I noticed something interesting recently, while trying to get a better grasp on who's who along one branch of my family tree (it felt like trying to differentiate between the Ptolemies with how often "William Barnett" came up).
Double checking census data and cross referencing death/birth/marriage records, I kept fixating on how often the reported birth locations changed between, my 2xGreat and 3x grandparents.
Ordinarily, I'd just chalk that up to the individual reporting said information for another person, and the former not knowing the exact details. It happens-- whatever. Same with misspellings, or mistaking a letter written in cursive.
I feel kind of dense, but when I finally started to scrutinize the ages in reference to census year for individuals that were giving me issues-- sure enough I found a concrete discrepancy.
For example: [barnett-203] and the new [Barnett-11786]. The former, Mr. William Theodore: For the longest time, I had thought him to be my great-great grandfather. Afterall, the dates were accurate enough, the locations? Lined up with what I knew, and bonus (and perhaps my folly) there was already information and connections to who I knew definitely to be related to me.
Turns out, it wasn't that simple. Not only was age just a bit off, but marriage data didn't line up. What I thought was a simple misreporting was, or likely is, different people entirely. It's just coincidence that their names are almost identical, and their general locations, professions are pretty much the same, also.
I could still be wrong about this new discovery-- hell, it'd probably be a lot easier to be wrong. There's definitely more information about one William than the other.