Reply #53: If you think that statistical "truths"

[Febble]

have "unambiguous" implications, then not only do I have a bridge to sell you, but I have a batch of scientific papers that I would be delighted for you to review.

Unfortunately most peer-reviewers tend to insist that even massively significant findings remain ambiguous. Statistical certainty is possible. Interpretational certainty is not.

And the only statistical methoology for establishing a causal relationship involves random assignment of an experimentally manipulated variable, for example drug and placebo in a double blind trial. Even then we may not be able to distinguish between proximal and distal causes.

It is nonetheless possible to use correlational statistics to test causal hypotheses. None of these, as far as I know, "unambiguous"ly point to fraud. Indeed, some point in the other direction.