Group Differences in Generalized Linear Models

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

This chapter deals with making comparisons between fixed groups in the framework of generalized linear models. First, we briefly introduce generalized linear models, the most common type of regression models. Next, we discuss a simple system for analyzing group differences in regression. We primarily focus on two types of comparisons—analyzing differences in the parameter vectors of the linear predictor and differences in the underlying distributions for the groups in the model. To illustrate such comparative methods for group differences, we perform analyses using real-world data. Theoretically, group differences in regression estimates can be viewed as an example of conditional causality. Practically, testing group differences in regression may see many useful applications in social science research.

Original languageEnglish (US)
Title of host publicationHandbook of Causal Analysis for Social Research
EditorsStephen L Morgan
PublisherSpringer
Chapter9
Pages153-166
Number of pages14
ISBN (Electronic)978-94-007-6094-3
ISBN (Print)978-94-017-9407-7, 978-94-007-6093-6
DOIs
StatePublished - Mar 27 2013

Publication series

NameHandbooks of Sociology and Social Research
ISSN (Print)1389-6903
ISSN (Electronic)2542-839X

Keywords

  • Generalize Linear Model
  • Exponential Family
  • Current Population Survey
  • Gamma Distribution
  • Link Function

ASJC Scopus subject areas

  • Psychology (miscellaneous)
  • Social Psychology
  • Social Sciences (miscellaneous)
  • Sociology and Political Science

Fingerprint

Dive into the research topics of 'Group Differences in Generalized Linear Models'. Together they form a unique fingerprint.

Cite this