Statistical Modelling

This group focuses on the development and application of statistical models to investigate substantive phenomena in the social sciences and beyond.

We are also interested in the continuing software developments for fitting statistical models.

As well as doing research that involves statistical models, we also run training courses and seminars to explain how the latest software developments allow us to fit statistical models to answer substantive research questions. These include advice on the preparation of data and the interpretation of results.

We are primarily based in Social Statistics, but we are also involved in interdisciplinary collaboration at the University of Manchester, as well as working with colleagues elsewhere in the UK, and overseas.

Recent research topics include the development and application of the following models:

  • Multilevel (hierarchical) Models for analysing data that have hierarchical, clustered, or cross-classified structure.
  • Bayesian hierarchical models for integrating traditional and new forms of data.
  • Longitudinal Models.
  • Models for Social Networks, such as Exponential Random Graph Models (ERGMs) for the analysis of social network structures.
  • Structural Equation Models.