Skip to main content

Undercounts of the number of household members in the AHS

HUD.GOV HUDUser.gov
eList
AHS
table>
Comparing 2003 AHS estimates to Census (2003 American Community Survey) I found that the number of people living in households according to the AHS is about 8% lower than the number according to Census.  It is my understanding that this downward bias is due to the fact that AHS data is weighted to correct for underreporting of housing units, but not within-housing unit underreporting.
 
I am wondering if anyone has looked at this within-household underreporting problem, and if so, devised a method to correct for it. For example, I would like to estimate the number of housing units with at least one child under 5 years old, but I am worried that my estimate will be too low without making some sort of adjustment.
 
Thanks.
 
-Matt mattlapenta@yahoo.com

Sell on Yahoo! Auctions - No fees. Bid on great items.