Talk:Correlation does not imply causation

Latest comment: 27 days ago by 2603:7000:6039:D397:5BC:EE9D:25E1:A5E6 in topic The title is just incorrect.

Untitled edit

ATTENTION: This page was moved after a vote at Talk:Correlation implies causation/Page title.

Can we put in a PSA that Correlation does imply causation, just not necessarily so? edit

If I had a dime every time someone told me that correlation does not equal causation as a means to dismiss correlation as evidence... Perhaps a section labelled: "Improper use as an argument" ?

Akiva.avraham (talk) 16:47, 15 March 2017 (UTC)Reply

Only if a reliable, published source makes that distinction. Wikipedia does not publish original thought. See Talk:Correlation does not imply causation/Archive 2 § Original research, below. —Sangdeboeuf (talk) 21:06, 29 March 2017 (UTC)Reply
Correlation does imply causation, but it doesn't mean causation. I think replacing 'imply' in the title by 'mean' would be better. Nederlandse Leeuw (talk) 01:15, 25 April 2017 (UTC)Reply
An implication in this context is the propositional sense of an implication, see Implicational propositional calculus.GliderMaven (talk) 03:47, 25 April 2017 (UTC)Reply
Yes, but encyclopædia article titles are expected to be in natural language, not propositional logic. Michael R Bax (talk) 20:59, 16 November 2017 (UTC)Reply
The two are not mutually exclusive, as here, where the natural language seems to have incorporated a propositional logic term.GliderMaven (talk) 21:33, 16 November 2017 (UTC)Reply
A better title could be "Correlation is not causation". Karlpoppery (talk) 23:47, 25 April 2017 (UTC)Reply
* "Correlation does not show causation"
* "Correlation does not necessitate causation"
* "Correlation does not prove causation"
* "Correlation does not demonstrate causation"
Michael R Bax (talk) 20:59, 16 November 2017 (UTC)Reply

Is it possible to prove causation, if correlation is negative sometimes? (talk) 09:48, 15 June 2017 (UTC)Reply

The title is fine if one uses the term "imply" according to standard use in the field of logic (wikt:imply verb use 1 and 2), whereas a natural language use (wikt:imply verb use 3) isn't right. The title "Correlation is not causation" has other problems, because the nature of 'equivalence' here is unclear. Given this, I suggest no page move. Klbrain (talk) 17:08, 28 June 2017 (UTC)Reply
Yes, but encyclopædia article titles are expected to be in natural language, not propositional logic. As such this title is misleading. In fact I was sent to this page by someone arguing that (in general rather than in logic) correlation does not imply causation! Michael R Bax (talk) 20:59, 16 November 2017 (UTC)Reply

Oxford Living Dictionaries: "imply, verb: … 1.1 (of a fact or occurrence) suggest (something) as a logical consequence." Using this definition, correlation does imply causation. Michael R Bax (talk) 20:59, 16 November 2017 (UTC)Reply

The reason that Michael R. Bax is 100% wrong here is simply that the phrase "correlation implies causation" can be interpreted in multiple ways. There are many ways to express the important point here clearly in natural language and without being ambiguous. The phrase "correlation implies causation" is not one of them. (talk) 19:57, 4 December 2020 (UTC)Reply
When a correlation and a causation correlate it is as much a coincidence as any correlation is to any causative effect.
It cannot be reworded to the satisfaction of all without someone inventing a new term to encompass the point, probably sticking something in front of imply in such a way as to make it not emulate standard conversation. Everybody would then have to agree on that term. Also, English isn't the only language, despite the irony of English (or more properly USA-speak) having become science's standard Lingua Franca...
The article currently makes the point about the technical use of the word 'imply' in this article.
In fact correlation never implies causation. It can seem to imply causation, it can appear to suggest causation. It does not state that any causative factor is implicit upon any correlation.
One dirty great big problem for use lay folk when it comes to statistics is that it is full of technical terms that in all probability possess a likelihood that on average said statistical term correlates in spelling and pronunciation with a non-technical term (in fact is the same word) which has a more general and sometimes more varied meaning, or even not having the same meaning at all. Further, even the standard use may not having the same meaning colloquially. It can be implied that someone is a liar but that does not mean the same thing as documentation providing implicit evidence of lying.
Correlation does not imply causation, it may suggest it but even that suggestion can be more subjective than objective. (talk) 14:58, 8 October 2023 (UTC)Reply

Extremely confusing sentence edit

In general I try to avoid removing citations, in favor of trying to rework problematic cited material. However, here I am removing a citation; this is an explanation of why I did.

The problematic content is the 'sentence' "The widely held (but mistaken) belief that RCTs provide stronger causal evidence than observational studies, the latter continued to consistently show benefits and subsequent analyses and follow-up studies have demonstrated a significant benefit for CHD risk in healthy women initiating oestrogen therapy soon after the onset of menopause.[5]", added in this edit: (May 2020 by User:Wikicize).

This sentence is problematic for (at least) two reasons:

  • It's grammatically incorrect (the sentence structure is not well-formed)
  • It's opinion-based due to the phrase "widely held (but mistaken) belief" which is presented without justification (the previous sentence sources an article that describes a "disparity between findings from observational studies and RCT" which "has created considerable debate among researchers").

Since the linked article ( is also linked from the article sourced in the previous sentence (, it seems reasonable to simply remove the link entirely instead of developing some workaround to maintain it. --Timothy.lucas.jaeger (talk) 01:01, 19 December 2021 (UTC)Reply

After further review, I was mistaken in thinking the reference I was planning to remove was linked from the other sourced article. I will need to rethink my plan of attack here. If someone takes a stab at handling this situation, I would also recommend moving this entire HRT example back to the opening summary from whence it came, as it makes no sense in its current section which is about the meaning of the work 'implies'. This move was done here: (May 2020 by User:Vsmith). --Timothy.lucas.jaeger (talk) 01:10, 19 December 2021 (UTC)Reply

I decided to simply remove this example entirely. It's hardly a poster child for correlation not implying causation, since there is dispute in the scientific community over whether the observed correlated data or randomly-controlled-trial causal data is actually the correct data to go by (assuming I have any understanding of what any of these cited papers are saying). --Timothy.lucas.jaeger (talk) 01:23, 19 December 2021 (UTC)Reply

Causation does not imply correlation, either. edit

The article currently states:

Where there is causation, there is correlation, but also a sequence in time from cause to effect, a plausible mechanism, and sometimes common and intermediate causes. While correlation is often used when inferring causation because it is a necessary condition, it is not a sufficient condition

Saying among other things that causation implies correlation. However, I don't think this is true.

Firstly, see . But secondly, consider this real world example:

I am driving a car on the motorway. The speed limit is 70mph, and I want to get where I'm going as fast as possible, so I'm sticking to the speed limit exactly. The motorway is hilly. When the car is going uphill I press harder on the accelerator to stay at 70, and when the car is going downhill I ease off so that I don't go too fast.

In an analysis of foot-angle vs vehicle speed, you would find no correlation between the two variables, since vehicle speed is constant even though foot-angle varies. However, there is an obvious causal relationship between foot-angle and vehicle speed - the foot-angle controls the accelerator, and the accelerator controls the speed.

This also applies to other control systems, such as room temperature vs AC activity - the AC will aim to maintain a constant temperature, such that you'll get a graph with varying AC activity and a flat room temperature, showing no correlation between AC and temperature, despite obvious causality.

Should I delete the quoted section? FalacerSelene (talk) 20:45, 10 April 2022 (UTC)Reply

No. In your example, there is a confounding variable which cancels a correlation that would otherwise be there. That is a very special case. Also, it is WP:OR, and we do not deleted sourced content because of OR. --Hob Gadling (talk) 05:49, 11 April 2022 (UTC)Reply

Thanks for your explanation. Here we go:

This wasn't an original thought to me, and I've traced it back to these links. Also, the section that I quoted doesn't have a conflicting citation - it is unsourced. FalacerSelene (talk) 08:42, 11 April 2022 (UTC)Reply

The title is just incorrect. edit

It does imply causation, it just doesn't prove it. Do you have any logical reason to dispute that? --fs 09:37, 8 June 2022 (UTC)Reply

See Correlation_does_not_imply_causation#Usage,_and_meaning_of_'imply'. Vpab15 (talk) 10:54, 8 June 2022 (UTC)Reply
And where does this article make it clear that it's purely a Statistics article? fs 08:30, 11 June 2022 (UTC)Reply
Why should it? Is it? --Hob Gadling (talk) 09:06, 11 June 2022 (UTC)Reply
"Dinosaur illiteracy and extinction may be correlated, but that would not mean the variables had a causal relationship." Examples like this use inconsistent language. Its strange to use "does not mean" and other colloquialisms after it's pointed out that can create confusion. It would be more consistent to say "Dinosaur illiteracy and extinction may be correlated, but that does not prove the variables had a causal relationship." and almost every example given would be clearer if they simply used "proved" or a synonym of 'proof' as used in a statistical context rather than present conclusions with colloquial language. Some of the examples presented feel like original ideas because their arguments are presented without citations or at face value. For example:
"Third factor C (the common-causal variable) causes both A and B" example A talks about drunkenness as 'an actual cause' when Alcohol is not even mentioned in the example presented. A lot of the third factor examples go against the provided section and should be made more consistent with the point of clarification. 2603:7000:6039:D397:5BC:EE9D:25E1:A5E6 (talk) 21:48, 27 March 2024 (UTC)Reply

Merge proposal edit

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
To not merge, on the grounds that the two articles have different purposes and readerships, and that hence readers are best served by having them separate. Manage the sad panda problem by removing content forking at Correlation does not imply causation. Klbrain (talk) 22:46, 27 December 2023 (UTC)Reply

Whole sections have been copy/pasted, i.e. forked. That makes me a saaaad panda.

{{merge|Causal analysis|date=July 2023}} (talk) 02:44, 14 July 2023 (UTC)Reply

Unfortunate. Nevertheless the concept is inherently significant and better placed within its own context instead of hidden away as a subset of causal analysis, or worse mixed into the general causal analysis article in such a way as to lose its discrete point.
Equally you could argue that it be merged as a subsection under pseudoscience as that is where the most significant occurrence of it occurs in practical terms.
An argument could equally be made to have it become a subsection under statistical analysis, and for all I know it may well be covered in that (or a similarly named) entry.
I do not know the wikipedia jargon for the forwarding hyperlinks which in text for would be italised 'see etc' or some acronym for a bit of latin in parantheses, however the density in wikipedia is just about right, it is easy to get lost in technical discussions if continual page/subject diversions occur.
It is not as if wikipedia articles on topics or people mentioned within another article do not often contain repeat information, seemingly often verbatim, sometimes a little paraphrased. In recent times I personally have noticed an increase in grammar breakdowns where sentences have something missing or simply do not make sense or it is not certain whether the sentence is saying yes or no anymore. This is a sign of copy pasting partial sentences into other sentences instead of using a new sentence, in other words doing proper editing.
If it is becoming hard for native English speakers to follow what hope have people for whom English is a second language?
Most of the discussion on this wikipedia topic bangs on about the symantics of the word 'imply' for no real reason, rewording it to Causation is not implict upon Correlation might work better than the often context dependent word 'imply'. However, most of the argument seems to be that correlation can imply causation.
This is something of a misunderstanding. Correlation correlating with causative processes is coincidence. It is within other contexts that connection is further suggested and via some form of methodological investigation that any underlying causality established and can be said to be the reason for the correlation having occurred. Possibly this may exit on a somewhat instinctive level as blatantly unconnected correlations are usually as plainly obvious as silly syllogisms or alternatively do not appear to suggest any linkage at all whilst linkage may appear later given extra outside information. (talk) 14:38, 8 October 2023 (UTC)Reply
I wouldn't be in favor of merging, as I think these are sufficiently different subjects to merit separate articles. Maybe a better option would be to remove the duplicate material and make each article reference the other. Dezaxa (talk) 14:41, 9 October 2023 (UTC)Reply
From a higher education industry perspective, this Article may probably be one of the most commonly introduced to learners of all statistics-related pages. Its monthly viewing is even more than correlation coefficient per the tool for page views. If you go to Google Scholar, you may find even a manuscript cited on this page although the quality of that is extremely questionable: Knoll, B. (2016). Youth Suicide Rates and Mormon Religious Context. Dialogue: A Journal of Mormon Thought, 49(2), 25-26. But I believe it's not the only one linked to here. I also believe there would be professors introducing this page to their students for further self-learning. It is NOT a good practice to consider merging without consideration of its context concerning educational functionality and viewing count. -- (Dasze) 10:06, 22 December 2023 (UTC)Reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I've helped to reduce the fork problem through the use of an excerpt template. Klbrain (talk) 22:46, 27 December 2023 (UTC)Reply