Message from @MOS MAIORVM (AKA Sola)

Discord ID: 311373428013924352


2017-05-09 05:22:33 UTC  

This sickness in science extends beyond medicine

2017-05-09 05:23:12 UTC  

The last real breakthroughs in physics must have been like 1920

2017-05-09 05:23:15 UTC  

Or earlier

2017-05-09 05:23:39 UTC  

@fallot it was a 10 year study as I said

2017-05-09 05:23:45 UTC  

That's quite a long time

2017-05-09 05:23:59 UTC  

there may be more factors to it unaccounted for but the premise is definitely eye opening

2017-05-09 05:24:41 UTC  

If you study something that could take 10 years to show its effect you need to study it for much longer than that. Thank you for making me aware of it at least.

2017-05-09 05:24:56 UTC  

I'm open to the idea, though I'd of course like to see more research

2017-05-09 05:25:08 UTC  

Be careful with that

2017-05-09 05:25:22 UTC  

2 years ago I would have said similarly

2017-05-09 05:25:37 UTC  

I think the point was that a 10 year study is far more reliable than a 1 year study

2017-05-09 05:25:45 UTC  

it's similar to sample size

2017-05-09 05:25:46 UTC  

Is it?

2017-05-09 05:25:54 UTC  

Another semi fallacy

2017-05-09 05:25:59 UTC  

All these things

2017-05-09 05:26:07 UTC  

how is that a semi fallacy?

2017-05-09 05:26:30 UTC  

Are you proposing that a 2 week study is as reliable as a 10 year study under the same conditions?

2017-05-09 05:26:35 UTC  

Because ultimately the validity does not depend on how big or blind your study is

2017-05-09 05:26:36 UTC  

I find that doubtful

2017-05-09 05:26:46 UTC  

Hence semi

2017-05-09 05:26:50 UTC  

That's not true at all

2017-05-09 05:26:51 UTC  

Re what you said

2017-05-09 05:27:03 UTC  

It is trivially true

2017-05-09 05:27:12 UTC  

Its validity depends on your priors

2017-05-09 05:27:25 UTC  

The mechanics of what you study

2017-05-09 05:27:29 UTC  

I'm not debating the validity of whether cheese causes cancer

2017-05-09 05:27:42 UTC  

Sure. I mean in general too.

2017-05-09 05:27:47 UTC  

I'm debating the validity of the results of any given study based on length and size

2017-05-09 05:28:07 UTC  

It is well established that such factors are important

2017-05-09 05:28:18 UTC  

Part of the scientific method if I recall properly

2017-05-09 05:28:34 UTC  

Not as important as your average person considers them

2017-05-09 05:28:54 UTC  

And how to do you come to that conclusion or measure such a thing?

2017-05-09 05:29:11 UTC  

Look

2017-05-09 05:29:17 UTC  

I'm not saying a long study makes it valid

2017-05-09 05:29:25 UTC  

but a long study under the same given conditions is superior to a short study

2017-05-09 05:29:28 UTC  

By observing the overall quality of science in terms of measurable impact or real insight

2017-05-09 05:29:35 UTC  

And again I say

2017-05-09 05:29:42 UTC  

Not necessarily

2017-05-09 05:29:51 UTC  

Long enough is the best one can say

2017-05-09 05:30:09 UTC  

Long enough is subjective

2017-05-09 05:30:24 UTC  

To some degree yes