Is it possible to set different weights for each individual character in a character type experiment and calculate a similarity matrix using a weighted categorical coefficient?
This is indeed possible. First, a character information field called ‘Weight’ needs to be added in the Character type window (select Fields > Add new field). The weights can either be entered manually or imported from a text file (choose Characters > Import character information). In the Comparison window, make sure the character type experiment is highlighted in the Experiments panel and choose Clustering > Weighted categorical clustering. This action calculates a similarity matrix and dendrogram using a weighted categorical coefficient. The weight is taken from the character information field named ‘Weight’. The analysis that is created with this command will be listed with the character type name (the name cannot be changed). Once the similarity matrix is calculated, these values can be used as input data for the calculation of other trees or networks (e.g. minimum spanning tree): select Clustering > Calculate > Advanced cluster analysis, and make sure to start from the Similarity matrix and not from the Character data.