Social Network Analysis for Social Work Research

The person-in-environment perspective which guides most all social work research is ideally suited to social network analysis. Social network analysis allows one to explicitly model the dynamics of a social environment surrounding individuals. This explicit modeling of social environment provides researchers with a rich understanding of how a social context shapes the opportunities for risk and resilience, action and inaction, growth and decay for individuals in those systems. Although social network analysis has been around in sociology, communication, public health and related fields for several decades, it has yet to penetrate far in social work research.

This workshop will focus on providing participants with an overview to the kinds of analytic techniques available in social network analysis that may inform social work research. There are several specific topics: (1) We will examine the differences in ego-centric network data (data about the social world collected from individuals) and sociometric network data (data about the social world collected from a group of interconnected individuals). We will discuss techniques for collecting each type of data. (2) Participants will be exposed to basic techniques for analyzing ego-centric data with SPSS or SAS, using statistical models. Some of these measures include density, centralization, and influence modeling. (3) Participants will be exposed to inexpensive and readily available social network software (UCINET) and how to use this software to generate common social network metrics used to model social networks in the context of sociometric data. Some of these metrics include density, centrality, betweenness, k-cores, and k-plexes. (4) Some basic techniques for creating visual representations of social network data will be discussed and software which easily allows for the creation of such images (NetDraw) will be introduced. (5) We will cover basic techniques for integrating sociometric social network measures into individual-level survey data. This last topic allows one to examine how structural attributes of a social system may be associated with individual-level psycho-social outcomes.

The workshop will be a mixture of didactic components and interactive engagement with actual data. The didactic component will focus on teaching core concepts in social network methods as well as basic steps in using the software packages. The interactive component will take the form of allowing participants to work through specific examples from my own published data with software packages, looking at how social network phenomenon impact the risk and resilience of homeless youth.

Providing more scholars in social work with knowledge in this area can improve the discipline's traction on problems at a macro and micro level. I bring considerable expertise in social network analysis. I have been working in conducting research on risk-taking behaviors in social networks since 1998. I have published several pieces on the social network processes affecting the risk behaviors of homeless youth, the disclosure practices or HIV infected women, and the implementation of evidence-based practices by organizations. I am currently the PI of a NIMH-funded R01 project looking at how the social network engagement of homeless youth affects their health and well-being over time.