Group Differences in Generalized Linear Models

  • Tim F. LiaoEmail author
Part of the Handbooks of Sociology and Social Research book series (HSSR)


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.


Generalize Linear Model Link Function Gamma Distribution Current Population Survey Exponential Family 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. Allison, P. D. (1999). Comparing logit and probit coefficients across groups. Sociological Methods & Research, 28, 186–208.CrossRefGoogle Scholar
  2. Blinder, A. S. (1973). Wage discrimination: Reduced form and structural estimates. The Journal of Human Resources, 8, 436–455.CrossRefGoogle Scholar
  3. Brée, D. S., Hogeveen, H., Schakenraad, M. H. W., Scheinemakers, J. F., & Michael Tepp, D. (1995). Conditional causal modeling. Applied Artificial Intelligence, 9, 181–212.CrossRefGoogle Scholar
  4. Chen, Y., Bressler, S. L., & Ding, M. (2006). Frequency decomposition of conditional Granger causality and application to multivariate neural field potential data. Journal of Neuroscience Methods, 150, 228–237.CrossRefGoogle Scholar
  5. Chow, G. C. (1960). Tests of equality between sets of coefficients in two linear regressions. Econometrica, 28, 591–605.CrossRefGoogle Scholar
  6. Czado, C., & Raftery, A. E. (2006). Choosing the link function and accounting for link uncertainty in generalized linear models using Bayes Factors. Statistical Papers, 47, 419–442.Google Scholar
  7. Davis, J. A. (1987). The logic of causal order. Thousand Oaks: Sage.Google Scholar
  8. Liao, T. F. (1994). Interpreting probability models: Logit, probit, and other generalized linear models. Thousand Oaks: Sage.Google Scholar
  9. Liao, T. F. (2002). Statistical group comparison. New York: Wiley-Interscience.CrossRefGoogle Scholar
  10. Liao, T. F. (2004). Comparing social groups: Wald statistics for testing equality among multiple logit models. International Journal of Comparative Sociology, 45, 3–16.CrossRefGoogle Scholar
  11. Lieberson, S. (1985). Making it count: The improvement of social research and theory. Berkeley: University of California Press.Google Scholar
  12. McCullagh, P., & Nelder, J. A. (1989). Generalized linear models (2nd ed.). London: Chapman & Hall.Google Scholar
  13. Oaxaca, R. (1973). Male-female wage differentials in urban labor markets. International Economic Review, 14, 693–709.CrossRefGoogle Scholar
  14. Oaxaca, R. L., & Ransom, M. (1998). Calculation of approximate variances for wage decomposition differentials. Journal of Economic and Social Measurement, 24, 55–61.Google Scholar
  15. Thomas, M. E. (1993). Race, class, and personal income: An empirical test of the declining significance of race thesis, 1968–1988. Social Problems, 40, 328–342.CrossRefGoogle Scholar
  16. Western, B., & Rosenfeld, J. (2011). Unions, norms, and the rise in U.S. wage inequality. American Sociological Review, 76, 513–537.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.Department of SociologyUniversity of Illinois Urbana-ChampaignChampaignUSA

Personalised recommendations