Skip to main navigation Skip to search Skip to main content

Correlation

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Correlation describes whether two variables ‘go together’. Correlation is a general term, covering various different kinds of association or relationship between two variables. Correlation does not imply causality. Correlations may be positive or negative, linear (looking like a straight line) or nonlinear (curved), and strong or weak. There are several different measures of correlation. Correlation measures range from +1, indicating perfect positive agreement to −1, indicating perfect negative agreement. The most appropriate measure depends on the type of measurement scale and distribution of both variables and the shape of the relationship between them.

Original languageEnglish
Title of host publicationEncyclopedia of Statistics in Behavioral Science
Subtitle of host publicationEveritt Behavioral
PublisherBlackwell
Pages1-3
Number of pages3
ISBN (Electronic)9780470013199
ISBN (Print)9780470065259
DOIs
Publication statusPublished - 1 Jan 2006

Keywords

  • association
  • bivariate data
  • correlation measures
  • linear and nonlinear relationships

Fingerprint

Dive into the research topics of 'Correlation'. Together they form a unique fingerprint.

Cite this