See also: Causality
Everybody knows that it's important to distinguish causation from association (correlation). Causal beliefs allow you to make prescriptions. If you are the sort of person who pays attention to this, then reading popular science (specifically clinical or behavioral studies) can be an exercise in patience.
I believe that much of the confusion is due to language, and would like to propose a pair of prescriptivist Thesaurus entries, designed to have zero overlap:
Words that imply or suggest a causal relationship in a specific direction
- "A causes B" (typically A,B events)
- "A leads to B" (typically A,B events)
- "A affects B" (typically B variable)
- "A has an effect on B" (same)
- "a high value of A is good for you" (typically A variable; B is implicitly considered "good")
- "A influences B" (typically B variable)
- action words, e.g. "being part of a club, even if it meets only once a month, increases your happiness as much as doubling your salary" or "joining a group that meets even just once a month produces the same happiness gain as doubling your income" source
- "has an impact" e.g. "poverty had the greatest negative impact on health." (a claim that is very hard to verify without strong assumptions)
- "The last 3 times volcano A erupted, it triggered an eruption of volcano B"
- "intergenerational transmission of divorce" http://www.bakadesuyo.com/if-your-parents-divorced-are-you-more-likely
Words that are neutral (a much smaller set)
- "A predicts B"
- "eruptions of A have been followed by eruptions of B".
- "often coincide with"
A linguist would probably explain that our persistent overuse of words loaded with causality is due to action words (which express causality) being more numerous and (therefore) more expressive than associational words... because actions are at core of why humans communicate at all.
suggestive word order
Andrew C Thomas writes:
<< So here's a modest proposal: when possible, beat back the causal assumption by presenting an associational idea in the order least likely to be given a causal interpretation by a layperson or radio host. Trying it this way, a random Google News headline reads: "Prolonged Use of Pacifier Linked to Speech Problems" and strongly implies a cause and effect relationship, despite the (weak) disclaimer from the quoted authors. Reverse that and you've got "Speech Problems linked to Prolonged Use of Pacifier" which is less insinuating, at least to me. >>