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

Assessing the significance of the correlation between two spatial processes.

P Clifford1, S Richardson, D Hémon

  • 1Mathematical Institute, University of Oxford, England.

Biometrics
|March 1, 1989
PubMed
Summary

New statistical tests account for spatial autocorrelation in data, reducing effective sample size for positively correlated processes. This method improves association analysis for lattice and nonlattice data, demonstrated in geographical epidemiology.

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

A quality improvement study: Optimizing pneumococcal vaccination rates in children with cochlear implants.

Vaccine·2022
Same author

Statistical properties of sketching algorithms.

Biometrika·2022
Same author

The molecular landscape and associated clinical experience in infant medulloblastoma: prognostic significance of second-generation subtypes.

Neuropathology and applied neurobiology·2020
Same author

Fast nonadiabatic dynamics of many-body quantum systems.

Science advances·2019
Same author

Time to cut the cord: recognizing and addressing the imbalance of DOHaD research towards the study of maternal pregnancy exposures - CORRIGENDUM.

Journal of developmental origins of health and disease·2019
Same author

Is there a role for pentoxifylline and tocopherol in the management of advanced osteoradionecrosis of the jaws with pathological fractures? Case reports and review of the literature.

International journal of oral and maxillofacial surgery·2019

Area of Science:

  • Statistics
  • Spatial Analysis
  • Epidemiology

Background:

  • Spatial autocorrelation is common in various data types, affecting statistical inference.
  • Traditional association tests often assume independence, which is violated by spatial autocorrelation.
  • Accurate analysis requires methods that account for spatial structures.

Purpose of the Study:

  • To develop modified statistical tests for association between spatially autocorrelated processes.
  • To introduce a method for evaluating the reduction in effective sample size due to spatial structure.
  • To assess the performance of these tests using simulations and real-world data.

Main Methods:

  • Development of modified tests using correlation coefficients or covariance.
  • Evaluation of effective sample size considering spatial autocorrelation.

Related Experiment Videos

  • Approximation of the variance of the correlation coefficient to quantify sample size reduction.
  • Monte Carlo simulations to assess test performance.

    • Main Results:

    • The proposed tests are applicable to both lattice and nonlattice spatial data.
    • Positively autocorrelated processes lead to a reduced effective sample size.
    • A method was developed to approximate this reduction.
    • Simulations demonstrated the effectiveness of the modified tests.

    Conclusions:

    • Modified association tests effectively handle spatial autocorrelation.
    • The concept of effective sample size is crucial for accurate spatial data analysis.
    • The method provides a practical approach for analyzing geographical epidemiological data.