Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Correlation does not imply causation
open-in-new
<p>The phrase "correlation does not imply causation" refers to the inability to legitimately deduce a <a href="/facts/Causality/Swh496X7">cause-and-effect</a> relationship between two events or <a href="/facts/Variable_(research)/vJm09hAC">variables</a> solely on the basis of an observed association or <a href="/facts/Correlation/egkluAEm">correlation</a> between them. The idea that "correlation implies causation" is an example of a <a href="/facts/Questionable_cause/8HA55aht">questionable-cause</a> <a href="/facts/Fallacy/IRVuLSef">logical fallacy</a>, in which two events occurring together are taken to have established a cause-and-effect relationship. This fallacy is also known by the Latin phrase <i>cum hoc ergo propter hoc</i> ('with this, therefore because of this'). This differs from the fallacy known as <i><a href="/facts/Post_hoc_ergo_propter_hoc/R7wtNthH">post hoc ergo propter hoc</a></i> ("after this, therefore because of this"), in which an event following another is seen as a <a href="/facts/Logical_consequence/AyDXp4dn">necessary consequence</a> of the former event, and from <a href="/facts/Conflation/ru3nDY96">conflation</a>, the errant merging of two events, ideas, databases, etc., into one. </p><p>As with any logical fallacy, identifying that the reasoning behind an argument is flawed <a href="/facts/Argument_from_fallacy/Q06DlXtd">does not necessarily imply</a> that the resulting conclusion is false. <a href="/facts/Statistics/DNPsKGYU">Statistical</a> methods have been proposed that use correlation as the basis for <a href="/facts/Statistical_hypothesis_testing/yv9RPF6U">hypothesis tests</a> for causality, including the <a href="/facts/Granger_causality/LAEvgA1A">Granger causality test</a> and <a href="/facts/Convergent_cross_mapping/usdIuYok">convergent cross mapping</a>. The <a href="/facts/Bradford_Hill_criteria/jYUM1Wyy">Bradford Hill criteria</a>, also known as Hill's criteria for causation, are a group of nine principles that can be useful in establishing epidemiologic evidence of a causal relationship. </p>