Group quarters categories where poverty==0

The universe for POVERTY in ACS says “not group quarters.” But if I look at GQ if POVERTY==0 I still get cases across all categories of GQ, though the largest numbers are certainly in 3 and 4. What do I make of the cases where gq==1 or 2?

In 2008 ACS I’ve got:

ta gq if poverty==0

1 | 3,002
2 | 278
3 | 51,242
4 | 21,921
5 | 25

Total | 76,468

It seems that the universe statement for POVERTY should be updated. After looking into the records with GQ==1,2, or 5 and POVERTY==0, it seems that these individuals also have the characteristics FAMUNIT>1, SUBFAM==0, and are AGE<15. So, basically, these records identify unrelated children who do not make up their own sub-family. Therefore, they should be considered NIU for POVERTY.

Ok, thanks so much!