The objective of this research is to demonstrate that measures of social structure can be used to enhance participant involvement in system change. The research concerns the application of graph theory based software to the measurement of network aspects of social capital in communities and in non-profit organisations, and with evaluating the short-term outcomes of specific organisational interventions. The epistemological framework being used in this research project is action research. Action research is real world or applied research as opposed to psychological research that might take place in a laboratory or other controlled environment. In general it is regarded as a qualitative rather than a quantitative research method. As a principle the people potentially affected by a social change intervention are directly involved in the action research. Action research does challenge traditional epistemological views of what constitutes legitimate science. In conventional science the expert researcher stays within the boundaries of their own discipline, follows strict scientific method setting up a hypothesis, developing an approved methodology, gathering and analysing data, producing a report and hoping that someone else will act on this expert information. This is not so in action research. In terms of the epistemological validity of action research, the process addresses themes rather than hypotheses and the role of the researcher is intimately involved in the change process. The learning that is developed is about both content and process thought of as an epistemological framework (F), transformed into a methodology (M) to research an area of concern (A), resulting in learning about F, M and A. In the action research process the researcher is more of a facilitator of the learning process than an expert. The use of techniques to identify social structures and the possibility of simulating proposed interventions using data generated by judgements of likely outcomes, raises the possibility of designing minimal interventions (micro-technology) in terms of cost and negative social impact, that may be able to deliver planned development of social capital. Furthermore, if network analysis software had the capacity to identify “unpredictable” adverse consequences of interventions in social systems, the potential might also exist for the development of planning tools and strategies to respond accordingly. The steams of theory and practice drawn upon included: Public health research on the social determinants of health (and other social outcomes) and community cohesion as a feature of these social determinants. The concept of social capital which stresses the importance of networks of connections involving trust and reciprocity between individuals in groups (bonding social capital), and across to other community groups and organisations (bridging social capital), and the way in which individuals and groups access resources and decision-making (linking social capital). Social capital is said to be stored in the networks of reciprocal and trusting interactions. Systems theory in which the access to, consumption and redistribution of resources (energy) is a fundamental process. Secondly, the concepts of dynamic equilibrium and constant disequilibrium are important with respect to “unpredictable” consequences of planned change. Systems theory is also important in terms of the definition of boundaries between groups and how mutual relationships form between groups within the same environment. Social network theory. Social network theory is a consolidation of anthropological, social psychological and sociological theorems concerning the structure of human relationships in large groups and organizations, and quantitative techniques to measure and describe these relationships. Social relationships, boundaries between groups and movement of resources can all be depicted in sociograms (diagrams of social interaction between people) as long as the numbers are small i.e. around 8-10 people. Sociograms are impractical when numbers are larger than this because sheer complexity makes them difficult to interpret. Combinational explosion of possible connections make them difficult to analyse. Thus large sociograms can only be “intuitively” analysed and the time required to produce diagrams by hand makes them cost prohibitive. Given the appropriate formulation of questions and the development of survey methodology, it appeared reasonable to conclude that network software could be used to analyse and graph relationships of trust and reciprocity between individuals within a group (bonding social capital), between groups (bridging social capital) and between individuals and groups in terms of their access to resources and decision making (linking social capital). The research hypothesis was that measures of social structure could be used to enhance participant involvement in system change designed to develop social capital, by enabling the development of innovative interventions directed at minimum disruption and maximum outcome for the organisation or community. The research methodology was action research based on applied case studies where knowledge and experience gained each time was applied in subsequent case studies. Tools and measures were modified as a result in a continuous process of case study, review of processes and results, reflection, and modification of subsequent applications. Case studies varied in size, scope and context each case study testing one or more elements of a general community/organisational change process. While this did not result in a single organisational intervention across the entire process, it did allow for testing the hypothesis in a range of contexts demonstrating that it is possible to develop tools to measure and describe network aspects of social capital and to build social capital in communities, community and other not-for-profit organisations. Theoretical and measurement issues in network aspects of social capital are reviewed and suggestions for further research and developments highlighted, are discussed.