Anyone can fudge data to get a result that objective evidence doesn't support.
The more self-confident scientists might actually be more likely to fudge data - they can self-justify by thinking "this idea must be true, so even if I put fake data in now, it doesn't matter because eventually real data will emerge to support it"
"this idea must be true" - this is the opposite of the type of person I'm talking about. There are some people who are able to completely detach their ideas from themselves, never losing confidence in themselves even if they lose confidence in one of their ideas.
The more self-confident scientists might actually be more likely to fudge data - they can self-justify by thinking "this idea must be true, so even if I put fake data in now, it doesn't matter because eventually real data will emerge to support it"