>>14265811It's very hard to prove a causal relationship. In fact there is no ''standard'' way to make a definitive proof as for almost all causal relationships you can always add argumentation and discussion for why due to factor x, y, z its not an exact causal observation.
With causal variables you often see a lot of consistency in the distribution of data and how one dataset relates to the other. Usually it needs to be shown in multiple experiment in various conditions to make a strong argument for causation.
>>14265876You can always continuously argue. Especially in the social sciences which the redditor loves. Therefore any highly likely causal relation that appears in these sciences but goes against the redditor his beliefs will be dismissed and labeled as ''fake science''. An argument using existing literature can also be typically readily be found in the social sciences to quote for the argumentation provided, but the redditor rarely goes to such efforts.
For plenty of things in medical sciences a causal relationship is assumed where you could also argue that a causal relationship doesn't exist. This depends on the current scientific consensus, patient reports of treatments and so on. Sometimes a causal relationship that was assumed turned out to be false, leading to different insights and different treatment methods.
Good example of this would be knee injuries in older people. Most older people have a degenerated meniscus with tears. Tears in meniscus are highly correlated with pain. But immediately sawing the torn parts of the meniscus off does not directly lead to an alleviation of pain symptoms in this group.
