Merged Naive empiricism vs skeptical empiricism

[jt512]

jt512, tell me how they can run multiple tests of a singleton fossil, or how many geologic maps are necessary before they become considered acceptable, and then we'll talk. Until then, this is nothing but what I call Physics Envy: the desire to force science as such to conform with the standards of one field, usually physics (hence the name). Unfortunately for you and your ilk, real science simply isn't done that way. But hey, I'm willing to be proven wrong--answer those two questions in the first sentence, or either one of them really, and I'll consider your idea.

What would it take for you to consider the notion that you're wrong? Would anything? If nothing would be sufficient, can you REALLY claim to be rational about this?

What would it take for you to stop caricaturing my position?

And it was you who brought up physics..

 

[Dinwar]

What would it take for you to stop caricaturing my position?

I'm not. A singleton fossil and a geologic map are scientific hypotheses, every bit as much as any medical study or physics hypothesis. Both are, however, accepted without replication--ie, only one fossil of a singleton species needs to be found in order to be accepted as a new species (that's what singleton means), and only one geologic map needs to be produced to be accepted by the scientific community. This directly contradicts the idea that a scientific study must be replicated in order to be acceptable, which is precisely what you stated earlier (I won't bother to quote it; if you argue against that point I'm simply going to stop responding to you all together, as it's trivially obvious to anyone who reads the thread).

If you wish to ammend your argument to be limited to specific branches of science, that's fine. If you wish to ammend it to deal with specific types of hypotheses, tests, and statements, that's fine too. Either are perfectly honest and legitimate ways to deal with this issue. But so long as you wish to discuss science as a whole you are obliged to address these issues, or to take the firm stand that historical sciences and geology are not sciences at all. You've painted yourself into the corner; I'm giving you ways out, but "That's not what I said!" isn't one of them.

It's your choice. But you don't get to not choose in this debate, because that choice is the very heart of this debate.

 

[Dinwar]

As for the physics, you're confusing an argument that illustrates a point with one that proports to proove it. I clearly was doing the former--in fact, I stated as much, explicitely. Please read my posts if you're goinig to get snarky.

 

[jt512]

If you wish to ammend your argument to be limited to specific branches of science, that's fine. If you wish to ammend it to deal with specific types of hypotheses, tests, and statements, that's fine too.

You seem to read every other sentence I write. I have made it clear in two separate posts that the need for replication ultimately stems from statistical inference, and in particular from frequentist significance testing. For instance, I wrote:

All I am doing is pointing out the fact that the scientific practices of some fields have resulted in those fields publishing a great deal of invalid results. One reason for this is their reliance on frequentist significance testing, which does not permit inferences to be made about the truth of a hypothesis from a single experiment alone. Frequentist significance testing requires replication.

The correct interpretation of a small p-value is that either the null hypothesis is false or an unusually extreme dataset was observed by random chance, but there is no way to distinguish between the two possibilities from the results of a single study. Ruling out the possibility that the dataset was unusual requires replication–and before you go on again assuming that I'm defining replication in the narrowest possible sense, please recall that I have stated several times that I don't mean that.

I don't know how to make it any clearer that the need for replication is an issue of drawing valid statistical inferences, especially from frequentist significance tests. If you're in a field where statistical inference is not required to draw valid conclusions—in other words, where there is no random sampling—then you can't have bad random samples that require replication to rule out.