An R Package for Generalized and Classical Blockmodeling of Valued Networks
Main page

Basic description
The package blockmodeling is primarily meant as an implementation of Generalized blockmodeling for valued networks. It aslo supports multirelational and multilevel blockmodeling. In addition, functions for computation of (dis)similarities in terms of structural and regular equivalence, plotting and other "utility" functions are provided.

If you find any bugs, please report to: email: ales{dot}ziberna{at}gmail{dot}com


The version of the package available on CRAN is very old and is not really useful for generalzed blockmodeling (it is OK for indirect approaches). In this version, some functions are extremely slow. They are practically unusable for networks with more than 30 units and 4 groups (AMD Athlon 1600 XP-M). These are: opt.these.par, opt.random.par, check.these.par (if there is a very large number of partitions). However, sum of squares homogeneity blockmodeling is relatively fast (also when used with above functions) when used with complete blocks only (structural equivalence). Networks with more than 100 units (in each more for two-mode data) can be analyzed in reasonable time (few hours).

The latest version (only available on Rforge) however supports parallel computing and includes functions written in C (mainly optParC, optParMultiC and optRandomParC) which are therefore much faster. Using them analysis of networks with approximately 200 units is feasible (more if you are patient). However, as these have no help-files yet available, they are not included in the CRAN version. Please e-mail me if you need help using them.


The lates version (which does includes the latest upgrades and bug fixes) is available on R-Forge repository. The package here is build during the night. There the package is available in 2 formats: Package source (.tar.gz) and Windows binary (.zip).
To install the package from R-Forge directly within R type:

Version 0.1.8 is available on CRAN. However, this version is very old and does not include patches, upgrades or bug fixes. This version of the blockmodeling package can be therefore instaled using install.packages using the CRAN repository by typing within R:

Some old versions are also available here, however mostly only as Package source and Windows binary.


Aleš Žiberna
assistant professor
Faculty of Social Sciences
University of Ljubljana
email: ales{dot}ziberna{at}gmail{dot}com
Web page at Faculty of Social Sciences


DOREIAN, Patrick, BATAGELJ, Vladimir, FERLIGOJ, Anuška. Generalized blockmodeling, (Structural analysis in the social sciences, 25). Cambridge [etc.]: Cambridge University Press, 2005. XV, 384 p., ISBN 0-521-84085-6.

WHITE, Douglas R., Karl P. REITZ (1983): Graph and semigroup homomorphisms on networks of relations. Social Networks, 5, p.193-234.

WHITE, Douglas R.(2005): REGGE (web page). (12.5.2005).

ŽIBERNA, Aleš (2007): Generalized Blockmodeling of Valued Networks. Social Networks, 29, p. 105-126.[]. Author accepted manuscript.

ŽIBERNA, Aleš (2007): Generalized Blockmodeling of Valued Networks. Doctoral dissertation. PDF (A typo on page 39 corrected on October 22, 2010)

ŽIBERNA, Aleš (2008): Direct and indirect approaches to blockmodeling of valued networks in terms of regular equivalence. Journal of Mathematical Sociology, vol. 32, no. 1, p. 57-84. []

ŽIBERNA, Aleš (2009): Evaluation of Direct and Indirect Blockmodeling of Regular Equivalence in Valued Networks by Simulations. Metodološki zvezki 6(2), 99–134. []

ŽIBERNA, Aleš (2013): Generalized Blockmodeling of Sparse Networks. Metodološki zvezki 10(2) (use this link to find supplementary materials), 99–119. []

ŽIBERNA, Aleš (2014): Blockmodeling of Multilevel Networks. Social Networks 39 (October): 46–61.[]. arXiv - Author accepted manuscript ŽIBERNA, Aleš, LAZEGA, Emmanuel (2016): Role Sets and Division of Work at Two Levels of Collective Agency: The Case of Blockmodeling a Multilevel (Inter-Individual and Inter-Organizational) Network. In Multilevel Network Analysis for the Social Sciences, edited by Emmanuel Lazega and Tom A. B. Snijders, 173–209. Methodos Series 12. Springer International Publishing. [].

email: ales{dot}ziberna{at}gmail{dot}com