GEE estimation procedure allows one to account for the correlation between adoption decisions
measured as a continuous variable while maintaining the integrity of the discrete choice model.
A National Integrated Field/Farm Production Practice,
Resource, Economic, and Environmental Survey
CEAP-ARMS integrates two producer-based surveys — one, the National Resources
Inventory (NRI) point-based production practice/environmental data survey (CEAP), and two, the
field/farm level production practice, resource use, farm household and economic survey (ARMS).1
ARMS, conducted for USDA’s ERS, is designed to primarily serve information objectives
involving cost-of-production, farm finances, and production practices. Using a streamlined
integrated questionnaire, CEAP-ARMS directly links more detailed production practice, program
participation, and site-specific environmental data from the NRCS CEAP, with the economic, farm
resource, and farm-household/operator characteristic data from ARMS.
The 2005 Phase II CEAP-ARMS included a sample of 489 NRI point-based farm fields (for
corn) across 4 States,2 with an average completion rate of 78 percent. When integrated with
associated NRI data, the usable Phase II sample was 380 observations with associated field-level
production practice, input use, program participation, and NRI environmental data. However, when
the Phase II/NRI data is integrated with the corresponding farm-level Phase III data, the usable
sample is 226 field/farm observations. This integrated production practice, program participation,
farm resource/economic, farm-household, operator, and environmental database provides the unique
opportunity to summarize initial characteristic differences for corn producers between conservation
1 CEAP, ARMS, and CEAP-ARMS are all surveys conducted by USDA’s National Agricultural Statistics Service.
ARMS is a crop-specific survey based on a list frame sample with the survey conducted in three phases: Phase I
involves survey planning/design and sample selection; the Phase II questionnaire collects field-level production
practice, input use, and cost-of-production data (for the annual survey crop of choice), and the Phase III follow-on
questionnaire collects associated farm-level resource, economic, and operator/household data. CEAP and CEAP-
ARMS, being NRI-point based, use an area frame sample design.
2 CEAP-ARMS for 2004 wheat included the States of Washington, Oregon, Idaho, Montana, North Dakota, South
Dakota, Nebraska, Colorado, Kansas, Oklahoma, Texas, Minnesota, Missouri, Illinois, Michigan, and Ohio. CEAP-
ARMS for 2005 corn included the States of Indiana, Illinois, Iowa, and Nebraska.
3