In science, honest error is not generally the result of ignoring data, or manufacturing it. It comes from weighing the data that supports your desired result more heavily than the data which contradicts it. It is always a subconscious process. The victim is never aware he is doing it.

This is a purely psychological phenomenon. It cannot be eliminated by either competence or integrity on the part of the researcher. Its why we have peer review and refereed publications; not because they're always right, but because they are more likely to be right.

You can't decide the truth by vote, and sometimes the maverick is right and the conventional wisdom is not.

But as a rule, he's either insane or irrational. Neither condition is usually obvious to the one who suffers from it, and it is impossible to demonstrate it to the victim. But for the observer, the ferocity with which the petitioner insists on presenting his case, and the contempt he directs toward those who dare challenge him, is persuasive enough evidence on which to base a tentative judgement.