Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Longitudinal data analysis. A comparison between generalized estimating equations and random coefficient analysis.

Jos W R Twisk1

  • 1Department of Clinical Epidemiology and Biostatistics, EMGO-institute, Vrije Universiteit medical centre (VUmc), Amsterdam, The Netherlands. jwr.Twisk@vumc.nl

European Journal of Epidemiology
|October 8, 2004
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

The effect of active vitamin D supplementation on body weight and composition: A meta-analysis of individual participant data.

Clinical nutrition (Edinburgh, Scotland)·2024
Same author

Prostate Cancer-related Events in Patients with Synchronous Metastatic Hormone-sensitive Prostate Cancer Treated with Androgen Deprivation Therapy with and Without Concurrent Radiation Therapy to the Prostate; Data from the HORRAD Trial.

European urology·2024
Same author

Microvascular flow imaging of fibroids: A prospective pilot study.

Acta obstetricia et gynecologica Scandinavica·2024
Same author

YELLOW RoUTIne prospective cohort study protocol: insight in the dynamics of bacteria in the elderly bladder.

BMC infectious diseases·2024
Same author

Neuropathological hallmarks in the post-mortem retina of neurodegenerative diseases.

Acta neuropathologica·2024
Same author

Association of rheumatoid factor, anti-citrullinated protein antibodies and shared epitope with clinical response to initial treatment in patients with early rheumatoid arthritis: data from a randomised controlled trial.

Annals of the rheumatic diseases·2024
Same journal

Re: "Does tattoo exposure increase the risk of cutaneous melanoma? A population-based case-control study".

European journal of epidemiology·2026
Same journal

Separability and identifiability as primary obstacles to substantively useful mediation analysis.

European journal of epidemiology·2026
Same journal

Intergenerational income mobility and psychotropic drug dispensation in a 1953 Stockholm cohort: a diagonal reference model approach.

European journal of epidemiology·2026
Same journal

Genetic risk score to enhance glaucoma case detection: a prospective double-blind screening study (EyeLife) in the population-based Lifelines cohort.

European journal of epidemiology·2026
Same journal

Different risks, same causes: educational attainment influences when, not from what, we die.

European journal of epidemiology·2026
Same journal

Loneliness around the world: patterns, predictors, and well-being implications.

European journal of epidemiology·2026
See all related articles

Generalized Estimating Equations (GEE) and random coefficient analysis offer comparable results for continuous longitudinal data. Dichotomous outcomes require cautious interpretation, particularly with missing data.

Area of Science:

  • Biostatistics
  • Longitudinal Data Analysis

Background:

  • Longitudinal studies involve correlated repeated measurements within individuals.
  • Specialized statistical techniques are necessary for accurate analysis.

Purpose of the Study:

  • Compare Generalized Estimating Equations (GEE) and random coefficient analysis for longitudinal data.
  • Evaluate performance with continuous and dichotomous outcomes, including missing data scenarios.

Main Methods:

  • Applied GEE and random coefficient analysis to a longitudinal dataset (n=147, 6 measurements).
  • Analyzed relationships between serum cholesterol and predictors: fitness, body fatness, smoking, gender.
  • Examined performance with simulated missing data, including selective missingness.

Main Results:

Related Experiment Videos

  • For continuous outcomes, GEE and random coefficient analysis yielded comparable results, equivalent to GEE with an exchangeable structure and random coefficient analysis with a random intercept.
  • No performance difference was observed between techniques for continuous data with missing values, even highly selective ones.
  • For dichotomous outcomes, random coefficient analysis produced larger regression coefficients and standard errors than GEE.
  • Analysis of dichotomous data with missing values yielded unpredictable results for both methods.

Conclusions:

  • GEE and random coefficient analysis are comparable for analyzing longitudinal data with continuous outcomes.
  • Longitudinal data analysis with dichotomous outcomes warrants caution, especially when missing data is present.