Is this not the Central Limit Theorem[1] ?
Nobody this side of Eternity is working with a full data set.
Without sufficient domain knowledge you don't even know you don't have the right data, perspectives, or even know the apparent obvious paths that are actually blind alleys in a subject. Every field of human knowledge is full of these.
The Dunning-Kruger effect is about misjudging your own abilities, leading to overconfidence in tasks or self-rated knowledge of domains.
Both tries to explain why people make mistakes, but one through the lens of cognition and information processing, the other to through arrogance and level of domain expertise.
If these concepts are useful is another matter.