot objects, to be extended with +.

The three analyses share one workflow: the analysis, interpret() for its axes, ggfacto() for its graph, and hierarchical_clust() for its clusters.

install.packages("ggfacto")

library(ggfacto)
data(tea, package = "FactoMineR")
mca <- multiple_correspondence_analysis(tea, 1:18)
interpret(mca)
ggfacto(mca, tea, sup_vars = c(sex, SPC), interactive = TRUE)

See the website, with the interactive graphs: https://bricenocenti.github.io/ggfacto/, and its guide, in English or en français.