The authors link High-Involvement Work Systems (HIWS) with social network research both theoretically and methodologically by conceiving of these work systems as networks and by using network methods to better understand and evaluate these arrangements. Their approach is to integrate the institutional perspective of industrial relations with the analytical methodology of social network analysis. They use a longi- tudinal data set collected before and after the introduction of an HIWS in a pharmaceutical company to measure the impact on patterns of employee communication as well as communication between the union and management. Improvements were found in customer satisfaction, and positive effects were seen in the pattern and structure of organizational communication with an increase in the density of lateral and vertical interdepartmental communication, and a decrease in hierarchy. These results were confirmed using survey data on employee perceptions of particular types of communication and interviews with employees at various levels of authority.