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

[Regression and correlation].

H M Johnsen1

  • 1Avdeling for klinisk kjemi, Regionsykehuset, Trondheim.

Nordisk Medicin
|January 1, 1990
PubMed
Summary
This summary is machine-generated.

Simple linear regression and correlation are often confused in medical research. Using regression for association when both variables are random is misguided; maximum likelihood estimation is the correct approach.

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 disease panorama of emergency surgery in Maniitsoq 1991-1992. An epidemiological study and an evaluation of surgical results].

Ugeskrift for laeger·1994
Same author

[Confidence intervals].

Nordisk medicin·1992
Same author

Ultrasound in the diagnosis of congenital dysplasia and dislocation of the hip joints in children older than two years.

Clinical orthopaedics and related research·1991
Same author

Ectopic pregnancy in Greenland. An epidemiological study.

Arctic medical research·1990
Same author

[Stochastic variable, expectation and variance--basic concepts in statistical analysis].

Nordisk medicin·1987
Same author

Protein concentration and oncotic pressure of interstitial fluid collected by wick technique.

Bibliotheca anatomica·1975

Area of Science:

  • Biostatistics
  • Medical Research Methodology

Context:

  • Simple linear regression and correlation are widely applied in medical literature.
  • Confusion between these statistical methods is prevalent due to similar calculations.

Purpose:

  • To clarify the distinct roles of simple linear regression and correlation in data analysis.
  • To highlight the inappropriate use of regression for assessing associations between two random variables.

Summary:

  • Simple regression models a non-random regressor's effect on a response variable for prediction.
  • Correlation measures the linear association between two random variables, serving as an investigative tool.
  • When both variables are random, least squares estimation is unsuitable; maximum likelihood estimation for linear structural models is recommended.

Related Experiment Videos

Impact:

  • Promotes accurate statistical application in medical publications.
  • Discourages the misuse of regression and correlation, improving the validity of research findings.
  • Guides researchers toward appropriate statistical methods for analyzing relationships between variables.