> I thought the modularity score is already a comparison to a null model?
I think that 1) the modularity score is not a rigorous statistical test that gives you a p-value (not that the method I proposed is any more rigorous ;)) and 2) the modularity score assesses the significance of a community structure as a whole while the methods discussed here assess the significance of a _single_ community. Best, T. _______________________________________________ igraph-help mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/igraph-help
