F. Husson website
IntroductionPrincipal Component AnalysisCorrespondence Analysis Multiple Correspondence Analysis ClusteringMultiple Factor AnalysisTo concludeForum
For each question, tick the correct answer or answers.
Q1) When describing obtained clusters with the help of quantitative variables: a quantitative variable characterizes a cluster if its mean for that cluster is significantly different to its overall mean a quantitative variable characterizes a cluster if the mean of that variable is the same in each cluster we need the hierarchical tree to help characterize obtained clusters
Q2) When describing the clusters obtained using qualitative variables: a chi² test helps to see whether overall, a qualitative variable characterizes well the clustering a certain category characterizes a cluster well if it is over-represented or under-represented in that cluster if a category is very common, it will characterize many clusters well
Q3) The model individual in a cluster is located at the exact barycenter of individuals in that cluster is one of the individuals from the data set is a virtual individual is the individual in the data set which is the closest to the barycenter of individuals in that cluster can be the model individual of several clusters
Q4) When characterizing clusters obtained using factor axes, the mean of the factor dimensions is always equal to 0 an axis characterizes a cluster if the individuals in that cluster have values significantly different to 0 an axis characterizes a cluster if the individuals in that cluster are spread out across the whole axis
Score = Correct answers: