Abstract
Despite decades of network research, the crucial question, How do networks evolve? has not been sufficiently explored. The authors explore this question by analyzing the co-authorship networks in the U.S. biotechnology firms. Building on network management and network inertia perspectives, the authors build a model predicting that the structural changes in the firms' co-authorship networks are dependent on the specific characteristics of firms' initial networks, the firm's age and size. The authors then extend the model by incorporating a measure of the impact of the quality of the knowledge produced by the network ties using the prominence and inertia perspectives, which lead to the incorporation of competing hypotheses and moderating relationships in the model of scientific network evolution. The authors then test the model using longitudinal analysis of 367 U.S. biotechnology firms over a span of 17 years. The authors find that firms' existing tie-specific characteristics in the form of a firm's existing network size, tie strength, and the knowledge quality are significant determinants of network evolution, but that this influence is tempered by organizational inertia.