Alessandro Lomi and / The analysis of social networks Alessandro Lomi is a professor of Organization and Management Theory at the University of Lugano where he is a member of the Institute of Computational Science. He is a Senior Research Fellow in the School of Psychological Sciences of the University of Melbourne, and a Life Member of Clare Hall College, University of Cambridge. In the recent past, he was an ordinary member of the Swiss National Science Foundation (Social and Human Science Division). His research interests include the analysis of social networks within and between organizations and the development of statistical models for the empirical study of social and other networks. He holds a PhD from Cornell University (New York).
Viviana Amati is a postdoctoral researcher at the Social Networks Lab, ETH Zurich. She received her Ph.D. in Statistics from the University of Milano-Bicocca and has previously worked as a postdoctoral researcher at the University of Konstanz. Her primary research interest is statistical analysis and modelling of dynamic networks with a focus on estimation and misspecification of stochastic models for relational data.
Data typically collected in the social sciences are based on the familiar case-by-variable research design, where "cases" (rows) represent various kinds of social actors, and "variables" (columns) contain measurements on a set of attributes of the actors or their context. Quantitative research based on this design typically adopts methods that emphasize relations among the "variables." Social network research, by contrast, focuses on relations among the "cases" by examining the social structure in which individual action is embedded. The methodological and substantive scope of social network research, therefore, is very general and extends to a wide range of social actors including individuals, organizations, sectors, and states. Against the backdrop of these general considerations, the course starts by introducing the basic theoretical and conceptual background of social network research, the fundamental ideas underlying the network approach, and discusses its many domains of empirical application. The course then proceeds to examine the basic analytical concepts needed to describe and understand the structure of social networks across various levels of analysis. Participants will learn how to visualize social network data to discover their main structural features, and how to implement different types of network research designs and approaches to data collection. The course goes on to illustrate contemporary statistical models for social networks, so that participants may learn how to test hypotheses using network data. Exponential Random Graphs models (ERGMs) and Stochastic Actor-oriented Models (SAOMs) will be introduced as examples of statistical models for studying network structure and connective behavior. The course will include practical examples and hands-on computer laboratories based on the analysis of real-life relational data. In the laboratories, the emphasis will be on the analysis of social networks in structured social and economic settings such as, for example, business companies, and other formal organizations like hospitals, universities and other educational institutions. Students will also be given the opportunity to work with their own data and consult with the instructors about their own research work.
The software packages used include the UCINET, PNet and RSIENA. The software used in the course is all publicly available. Depending on the interests of the participants, other software resources developed in the R environment may also be adopted.
Participants taking this course are expected to be familiar with the basic concepts of descriptive statistics and have an active interest in statistical inference.