Social network analysis
Theme Co-ordinator: James Hargreaves
Social network analysis (SNA) enables the empirical study of systems of social relationships. Modern SNA enables the visualisation and quantitative description of social ties between actors – called ‘nodes’ in SNA – that make up a network (for some examples have a look at http://tinyurl.com/bf2gc4). It can be used to examine the influence of these ties and the attributes of those to whom an individual of interest (‘ego’ in SNA) is connected (ego’s ‘alters’) on, for example, their health, attitudes and behaviour. By extending analysis beyond an individual to their position within the network and the attributes of alters, SNA offers the potential for new insights into the distribution of health and behaviour through a population. Local network (or egonet) studies collect information from individuals on those to whom they are directly connected, without the need to identify these individuals. Complete network studies identify individuals to whom research participants are connected, and then map connections among all individuals in the “bounded” study population. These latter studies allow the generation of empirical measures of network structure such as density (the proportion of all potential social ties that actually are present in the network), and of the social network position of nodes that might influence their behaviour or health, such as in/out degree (the number of ties an individual receives/nominates) and measures of centrality (which capture the extent to which a node is positioned in the centre of a network).
Work on social networks at LSHTM is in its early stages but there is growing interest in relation to a wide variety of topics. An occasional seminar series has been running for the last couple of years. Current topics of interest include social and sexual networks and how these are associated with infectious disease spread, immunisation behaviour, social support and drug and alcohol use among adolescents.