Many scholars have explored ways to identify the positions and roles of companies in a patent citation network but their analyses lack structure. Accordingly, this study proposes a structured approach to identify these positions and roles in a network by integrating social network and multivariate analysis. First, an adjacency matrix is constructed based on the graph theory to indicate the correlation between collected data. The next step is to conduct the network analysis and compute the statistics of network centrality. Then, the principal component analysis was made to break down these statistics to a few principal components. These selected principal components are then used as cluster variables for a two-step cluster analysis. Hierarchical cluster analysis was first made to determine the proper number of clusters and then K-means clustering was used for dividing actors into k proper positions. In addition, the multivariate analysis of variance (MANOVA) is conducted to test the significance between those positions. After that, a new adjacency matrix was built upon the rearrangement of k positions. The frequency within and between these positions is then computed and the cut-off value is determined to distinguish the difference between these frequencies. Finally, each position will be labeled based on its characteristics and the relationship within and between these positions. After the structured approach is constructed, the litigation-related network of smartphone makers will be used as empirical evidence. The results show that this structured approach can effectively distinguish the position and role of a company in a network.
