Agricultural Chemical Useage ARMS II Statistical Methodology

Statistical Methodology
Name of Survey: Agricultural Resource Management Survey (ARMS) Phase II: Corn and
Organic Corn Production Practices and Costs Report (PPCR), Upland Cotton and Fall Potatoes
Production Practices Report (PPR)
Name of Summary: Agricultural Chemical Usage, 2010 Corn, Organic Corn, Upland Cotton,
and Fall Potatoes Summary
(Access at USDA-NASS Quick Stats 2.0: http://quickstats.nass.usda.gov/. See Program: Survey,
Sector: Environmental, Groups: Field Crops and Vegetables)
Data Collection Period: September to December 2010
Target Population and Sampling Frame: The target population for the ARMS Phase II was the
official USDA farm population, within the pre-defined Program States, which is defined as “all
establishments that sold or would normally have sold at least $1,000 of agricultural products
during the year,” excluding abnormal or institutional farms.
Program States are selected to both minimize the total number of States selected, and ensure at
least 80 percent coverage of target-commodity acreage.
The USDA-NASS list sampling frame, including all known U.S. operations qualifying as a farm
under the USDA-NASS definition, theoretically accounting for 90 percent of total U.S. land in
farms, was the frame used to derive the sample. Control data was used to exclude abnormal or
institutional farms.
Sampling Methods: Through ARMS Phase I screening, only operations which both qualified for
the target population, and reported the commodities of interest, were eligible for the ARMS
Phase II sample.
From the resulting subpopulation of farms with positive acres of the target commodities, samples
for the PPCR commodities corn and organic corn were drawn through Sequential Interval
Poisson sampling - a method used primarily to control overlap within the ARMS program and
overlap with concurrent USDA-NASS surveys. Samples for the PPR commodities upland cotton
and fall potatoes were drawn through Multiple Probability Proportional to Size sampling
(acreage of the target commodities was the measure of size). This methodology ensured that (1)
target sample sizes and commodity coverage were met, and (2) overlap was maximized in order
to decrease the total number of contacts.
Chemical use and production practice data for one randomly selected field of the target
commodity(s) in each sampled operation were collected through personal interview.
Sample Unit and Reporting Unit: The sample unit was the individual farm. The reporting unit
was a single, randomly selected field from each sampled farm.

Modes of Data Collection: Personal interview

Selected Terms and Definitions:
Active Ingredient: The specific pesticide ingredient which kills or controls the target pest(s) or
other target material(s), or otherwise results in the pesticide effect(s). All pesticide-use estimates
in report are at the active ingredient level; one or more active ingredients are present in known
amounts in the pesticide products reported in survey.
Estimates of active ingredient use were reported in a single unit of equivalence, per ingredient.
For salt, ester, or amine active ingredients, estimates were reported in the parent acid equivalents.
For example, the acid derivatives glyphosate isopropylamine salt and 2,4-D, 2-ethylhexyl ester
were reported in the glyphosate and 2,4-D equivalents, respectively. For copper compounds,
estimates were reported in the metallic copper equivalent.
Active Ingredient Code: A unique code assigned to each active ingredient upon registration with
the Environmental Protection Agency’s Office of Pesticide Programs, to facilitate pesticide
regulation.
Area Applied, Percent: Percent of total planted acres which received one or more applications of
a specific fertilizer nutrient or pesticide active ingredient. (In Quick Stats: Treated, Measured as
Percent of Area Planted)
Avoidance: A strategy in which the detrimental effects of pests on crops are mitigated or
eliminated solely through various cultural practices. Avoidance is one of four classes of pestmanagement practices for which data are included in report.
Beneficial Insects: Insects (small invertebrate animals, mostly of arthropod classes Insecta and
Arachnida), which are collected and introduced onto crop acres because of their value in
biological control as predators on harmful insects and parasites.
Chemigation: Application of agricultural chemicals, including pesticide products, by injection
into irrigation water.
Crop Year: The period starting immediately after harvest of the previous year’s crop and ending
at harvest of the current year’s crop.
Farm: Any place from which $1,000 or more of agricultural products were produced and sold, or
normally would have been sold, during the year. Government payments are included in sales.
Fertilizer: A soil-enriching agricultural input which contains one or more plant nutrients. Data
for three primary macronutrients nitrogen (N), phosphate (P), potash (K), and the secondary
macronutrient sulfur (S) are included in report.

Fungi: Various organisms of the kingdom Fungi, which obtain nutrients by decomposing plant
or other organic life. This pest group includes mushrooms, molds, mildews, smuts, rusts, and
yeasts. Fungal infestations have the potential to reduce crop production and/or lower the grade
quality of the host crop.
Mechanism of Action (MOA): The method or biological pathway by which the pesticide or
active ingredient kills or controls the target pest(s) or other target material(s).
Monitoring: A strategy involving the observance or detection of pests through systematic
sampling, counting, or other forms of scouting. Monitoring may include prediction of pest
population levels through the observance of environmental factors such as weather or soil and
crop quality. Monitoring is one of four classes of pest-management practices for which data are
included in report.
Nematodes: Unsegmented, parasitic worms of the phylum Nematoda. Prominent animal pest of
wheat and other field crops with the potential to be highly destructive, lowering crop production
and grade quality significantly.
Number of Applications: The average number of times a treated acre received a specific fertilizer
nutrient or pesticide active ingredient. (In Quick Stats: Applications, Measured in Number)
Pesticide: Defined by the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) as “(1)
any substance or mixture of substances intended for preventing, destroying, repelling or
mitigating any pest, (2) any substance or mixture of substances intended for use as a plant
regulator, defoliant, or desiccant, and (3) any nitrogen stabilizer…”(Title 7, U.S. Code, 136).
Under FIFRA, pesticides are registered and regulated through the Environmental Protection
Agency’s Office of Pesticide Programs. Four classes of pesticides are included in report: (1)
herbicides targeting weeds, (2) insecticides targeting insects (3) fungicides targeting fungi, and
(4) other chemicals targeting all other pests or other materials (including extraneous crop
foliage).
Pheromone: A chemical substance produced by an insect which serves as a stimulus to other
individuals of the same species for one or more behavioral responses.
Prevention: A strategy in which a pest population is kept from infesting a crop or field, by
taking various preceding actions. Prevention is one of four classes of pest-management practices
for which data are included in report.
Rate per Application: Ratio indicating pounds (lbs) of a fertilizer primary nutrient or pesticide
active ingredient applied, per single application, per planted acre. (In Quick Stats: Applications,
Measured in Lb/Acre/Application)

Rate per Crop Year: Ratio indicating pounds (lbs) of a fertilizer primary nutrient or pesticide
active ingredient applied, counting all applications per crop year, per planted acre. (In Quick
Stats: Applications, Measured in Lb/Acre/Year)
Suppression: A strategy which involves the control or reduction of existing pest populations in
order to mitigate crop damage. May include physical or biological controls, or management of
resistance build-up through pesticide rotation. Suppression is one of four classes of pestmanagement practices for which data are included in report.
Data Review and Estimation Procedures: Estimates were based on data collected, reviewed,
and verified through the cooperative efforts of the USDA-NASS Environmental and
Demographics Section and Program-State Field Offices. NASS maintains chemical use
databases which contain product recommended use ranges and active ingredient concentrations
per product. These databases are used to review pesticide product usage data and to convert
pesticide product usage data to the equivalent active ingredient levels for publication. Review
and finalization of all data proceeded with assessment of reasonableness and consistency at the
record, State, and U.S. levels.
Total Applied estimates were calculated and published at the population-level using official
USDA-NASS 2010 acreage estimates by State (published in “Crop Production – 2010
Summary” [Cr Pr 2-1 (11)]). For this reason, all published estimates apply at the populationlevel defined by these official estimates. Note that although official USDA-NASS crop acreage
estimates are subject to revision, the Total Applied estimates are not subject to revision.
Indications were published if five or more reports were received and summarized for a given
variable, and if the indication was greater than or equal to one half of the pre-determined
rounding unit for that variable. Distribution and reliability measures were published if 30 or
more reports were received and summarized for a given variable.
Reliability: Estimates were subject to sampling variability; sampling variability was measured
by the coefficient of variation (cv), expressed as a percent of the estimate. Coefficients of
variation were derived and published for Percent Treated, Number of Applications, Rate per
Application, and Rate per Crop Year pesticide-usage variables. In general, the lower the cv, the
better the indication. Coefficients of variation differed considerably by variable, chemical, and
crop. The narrower the numerical range of responses per variable, and the larger the number of
positive responses per variable, the smaller the sampling variability. For these reasons, cv’s were
generally lower for active ingredient Rate of Application estimates, and for estimates associated
with the most often reported active ingredients (application rates reported almost always fell
within the manufacturer’s relatively narrow recommended usage range, and a relatively large
number of reports were received for the most widely used active ingredients).
Estimates were additionally subject to non-sampling errors. Non-sampling errors result when the
target population is mis-defined through list duplication or incompleteness, or sample unit data is

mis-recorded through mistakes in reporting, recording, or processing the data. Strict quality
controls implemented at each step of the survey and data review process minimized the
occurrence and magnitude of non-sampling errors.
Revision Policy: Estimates are final at first publication, and are not subject to revision.