Previous Chapter Chapter 40: Generalized Linear Models Next Chapter
Chapter 40: Generalized Linear Models
Generalized linear models (GLMs) were proposed by Nelder and Wedderburn (1972) and represent a family of statistical techniques that can be used to analyse a wide variety of research problems. They are sufficiently general to be applicable to much social science data and provide a comprehensive set of analytical tools. Of particular importance is the unified theoretical framework that enables certain ‘economies of scale’ to be realised (e.g. the interpretation of the parameter [Page 133]estimates and model-fit statistics are similar for all techniques under the GLM umbrella, as are many of the model-building techniques and model diagnostics) and enable a full set of analyses to be taught within the confines of a typical postgraduate statistics course.
GLMs are univariate models as ...