The more independent experiments you conduct, particularly if you use a different experimental set-up, the more confident you can be about validating (or invalidating) a particular theory.