This is such a pervasive problem. Here we are years later and we have yet another scientist clinging to old data when they have been proven wrong.
I think part of early science education needs to be the concept that you can never prove theories right, only ever prove them wrong and that's it's ok that things are that way. People hate this concept, because they like building rigid belief systems where they know for sure the things they believe are right and never have to consider alternate possibilities. It's ok to do that, as long as you know how to spot cases when you are wrong and are willing to re-think things when they occur. Some people are unwilling to do even that.
The prime example of this should be Newtonian physics' most basic formula, Force = Mass * Acceleration. It's been tested and experimentally verified for hundreds of years by millions of people, and it's wrong. It always was, and always will be wrong. It is so close to right it is still used extensively, but only when we know the margin of error it introduces is negligible, and we know how to calculate that margin of error now so we know when it will be small enough to ignore and when it won't. There is no proving things right, there is only knowing that things have been right so far. The second the real world proves you wrong, you are wrong.