Chasing shadows: Defining network boundaries in qualitative social network analysis

Defining network boundaries is a key challenge in social network analysis. In our recent qualitative study of network influences on educational decision-making - based on interviews with 107 individuals from 16 case study networks - the set of members with whom interviews were secured in each case represented only a sub-set of the broader networks from which they were drawn. Following an introduction to our study and an outline of our approach, we consider some of the processes of filtering and selection that affected the specific composition of our network sample, and reflect upon what this tells us about the processes by which participants in network-based research make decisions about the representation of their networks within research contexts. We then explore the question of whether the partiality of our data actually matters, and conclude that it reflects the permeable, partial and dynamic nature of social networks, characteristics which are central to qualitatively-informed understandings of SNA. © 2009 SAGE Publications.