clustering_tm: Redefined clusering coefficient for two-mode networks


This function calculates the two-mode clusering coefficient as proposed by Opsahl (2010).
Note: If you are having problems with this function (i.e., run out of memory or it being slow for simulations), there is a quicker and much more memory efficient c++ function. However, this function is not fully integrated in R, and requires a few extra steps. Send me an email to get the source-code and Windows-compiled files.


clustering_tm(net, subsample=1, seed=NULL)



A binary or weighted two-mode edgelist


Whether a only a subset of 4-paths should we used when calculating the measure. This is particularly useful when running out of memory analysing large networks. If it is set to 1, all the 4-paths are analysed. If it set to a value below one, this is roughly the proportion of 4-paths that will be analysed. If it is set to an interger greater than 1, this number of ties that form the first part of a 4-path that will be analysed. Note: The c++ functions are better as they analyse the full network.


If a subset of 4-paths is analysed, by setting this parameter, the results are reproducable.


Returns the outcome of the equation presented in the paper


version 1.0.0, taken, with permission, from package tnet


Tore Opsahl;


Opsahl, T. 2010. Triadic closure in two-mode networks: Redefining the global and local clustering coefficients. arXiv,1006.0887

Questions? Problems? Suggestions? or email at

All documentation is copyright its authors; we didn't write any of that.