II Department of
SRB Research Report
An Analysis of the
Sampling Frame for the
Chemical Use and Farm
AN ANALYSIS OF THE SAMPLING FRAME FOR THE CHEMICAL USE AND
FARM FINANCE SURVEY, by Susan Cowles and Susan Hicks, Sampling and Estimation
Research Section, Survey Research Branch, Research Division, National Agricultural
Statistics Service, United States Department of Agriculture, Washington, D.C. 20250-2000,
February 1994, Report No. SRB-94-05.
The National Agricultural Statistics Service (NASS) plans to reduce respondent burden and
improve data quality by combining the Objective Yield Cropping Practices Survey (Form H)
and the Cost of Production Survey (COPS) versions of the Farm Cost and Returns Survey
(FCRS) to form the Chemical Use and Farm Finance Survey (CUFFS). A pilot survey for
CUFFS was conducted in 1991 in Minnesota and both Minnesota and Louisiana in 1992.
Most NASS surveys utilize a multiple frame design -- a combination of list and area frames.
To evaluate the necessity of a multiple frame sample for this survey, each State's Form H
data for selected commodities was partitioned into overlap (OL) and nonoverlap (NOL)
domains. Multiple frame (MF) and OL estimates of percent of acres treated and rate of
application were compared. Comparisons were then made between the multiple frame
components--OL and NOL domains. Significant differences were found but they were not
consistent across commodity, chemical, State or time.
Sample frame; Bootstrap-t confidence intervals; Overlap; Nonoverlap.
This paper was prepared for limited distribution to the research community outside the
U.S. Department of Agriculture. The views expressed herein are not necessarily those of
the USDA or NASS.
The authors would like to thank Fatu Wesley for her assistance as well as Larry Pope and
others who assisted in compiling this data.