This paper focuses on online party affiliation from the perspective of Social Network Analysis (SNA). It examines the structure of the information channels of politicians on Twitter in order to map their online social network, relying on data gathered for all elected representatives in the municipal councils and executives of the Brussels Region (Belgium). The paper adds a temporal dimension by looking at the development of the network over time, and its determinants. The paper starts by analyzing and categorizing the patterns in the structure of the information circulation, and its evolution over time based on three criteria: centrality, density, and development. It then investigates the factors that predict the evolution of these patterns over time. More specifically, we test whether the structural development of the online networks are related to the socialization role of parties, local coalition governments, and local vs. regional politics. Our results support all three hypotheses, but show that party affiliation is a stronger determinant of the structural development of online communities. ; info:eu-repo/semantics/nonPublished