Thank you for your reference to that link. Having now seen it I do recall that controversy albeit not in that level of detail (such work isn't my profession).
I know from practical experience that most of the points Fischer et al have made are often valid or the case including those about producing 'tailored' results for a specific target audience. Often one comes across most of them in one's own work and it's usually when one is tying to figure out whether one's actually deluding oneself given, say, that a set of results looks too good. Alternatively, one may have a dataset that's full of garbage and not yet realized it.
Then there are always those cases where most of the data fits within the expected distribution except for some wayward measurement or two, which if included, would throw the stats and make the overall experiential results look less precise. What does one do, ditch out the odd/atypical results or rerun the whole experiment again in the hope that those large wayward deviations don't reappear? Anyone who has done experiments is overly familiar with the problems of processing experiential data—as they say, they're just part of the territory.
Clearly, I've no hands-on expertise in Mendel's work so I can only make general comment which is that as Mendel is no longer around to defend himself or his work, as such we must be particularly careful about accusing him of doing something of which he may not have done. From that Wiki, despite the controversy, it seems that the passage of time has meant that Mendel's work seems to have been examined with reasonable rigor and care given that it's been re-examined and reanalyzed on multiple occasions by different personnel. In the absence of better and more accurate data there's little more we could reasonably do.
I suppose I'm overly sensitive to accusations of fraud because of an incident that happened eons ago back in my student days. This was when a chemistry tutor blatantly accused me of cheating in my write-up of an experiment and he did so in the presence of other students. What he objected to was that my experimental results (dataset) of a titration were too good to be true and that I'd worked back and substituted data from the theoretical/ideal model.
This wasn't the case at all as I had recorded the titration's dataset exactly as I'd measured it. Needless to say, I was furious at his unfounded accusation and I demanded that he say back over lunchtime and watch me whilst I reran the experiment, this he did. The results I obtained on the rerun were just marginally better than my first attempt.
I lay no claim to being an experiential genius—absolutely far from it. I'll just say this: some people are intrinsically much better at experimental work than are others and those who doubt their results are perhaps basing their doubt on their own less-than-optimal performance. That said, one's always right to doubt any claim or sets of results until they're independently verified.
