|
Activity Number:
|
65
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #306888 |
|
Title:
|
NASS/USDA Area Frame Sample Allocation for Estimation of Number of Farms Not on the Ag Census Mailing List
|
|
Author(s):
|
Floyd Spears*+ and Raj S. Chhikara and Charles R. Perry and Phillip S. Kott
|
|
Companies:
|
Harding University and University of Houston-Clear Lake and National Agricultural Statistics Service and National Agricultural Statistics Service
|
|
Address:
|
Box 10764, Searcy, AR, 72149,
|
|
Keywords:
|
optimum sample allocation ; area frame sample ; modeling standard deviations ; agriculture census ; not on mailing list
|
|
Abstract:
|
USDA/NASS needs to estimate the number of farms that are not on the census mailing list (NML) for the 2007 Ag Census. Additional sample segments are planned during its 2007 annual ag survey to determine estimates for several NML items (number of farms, and subdomains corresponding to several minority and specialty farms). Stratum standard deviations from the 2002 area frame sample data are modeled in terms of certain NML farm characteristics. The NASS multivariate allocation procedure is applied to determine sample allocations for several different sets of ag items, which include NML farms and the regular annual ag survey items. Various combinations of the actual 2002 design allocation and those obtained using the NML items are investigated. The allocation that most closely meets the sample size and precision goals is identified.
|
