Evolution of collaborative academic networks

To a substantial degree intellectual innovation occurs as a result of collaboration in interconnected research relationships. The production of knowledge is a social process involving interactions among people and organisations with different backgrounds, resources, predispositions and insights. The members of such a community interact and develop and exchange new knowledge and through these processes shape their community over time. In the last 20 years there has been an increasing amount of collaboration in research and research into collaboration. The ways that research can be fostered includes collaboration internationally (Hakala et al. 2002), across sectors, between university and industry (Anonymous 2004), between science and technology, across scientific disciplines (Forman and Markus 2005), and at individual and institutional levels. Much work has been done on the link between research productivity and research collaboration showing that more prolific researchers collaborate more frequently and researchers at all levels of productivity tend to collaborate more with highly productive researchers. Other studies have focused on patterns of spatial proximity and scientific interaction at regional and country levels (Andersson and Persson 1993). Additionally there have been studies that have considered the nature of the social networks that collaboration engenders including the social networks of scientists and more specifically the collaborative network of the IMP group. Work on the nature, role and impact of collaboration contributes to a growing literature in areas that examines the nature and characteristics of complex adaptive systems in which network structures play a fundamental role. In this empirical paper we examine the emergence of collaborative structures in academic networks using a large time-series database of collaboration patterns in academic communities. We are particularly interested both in isolating the types of collaborative structures and how they evolve and in considering the implications of these patterns for the productive output of the collaborative network. The data is supplemented by stories of collaboration narrated by key actors within collaborative groups. These stories illuminate the structural evidence in often-surprising ways, hinting at the wide variety of 'roles' and 'relationships' that may coexist within even a single collaborative grouping. The paper concludes with a discussion of the implications of this work for the production of knowledge in academic communities and beyond, as well as more general discussion of the implication for innovation. Specifically, the ideas of network competence developed by Ritter et al (2002) and others are considered.