|
Activity Number:
|
170
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2003 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #300641 |
|
Title:
|
Redesign of the Survey of Construction
|
|
Author(s):
|
Bonnie E. Kegan*+ and James D. Ashley
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau
|
|
Address:
|
2915 Greenwich Ct., Crofton, MD, 21114-2850,
|
|
Keywords:
|
optimal allocation ; overlap ; stratification
|
|
Abstract:
|
The Survey of Construction (SOC) is designed to estimate the number of new privately owned housing units started, under construction and completed, the number of new single-family houses sold and for sale, and other characteristics of new residential housing. SOC samples permits from preselected permit offices and canvasses selected nonpermit (NP) areas for new construction. The last redesign of SOC was completed in 1984 before the conversion of the survey to Computer Assisted Personal Interviewing. The SOC PSUs are a subsample of PSUs selected by the Current Population Survey (CPS) as the two surveys share field representatives for data collection. The samples of permit offices and NP areas are drawn from this subsample. This paper details the first stage of the redesign in which the SOC PSUs are selected. It describes the formulation of cost and variance models used in the optimal allocation, as well as the criteria used to identify CPS PSUs as self-representing or nonself-representing (NSR) for SOC. The stratification of NSR PSUs and their selection using overlap methods is also described. Finally, the redesigned sample of PSUs is compared to the 1984 sample of PSUs.
|
