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

Two eyes or one? The data analyst's dilemma.

J Katz1

  • 1Wilmer Institute, Johns Hopkins Medical Institutions, Baltimore, Maryland.

Ophthalmic Surgery
|August 1, 1988
PubMed
Summary

Ophthalmic studies often analyze individual eyes, but similar eye characteristics can skew results. This research highlights the impact of eye pair correlation on statistical analysis for accurate ophthalmic data interpretation.

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

[Pain and dysfunction of the temporomandibular joint: etiology and treatment].

Harefuah·1992
Same author

Prenatal diagnosis of tetrasomy 12p by in situ hybridization: varying levels of mosaicism in different fetal tissues.

Prenatal diagnosis·1992
Same author

Use of esterase inhibitors and zone electrophoresis to define bacterial esterases in amniotic fluid.

American journal of obstetrics and gynecology·1992
Same author

Phantom limbs still a ghostly phenomenon.

CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne·1992
Same author

Impact of vitamin A supplementation on the incidence of infection in elderly nursing-home residents: a randomized controlled trial.

Age and ageing·1992
Same author

The increased risk of ulcerative keratitis among disposable soft contact lens users.

Archives of ophthalmology (Chicago, Ill. : 1960)·1992

Area of Science:

  • Ophthalmology
  • Biostatistics
  • Clinical Research

Background:

  • Ophthalmic studies frequently collect data on disease status, visual function, and ocular physiology for both eyes of each patient.
  • Ocular characteristics often exhibit high similarity between the two eyes of the same individual, despite inter-individual variability.
  • Standard statistical analyses in research commonly treat individual eyes as the primary unit of analysis, rather than the patient.

Purpose of the Study:

  • To discuss the correlation observed between pairs of eyes for various ocular measures.
  • To evaluate the impact of this eye-pair correlation on the outcomes of common statistical procedures used in ophthalmic research.
  • To inform researchers about potential statistical challenges and provide guidance for accurate data analysis.

Main Methods:

  • Review of existing literature on ocular data collection and statistical methodologies in ophthalmology.
  • Discussion of the statistical implications of correlated data, specifically focusing on paired eye measurements.
  • Exploration of how intra-individual eye similarity affects standard statistical techniques.

Main Results:

  • A significant correlation is often present between paired ocular measures within the same individual.
  • The inclusion of both eyes in analyses without accounting for correlation can lead to misleading statistical results.
  • Standard statistical methods may violate assumptions of independence when applied to paired eye data, affecting significance levels and effect size estimations.

Conclusions:

  • Researchers must acknowledge and address the correlation between paired eyes in ophthalmic studies.
  • Appropriate statistical methods that account for paired or correlated data are essential for valid interpretation of ophthalmic research findings.
  • Careful consideration of the unit of analysis (eye vs. patient) is crucial for robust study design and accurate conclusions in ophthalmology.

Related Experiment Videos