Skip to main content

Efficiency apts in 1998 DC metro

HUD.GOV HUDUser.gov
eList
AHS
We have found what appears to be a large discrepancy between the number of
efficiency (i.e., no bedroom) apartment units reported by the 1998 AHS metro
survey and the number of such units reported by the Census.  For the
Washington, DC metro area, table 2-3 of the AHS printed report for 1998
indicates that there are about 4,300 renter-occupied efficiency housing
units.  The Census reports, however, that in the same counties corresponding
to the AHS metro area there were 47,767 and 69,400 renter-occupied
efficiency housing units in 1990 and 2000, respectively.

The AHS report we are accessing is at
http://www.census.gov/prod/2000pubs/h170-98-18.pdf.

Any ideas as to why the AHS estimate is so far from the Census estimates?
The definitions appear to be similar across the two sources.  The estimates
for units of 1 bedroom or larger reported by the AHS seem to be consistent
with those reported by the Census.  So, the problem seems to be isolated to
the no bedroom units.

We have reviewed the AHS data anomalies report for 1998 on the web site and
this issue is not mentioned.  We have not checked the numbers for other
metro areas to see if a similar problem occurs, however.

This leads me to a more general question about data checking for the AHS.
Is there any systematic procedure for comparing AHS estimates to
corresponding Census ones to see if there are large discrepancies?

Thank you for any insights that anyone may have.

Peter A. Tatian
Senior Research Associate
The Urban Institute

ptatian@ui.urban.org
http://www.urban.org