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

Cluster effects and simultaneity in multilevel models

R Blundell1, F Windmeijer

  • 1Department of Economics, University College London, UK.

Health Economics
|July 1, 1997
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

Predictors of post-thyroidectomy hypocalcaemia: a systematic and narrative review.

The Journal of laryngology and otology·2020
Same author

The Oxylipin Signaling Pathway Is Required for Increased Aphid Attraction and Retention on Virus-Infected Plants.

Journal of chemical ecology·2020
Same author

Bilateral Thyroid Carcinosarcoma in a Cat.

Journal of comparative pathology·2019
Same author

IRC +10 216 in 3-D: morphology of a TP-AGB star envelope.

Astronomy and astrophysics·2018
Same author

Detection and Isolation of Digital Dermatitis Treponemes from Bovine Pressure Sores.

Journal of comparative pathology·2016
Same author

The association between contact and intellectual disability literacy, causal attributions and stigma.

Journal of intellectual disability research : JIDR·2015
Same journal

Health on the Move: The Impact of Poverty Alleviation Relocation on Healthcare Utilization in China.

Health economics·2026
Same journal

The Effects of Compulsory Licensing: A Case Study of HIV Drugs.

Health economics·2026
Same journal

Beyond Tobacco Prevention: The Effects of Tobacco 21 Laws on Young Adults' Body Weight.

Health economics·2026
Same journal

Assessing the Estimands and Estimates of Hospitalization Rates in Health Economics and Clinical Medicine.

Health economics·2026
Same journal

The Impact of Unemployment Insurance Benefit Cuts on Mental Health: Evidence From Early Pandemic Program Expirations.

Health economics·2026
Same journal

Do Patients Value the Service Provided by Physicians Who Overbill? A Willingness-to-Pay Study Using French Survey Data.

Health economics·2026
See all related articles

For small group sizes, the Generalized Least Squares (GLS) estimator in multilevel models is biased. A fixed effects approach offers consistent slope parameter estimates, especially when cluster effects correlate with regressors.

Area of Science:

  • Econometrics
  • Biostatistics
  • Health Services Research

Background:

  • Multilevel models are widely used in various scientific fields.
  • The Generalized Least Squares (GLS) estimator is a common method in these models.
  • Bias and inconsistency issues arise with GLS when cluster effects correlate with regressors, particularly in small sample sizes.

Purpose of the Study:

  • To investigate the bias and inconsistency of the GLS estimator in multilevel models with small group sizes.
  • To compare the performance of GLS with a fixed effects approach under specific conditions.
  • To apply these estimation methods to real-world healthcare utilization data.

Main Methods:

  • Comparison of Generalized Least Squares (GLS) and fixed effects estimators.

Related Experiment Videos

  • Analysis of estimator performance based on group size and correlation between cluster effects and regressors.
  • Application of estimators to UK acute care hospital utilization data, accounting for health authority district effects.

    • Main Results:

    • The GLS estimator is biased and inconsistent for small group sizes when random cluster effects are correlated with regressors.
    • A fixed effects approach provides consistent slope parameter estimates under these conditions.
    • GLS and fixed effects estimators yield equivalent results for large group sizes.

    Conclusions:

    • Fixed effects methods are preferable to GLS for multilevel models with small group sizes and correlated cluster effects.
    • The choice of estimator significantly impacts the reliability of results in multilevel modeling.
    • Findings have implications for analyzing healthcare utilization and other complex data structures.