A step-by-step guide to generalized estimating equations using SPSS in the field of dentistry

TL;DR

This paper provides a comprehensive, step-by-step guide to applying Generalized Estimating Equations (GEE) in dentistry research using SPSS, focusing on correlated data analysis from multiple patient outcomes.

Contribution

It introduces the basic concepts of GEE, explains covariance and correlation matrices, and demonstrates model selection using QIC and QICu criteria in the context of dentistry.

Findings

GEE effectively handles correlated dental data.

QIC and QICu assist in selecting optimal models.

The guide facilitates practical application of GEE in SPSS.

Abstract

The Generalized Estimating Equations (GEE) approach is a widely used statistical method for analyzing longitudinal data and clustered data in clinic studies. In dentistry, due to multiple outcomes obtained from one patient, the outcomes produced from individual patients are correlated with one another. This study focuses on the basic ideas of GEE and introduces the types of covariance matrix and working correlation matrix. The quasi-likelihood information criterion(QIC) and quasi-likelihood information criterion approximation(QICu) were used to select the best working matrix and the best fitting model for the correlated outcomes.