The article continues to present the results of the analysis of collaboration networks of Russian sociologists in 2010–2021. It was conducted on the basis of data on co-authorship of scientific articles indexed in the electronic library eLibrary (75,232 scientific publications on sociology). The methodology of bibliometric network analysis implies the construction of several types of collaboration networks. The resulting networks include 37,790 unique authors. In the first part of the article, based on the analysis of the publication activity of the authors, current trends in co-authorship (macro-level) are defined and the most “visible” representatives of the sociological community (micro-level) are highlighted. The second part of the article provides a meso-level analysis that helps to characterize the structure of the collaboration and identify groups of collaborating authors. There are many isolated groups of authors – more than 4 thousand groups ranging in size from 2 to 100+ nodes, as a rule, these are dyads and triads. However, a third of all authors belong to the main connected component and can reach each other. The members of this group form a unified network of co-authorship based not only on direct but also indirect relationships and have access to collective social capital and common community resources. The selected groups are very diverse in terms of their structure; the article provides examples of groups allocated for different reasons. Further research may include examining collaboration at the organizational level, exploring the global structure of connections using blockmodeling, and studying the network dynamics in a longitudinal perspective.