On: A French love affair... with graphology

[brazzy]

> And just because we cannot measure its success rate using mathematics or statistics - that doesn't mean it is not a valid tool.

Wrong. It means exactly that.

[bnegreve → brazzy]

Usually when something is not statistically relevant you can also prove it. I am not sure how you would do it in this case.

It probably works to some limited extents: for example, I am sure they can guess rather accurately whether the candidate is male or female.

So I do see a difference between a non valid tool and a tool whose efficiency is hard to evaluate.

Now, don't get me wrong, I wouldn't use graphology.