Every persistent genealogical researcher eventually reaches that point in their research where they have to confront the existence of mistakes and errors, in their own records and those of others.
One example is the record for my Grandfather, LeRoy Parkinson Tanner and his family in the U.S. Federal Census record for 1930. After searching in the indexes for his name, I did not find him in the census. However, this was an unreasonable conclusion. I might have had a different attitude had I no idea where he was living, but I knew where he was living in 1930. So, I went through the Census record, page by page. Sure enough, there he was. His last name, Tanner, had been transcribed as "Tamer" in the indexes, just enough difference to make it seem like he was not there.
Not all mistakes are that easy to avoid. Sometimes the only information you can find is not accurate. It may take years to find additional information to correct the initial problems caused by the inaccuracy. I have repeated my basic premise, that is, to doubt all information from the beginning as the way to proceed with all research. Always be ready to find contradictory information and have the patience to accept the fact that the contradictions may never be resolved.
Here is an example of an unresolvable contradiction. My 3rd Great-Grandmother's name is reported variously as Roxanna Ray or Roxalanna Ray, depending on the source. Various of my relatives have transcribed the name as follows:
Also Known As
All of these variations are presently listed in New.FamilySearch.org and FamilySearch Family Tree. Interestingly, her marriage record in Vermont lists her name as Roxy.
Additional research may finally resolve this issue or it may not. If the source information shows multiple names with different spellings, this would not be unusual nor cause any problems with future research. But it is unfortunate that so many researchers have simply copied and re-copied the records without trying to verify the information. I view contradictions and errors in the records, like I would a large rock in a field; I just plow around it. The key is to not abandon the field simply because it has a rock. In other words, don't give up the research because of the errors, do your research in spite of the errors.