[lutusp]

A quote from the article: "We have evolved for it, to the extent that those surrounded by a tight-knit group of friends who regularly gather to eat—and, crucially, gossip—live an average of 15 years longer than loners."

A = alive 15 years longer than the average.

B = meets friends and gossips.

The claim, made by a psychologist (of course), fails to take into account the fact that correlation doesn't equal causation. It may be that some unexamined factor C causes people to (B) meet and gossip, and (A) live longer as well. Meaning before we change our lifestyle with the expectation of a longer life, we should first do some actual science.

[groby_b]

And before we roll out Internet tropes about correlation and causation, maybe we should first read the underlying studies.

The article fails horribly by not quoting any of them, but here's a good starting point: http://www.encyclopedia.com/doc/1G2-3402200381.html

Or, if you want a list from the guy who did the initial investigations into the phenomenon, start here: http://www.psc.isr.umich.edu/people/profile/478/James_S_Hous...

I'm far from an expert on the topic, but from what I read, a causal relationship is at the least very strongly implied.

[lutusp]

> I'm far from an expert on the topic, but from what I read, a causal relationship is at the least very strongly implied.

Scientists don't imply cause-effect relationships, they demonstrate them, using evidence, to the exclusion of alternatives. Then they offer an explanation -- a theory about the evidence. These properties are rarely present in psychological studies, especially the explanation requirement.

Here's an example. Let's say I'm a doctor who believes he has cured the common cold. My cure is to shake a dried gourd over the patient until he's all better. My cure always works -- it's perfectly reliable, even though it sometimes takes a week. So, where's my Nobel Prize? I've met the same evidence requirement the linked article does, and the latter is being described as science.

> ... the guy who did the initial investigations into the phenomenon ...

It's not a phenomenon, it's an observation, one without any effort to explain it or demonstrate a cause-effect relationship.

> And before we roll out Internet tropes about correlation and causation, maybe we should first read the underlying studies.

After reading the original work, one is left with the same impression the linked article provides -- a description without an explanation, and no effort to meet the evidence required to move from a correlation to a cause-effect relationship.

"The first principle is that you must not fool yourself and you are the easiest person to fool." -- Richard P. Feynman

[groby_b]

> Scientists don't imply cause-effect relationships, they demonstrate them, using evidence, to the exclusion of alternatives.

You mean like Einstein did with the theory of relativity? Oh. Wait. That was only proven 30+ years later by Ives–Stilwell.

You mean like String Theory is completely proven to the exclusion of alternatives? Oh. Wait. It isn't.

Science observes, formulates hypotheses, and then falsifies or shows them to be true. It is never "to the exclusion of alternatives", or we'd be done with science by now. It is about forming the model that best fits the world, given our current knowledge.

> I've met the same evidence requirement the linked article does, and the latter is being described as science.

The "linked artice" is journalism, not science. Try following up on the links in there, to the actual science.

> one without any effort to explain it or demonstrate a cause-effect relationship.

You didn't read any of the papers, did you?

> After reading the original work, one is left with the same impression the linked article provides

Ah. You digested 35 years of research, and found no evidence? Or you skimmed the overview I provided, and chose to not further investigate?

Yes, there is no clear mechanism, yet. That's because we don't understand psychology well enough to always define clear mechanisms. That's what science is about - furthering our understanding.

The work on establishing actual causal pathways is still going on, and probably will for quite a while. The metastudies indicate that research is doing a decent job to control for other factors, and still reproduces nicely.

[lutusp]

> You mean like Einstein did with the theory of relativity? Oh. Wait. That was only proven 30+ years later by Ives–Stilwell.

You're registering agreement with my point. Or didn't you notice? No one accepted relativity until it was confirmed by experiment.

> You mean like String Theory is completely proven to the exclusion of alternatives? Oh. Wait. It isn't.

Notwithstanding its popular name, string theory is not a scientific theory, it's a conjecture without an empirical basis.

> Science observes, formulates hypotheses, and then falsifies or shows them to be true.

Yes to the first, no to the second. Science never proves anything true, only false. This idea was perhaps best expressed by philosopher David Hume, who said, "No amount of observations of white swans can allow the inference that all swans are white, but the observation of a single black swan is sufficient to refute that conclusion."

> The "linked artice" is journalism, not science.

So science journalism is free from the responsibilities of science itself? If that were true, it wouldn't be either possible or responsible to call it science journalism.

> Ah. You digested 35 years of research, and found no evidence? Or you skimmed the overview I provided, and chose to not further investigate?

There is no demonstrated cause-effect relationship, it's a correlation. If there were evidence for a cause-effect relaitonship, it would have been included even in the popular accounts.

> You didn't read any of the papers, did you?

This is what you think constitutes scientific debate? If the evidence existed, you would be linking to it rather than arguing in bad faith.

> Yes, there is no clear mechanism, yet. That's because we don't understand psychology well enough to always define clear mechanisms. That's what science is about - furthering our understanding.

Psychology isn't science, and until it tries to explain what it has until now merely described, that status won't change.

> The metastudies indicate that research is doing a decent job to control for other factors, and still reproduces nicely.

If reproducing unexplained results constituted science, astrology would be a science.