As mentioned earlier, I do family history research, and many of my people come from poor rural farming backgrounds. It wasn't uncommon for people to get married at 18 and have 10 kids, every two or three years like clockwork, with the last being born when the mom was over 40. I think biologically if you start early and you're good at it and you do it regularly, it's not the same kind of "big deal" to have a child at 44, even in the 1920s, as it would be today for someone having their first
child at 44. Plus, I suspect the standards for a good life were unfortunately lower back then as well.
Anyway, these huge families completely mess up the concept of "generations." There could easily be 20 or 25 years between the oldest and youngest child. Most likely the eldest children will be married and having kids of their own while their mom is still having her youngest kids. Plus everyone tends to marry their relatives (because that's who's around in their small rural area with bad roads) so you get couples like "second cousins once removed"--the "removed" refers to different generations relative to their common ancestor. For example, there's my first cousins, and their
children are my first cousins, once removed. (My kids would be second cousins to those kids.) At least that's my understanding of it.
Now, in my family, amazingly enough the majority of babies in these big families tended to survive to older childhood at least, and be caught on a census or remembered by the other children. But infant/young child mortality was so high in some places, that a five-year gap between kids might indicate two or three babies who died very young.
And although I'm sure the parents remember those children in their hearts, the official documents *I* have access to could be spottier, and if the surviving children were very young themselves, they may not have much memory of those babies to pass on either. That's what *I* tend to first suspect when I see a big gap between kids--kids in between who died young; or, one parent was away from home for a while or had a long illness, neither of which might be caught in the records.
I think in all my research I've only seen one case where a modern researcher suspected a young "child" was actually a "grandchild." What I find much more in my tree is grandchildren living with grandparents, as fully acknowledged on the census records, and then I have to try and figure out what child they belonged to, and where that child is now instead of with their own kid. One of my direct ancestors had something like six daughters and then finally a son, and then one day there's a grandson living with them, with their last name, only their son is not old enough to have fathered him. Must have been one of the girls who had the child out of wedlock. I figured out which one it was by checking who wasn't
married when the child was born--and it turned out she was now married with new kids living just down the street. But her oldest child out of wedlock (presumably with a different guy than the one she was now married to) couldn't live there, he had to live with his grandparents. I think that must have been an interesting story.