top | item 36799646

(no title)

cowoder | 2 years ago

What would make correlation research worthless? Wouldn't that just be a first step to see if researching causation is even worth it?

discuss

order

ethanbond|2 years ago

This is pretty much a canned response from LatteLazy. It ignores the fundamental fact that science has never “proved causation” of anything ever. All we have is lower and higher confidence correlations and that’s all we’ll ever have.

LatteLazy|2 years ago

I'll take any confidence at all. This has a zero percent confidence because causation is fundamentally different to correlation. And showing causation is pretty much the whole point of experiments in science. Hence, this is not even science really. Just advertising...

Bostonian|2 years ago

I would not call study worthless, but to establish causation one could do an experiment where elderly adults are given (1) an opportunity for a free hearing test and hearing aids if needed (2) the equivalent amount of cash. Then follow up the two groups to see if group 1 is less likely to develop dementia. My guess is that the results will not be strong.

LatteLazy|2 years ago

It becomes worthless when we never take the second step.

It is worthless because why not just jump straight to step 2?

The answer of course is that that would be expensive and hard and might not prove anything. Much safer and cheaper to NOT actually test you hypothesis and you can still use it to up your publication figure and your funder can still use it to sell hearing aids etc...

ethanbond|2 years ago

> It is worthless because why not just jump straight to step 2?

Can you explain what this means?