There is pretty good data on this from pre-industrial England. What you are asserting is what historians widely believed before they actually looked at the dat

There is pretty good data on this from pre-industrial England. What you are asserting is what historians widely believed before they actually looked at the data, around the 1960s. They said this by looking at the more recent past of less developed places (like rural Russia, pre-1917) and assuming that the far past of England looked the same, and thus that nuclear families became common only recently. And they were wrong. Of course, there are many places even further away than Russia. Lots of other cultures did have multi-generational families as the pattern, and many still do.