We all know the truth of the old adage "Correlation is not Causation."

Except when sometimes we don't because sometimes we all know that correlation IS causation.

Like, the time you blew off studying for the next day 8 am calc test to go to the frat party where you proceeded to out drink the entire pledge class all night.

In a scientific sense, causality must meet specific criteria. The most widely accepted of these are commonly referred to as Hill's Criteria for Causation (alternately called Bradford Hill Causation) and were established by Sir Austin Bradford Hill, British epidemiologist, in 1965. The criteria are as follows:

Temporality--the presumed effect occurs after the presumed cause

Strength--a statistical calculation of the likeliness of association between two variables (there are a number of such calculations depending on the relationship and data in question)

Reproducibility (or Consistency)--consistent results by different researchers using different methodology

Specificity--this relates to the elimination of variables by connecting other likely explanations

Biological gradient--a relationship between cause (such as exposure to a particular chemical) and effect (a greater exposure leads to a proportional or inversely proportional result); in epidemiological studies this is generally called the dose-response relationship

Plausibility--likeliness of the proposed mechanism between the presumed cause and effect

Coherence--epidemiological and laboratory data are in agreement; the association is consistent with current theory and knowledge

Experiment--the ability to purposely/purposefully create or ameliorate the effect as a result of the presumed cause (or preventing the presumed cause)

Analogy--whether or not it follows similar relationships

Hill's Criteria is a process of critically thinking about the relationship between variables to establish causality. They are not definitive, nor