Subject specific and population average models for binary longitudinal data: a tutorial

Authors

  • Camille Szmaragd School of Clinical Veterinary Science, University of Bristol
  • Paul Clarke Centre for Market and Public Organisation, University of Bristol
  • Fiona Steele Graduate School of Education, University of Bristol

DOI:

https://doi.org/10.14301/llcs.v4i2.249

Keywords:

Autocorrelation, British Household Panel Survey, hierarchical models, logistic regression, marginal models, mixed effects models, random effects models, repeated-measures analysis

Abstract

Using data from the British Household Panel Survey, we illustrate how longitudinal repeated measures of binary outcomes are analysed using population average and subject specific logistic regression models.  We show how the autocorrelation found in longitudinal data is accounted for by both approaches, and why, in contrast to linear models for continuous outcomes, the parameters of population average and subject specific models for binary outcomes are different.  To illustrate these points, we fit different models to our data set using both approaches, and compare and contrast the results obtained.  Finally, we use our example to provide some guidance on how to choose between the two approaches.

Author Biographies

Camille Szmaragd, School of Clinical Veterinary Science, University of Bristol

Research Associate, School of Clinical Veterinary Science, University of Bristol

Paul Clarke, Centre for Market and Public Organisation, University of Bristol

Senior Lecturer, Centre for Market and Public Organisation, University of Bristol

Fiona Steele, Graduate School of Education, University of Bristol

Professor of Social Statistics, Graduate School of Education, University of Bristol

Downloads

Published

2013-05-22