A step-by-step guide to Generalized Estimating Equations using SPSS in dental research

Lim,  Hoi-Jeong; Park,  Su-Hyeon

J Korean Dent Assoc > Volume 54(11); 2016 > Article

The Journal of The Korean Dental Association 2016;54(11):850-864.
Published online November 30, 2016.

치의학 분야에서 SPSS를 이용한 일반화 추정방정식의 단계별 안내

임회정¹, 박수현²

^¹전남대학교 치의학전문대학원 치과교정학교실, 치의학 연구소
^²전남대학교 자연과학대학 통계학과

A step-by-step guide to Generalized Estimating Equations using SPSS in dental research

Hoi-Jeong Lim¹, Su-Hyeon Park²

¹Department of Orthodontics, Chonnam National University School of Dentistry
²Dental Science Research Institute, Chonnam National University

Abstract

The Generalized Estimating Equations (GEE) approach is a widely used statistical method for analyzing longitudinal data and clustered data in clinical studies. In dentistry, due to multiple outcomes obtained from one patient, the outcomes produced from an individual patient are correlated with one another. This study focused on the basic ideas of GEE and introduced the types of covariance matrix and working correlation matrix. The quasi-likelihood information criterion (QIC) and quasi-likelihood information criterion approximation (QIC_u) were used to select the best working correlation matrix and the best fitting model for the correlated outcomes. The purpose of this study is to show a detailed process for the GEE analysis using SPSS software along with an orthodontic miniscrew example, and to help understand how to use GEE analysis in dental research.

Key Words: Generalized Estimating Equations;clustered data;dental research;