In the case of free sorting data, a list of dissimilarity matrices (the length of the list is equal to the number of subjects).
In the case of multiple sorting, dissimilarity matrix for a subject is the sum of the dissimilarity matrices computed from each of the different partitions given by this subject.