Census of Agriculture Methodology - Appendix A

Appendix A.
Census of Agriculture Methodology
The purpose of a census is to enumerate all objects
with a defined characteristic. For the census of
agriculture, that goal is to account for “any place from
which $1,000 or more of agricultural products were
produced and sold, or normally would have been sold,
during the census year.” To do this, NASS creates a
Census Mail List (CML) of agricultural operations
that potentially meet the farm definition, collects
agricultural information from those operations,
reviews the data, corrects or completes the requested
information, and combines the data to provide
information on the characteristics of farm operations
and farm producers at the national, State, and county
levels. In this appendix, these census processes are
described.
THE CENSUS POPULATION
The Census Mail List
The National Agricultural Statistics Service (NASS)
maintains a list of farmers and ranchers from which
the CML is compiled. The goal is to build as complete
a list as possible of agricultural places that meet the
farm definition. The CML compilation begins with
the list used to define sampling populations for NASS
surveys conducted for the agricultural estimates
program. Each record on the list includes name,
address, telephone number, and email plus additional
information that is used to efficiently administer the
census of agriculture and agricultural estimates
programs.
NASS builds and improves the list on an ongoing
basis by obtaining outside source lists. Sources
include State and federal government lists, producer
association lists, seed grower lists, pesticide
applicator lists, veterinarian lists, marketing
association lists, and a variety of other agriculturerelated lists. NASS also obtains special commodity
lists to address specific list deficiencies. These outside
source lists are matched to the NASS list using record
linkage programs. Most names on newly acquired
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

sources are already on the NASS list. Records not on
the NASS list are treated as potential farms until
NASS can confirm their existence as a qualifying
farm. Staff in NASS regional and field offices
routinely contact these potential farms to determine
whether they meet the farm definition. For the 2017
Census of Agriculture, NASS made a concerted effort
to work with community-based organizations not only
to improve list coverage for minorities but also to
increase census awareness and participation.
List building activities for developing the 2017 CML
started in 2014 by updating list information from
respondents to the 2012 Census of Agriculture.
Between 2015 and 2017, NASS conducted a series of
National Agricultural Classification Surveys (NACS)
on approximately 1.6 million records, which included
nonrespondents from the 2012 census and newly
added records from outside list sources. The NACS
report forms collected information that was used to
determine whether an operation met the farm
definition. If the definition was met, the operation was
added to the NASS list and subsequently to the CML.
Addressees that were nonrespondents to a NACS
were also added to the CML and identified with a
special status code.
Measures were taken to improve name and address
quality. Additional record linkage programs were run
to detect and remove duplicate records both within
each State and across States. List addresses were
processed through software programs that utilize the
United States Postal Service’s National Change of
Address System and the Locatable Address
Conversion System to improve mail delivery.
Records on the list with missing or invalid phone
numbers were matched against a nationally available
telephone database to obtain as many phone numbers
as possible. To reduce costs, operations with
characteristics that indicated they were unlikely to be
farms, according to the farm definition, were removed
from the list.
Appendix A A - 1

The official CML for the 2017 Census of Agriculture
was established on September 3, 2017. The list
contained 2,999,098 records. Of these, 2,259,750
records were thought to meet the NASS farm
definition and 739,348 were potential farm records,
which included NACS nonrespondents, other records
added to the CML by the NASS regional field offices
after the record linkage process, and late adds to the
CML that were not included in any previous NACS or
State screening survey.

Not on the Mail List (NML)
Extensive efforts are directed toward developing a
CML that includes all farms in the U.S. However,
some farms are not on the list, and some agricultural
operations on the list are not farms. NASS uses its
June Area Survey (JAS) to quantify the number and
types of farms not on the CML. The records in the
JAS that are not on the CML are said to be in the Noton-the-Mail List (NML) domain. If a JAS record in
the NML domain is determined to be a farm during
the census, it is an NML farm. The NML farms are
used to measure coverage associated with the census.
The JAS is based on an area frame, which covers all
land in the U.S. and includes all farms. The land in the
U.S. is stratified by characteristics of the land. A
probability sample of segments is drawn within each
stratum for the JAS. Segments of approximately equal
size are delineated within each stratum and designated
on aerial photographs. The JAS sample of segments is
allocated to strata to provide accurate measures of
acres planted to widely grown crops, farm numbers,
and inventories of cattle. Sampled segments in the
JAS are personally enumerated. Each operation
identified within a segment boundary is known as a
tract.

The 2017 JAS sample was increased to improve the
farm counts for operations that produced specialty
commodities or had socially disadvantaged or
minority producers. The total JAS sample consisted
of 13,972 segments of which 3,012 were additional
segments. This set of additional segments is referred
to as the Agricultural Coverage Evaluation Survey
(ACES) segments. The ACES segments were selected
using a multivariate sampling design that targeted
specific items at the U.S. level. The 2017 JAS
A - 2 Appendix A

consisted of sample segments from all States, with the
exception of Alaska where NASS does not maintain
an area frame.
During the JAS/ACES enumeration process, each
tract is identified as either agricultural or nonagricultural. Each JAS/ACES agricultural tract is
identified as a farm or non-farm in June based on the
farm definition of $1,000 of sales or potential sales of
agricultural products. Non-agricultural tracts are
further classified into categories: with farm potential,
with unknown farm potential, or with no farm
potential. The names and addresses collected in the
2017 JAS/ACES were matched to the CML. Those
from the 2017 JAS/ACES that did not match were
determined to be in the NML domain and sent a
yellow census report form so that they could be
differentiated from the green report form sent to those
addressees on the CML. Instructions on the census
report form directed any respondent who received
duplicate forms to complete the CML form and to
mail all duplicate forms back together. Those who
returned a CML and an NML form had been
misclassified as NML and were removed from the
NML domain.
The initial NML mailout consisted of 42,430 records.
A total of 41,787 NML records were summarized of
which 2,799 records were confirmed to be NML and
in-scope.
The farm/nonfarm status of each NML domain
operation was determined based on the reported data
in the census form. An operation in the NML domain
that was determined to be a farm is referred to as an
NML farm. Characteristics of NML farms and their
producers provided a measure of the undercoverage
of farms on the CML. The percentage of farms not
represented on the CML varied by State. In general,
NML farms tended to be small in acreage, production,
and sales of agricultural products. Farm operations
were missing from the CML for various reasons,
including the possibility that the operation started
after development of the CML, the operation was so
small that it did not appear in any agriculture-related
source list, or the operation was misclassified as a
nonfarm prior to census mailout. The CML was used
with the NML in a capture-recapture framework to
represent all farming operations across all States in
the JAS sample.
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

DATA COLLECTION OUTREACH AND
PROMOTIONAL EFFORTS
NASS planned and executed a multi-phase strategic
communications campaign for the 2017 Census of
Agriculture, to increase the level of awareness and
response among all U.S. agricultural producers.
 Phase 1 ran from December 2016 − June 2017. It
raised awareness about the census and list building,
encouraged producers to sign up in response to
NASS mailings and at community, association,
and other stakeholder meetings where NASS
partners reached out.
 Phase 2 ran from July 2017 − December 2017. It
notified farm producers and agricultural
organizations that the census would be mailed in
December, and encouraged communications
regarding the census.
 Phase 3 ran from December 2017 – July 2018. It
focused on census data collection with messaging
urging response, reminding producers that it was
not too late to respond.
 Phase 4 ran from August 2018 – February 2019. It
thanked producers for their participation and
NASS partners for their support, and informed all
of the February 2019 data release plan.
The communications campaign focused on these
primary areas: partnership building, local-level
outreach, public relations, media relations, paid
media, and social media. Some external support was
provided by a private communications agency (i.e.
primarily assistance with paid media/advertising
strategy and ad creation) and a freelance writer.

The unifying force behind the 2017 communications
campaign was the theme “Your Voice. Your Future.
Your Opportunity.” This was accompanied by
supporting messages and artwork that created a
consistent look and feel for all census
communications. All messages and materials served
the purpose of inspiring action: Grow Your Farm
Future - Shape Your Farm Programs - Boost Your
Rural Services - Fill out your Census of Agriculture Do your part to be counted - The Census of
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Agriculture is Your Voice, Your Future, Your
Opportunity.

Partnership and Local-Level Outreach
At the national level, NASS officials met with leaders
from dozens of agricultural organizations, State
Departments of Agriculture, and other USDA
agencies to successfully secure their support in
promoting the census among their constituencies.
Stakeholders partnered with NASS to promote the
2017 Census of Agriculture through publications (e.g.
newsletters), special mailings, speeches, social media,
websites, and other communications. In addition,
through grassroots-level outreach and efforts, NASS
partnered with a number of community-based
organizations to reach minority and limited-resource
farmers and ranchers. National-level outreach was
encouraged and mirrored at the regional, State, and
local levels. Among the highlights of these
partnership efforts was the production of multiple
television and radio public service announcements
featuring the U.S. Secretary of Agriculture, State
secretaries, directors, and commissioners of
agriculture and leaders from community-based
organizations.
Coverage of American Indian and Alaska
Native Farm Producers
To maximize coverage of American Indian and
Alaska Native agricultural producers, special
procedures were followed in the census. A concerted
effort was made to get individual reports from every
American Indian and Alaska Native farm or ranch
producer in the country. If this was not possible within
some reservations, a single reservation-level census
report was obtained from knowledgeable reservation
officials. These reports covered agricultural activity
on the entire reservation. NASS staff reviewed these
data and removed duplication with any data reported
by American Indian or Alaska Native producers who
responded on an individual census report form.
Additionally NASS obtained, from knowledgeable
reservation officials, the count of American Indian
and Alaska Native producers (on reservations) who
were not counted through individual census report
forms, but whose agricultural activity was included in
the reservation-level report form.
Appendix A A - 3

Table D, American Indian and Alaska Native
Producers: 2017 provides the number of producers
(1) reported as American Indian or Alaska Native in
the race category, either as a single race or in
combination with other races, on the individual
census report forms (for up to four per farm) and (2)
identified as American Indian or Alaska Native
producers farming on reservations by reservation
officials. The count from the individual report forms
is summarized in the “Individually reported” column.
It includes up to four producers on or off reservations.
The “Other” column provides counts of producers on
reservations as reported by a reservation or tribal
official. The “Total” column is simply a sum of the
“Individually reported” and the “Other” columns.
Tables in other parts of the publication count the
reservation-level reports as single farms.

Public Relations
In the public relations arena, NASS worked with
internal and external stakeholders to equip them with
communications tools and resources to deliver the
census communications message to their audiences.
NASS utilized its Intranet and the Partner Tools page
on the census website to deliver materials to the 12
regional and 46 field offices as well as to external
stakeholders. The materials included but were not
limited to: customizable news releases, public service
announcement scripts, and a PowerPoint template;
Secretary of Agriculture video public service
announcements, and drop-in advertisements;
informational, instructional, and testimonial videos;
website buttons and banners; brochures in multiple
languages; flyers; posters; FAQ sheets, talking points,
and more. In addition, at the national level, NASS
issued six news releases during data collection (three
more were produced before data collection to inform
and prepare producers) citing department and agency
spokespeople, published half a dozen timely and
relevant pieces to the USDA blog highlighting the
census, and conducted three social media campaigns.
These public relations efforts at the national and locallevels helped ensure that NASS’ message about the
census was continually in the media, including print
and online publications, a variety of social media,
radio, and some television programs. Media outlets
included both those specializing in agriculture and
more general outlets.
A - 4 Appendix A

Paid Media
Even with increasingly limited budgets and resources,
NASS was able to apply a small portion of funds
toward paid media. For the 2017 Census of
Agriculture, NASS strategically advertised in
regional print publications, online, and with national
agriculture news services (i.e. TV, radio) to bolster
reach both in general and within geographicallyspecific, previously under-represented populations
and lower response areas.
DATA COLLECTION
Method of Enumeration
Data collection was accomplished primarily by mail,
Computer-Assisted Self Interview (CASI) on the
Internet, and personal enumeration for special classes
of records in the census operations. Personal
enumeration (interviewing) involved the use of both
Computer-Assisted Telephone Interview (CATI) and
Computer-Assisted Personal Interview (CAPI) data
collection instruments. Enumerators at the five NASS
Data Collection Centers conducted CATI data
collection. In addition, enumerators under contract
with NASS through the National Association of State
Departments of Agriculture (NASDA) conducted
phone and personal interviews with respondents. For
the 2017 Census of Agriculture, NASS implemented
a pre-notification strategy in an effort to increase
awareness, improve overall responses, and encourage
respondents to report early to avoid continued
correspondence. All records with an e-mail address
received an e-mail message marketing the improved
web form and announcing the census mail packets
were coming.
Report Forms
Four versions of report forms were used for the 2017
Census of Agriculture:
 General form (17-A100)
 Short form (17-A200)
 Hawaii form (17-A101)
 American Indian form (17-A300)
The general form facilitated reporting crops and
livestock most commonly grown and raised in the
U.S. The short form expedited reporting specific
crops or livestock for pre-identified farms and ranches
in the U.S. The Hawaii form targeted crops and
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

livestock specifically grown or raised on farms and
ranches in Hawaii. The American Indian form
focused on crops and livestock for farms and ranches
on reservations in Arizona, New Mexico, and Utah.
All of the report forms allowed respondents to write
in specific commodities that were not prelisted on
their report form.
Report Form Mailings
Pre-notification of census data collection began on
November 17, 2017. Approximately 600,000
producers with an active e-mail address on the census
mail list received a message informing them of the
upcoming census data collection period and
encouraging them to utilize the new census web form.
Between November 27 and November 30, 2017,
approximately 1 million producers received a letter
with their survey code and instructions for completing
their census online. The letter encouraged producers
to report online early to avoid receiving mail and
phone follow-up. Approximately 3 million mail
packets were mailed in December 2017 and January
2018. Each packet contained a cover letter, instruction
sheet, a labeled report form, and a return envelope.
The Census Bureau’s National Processing Center
(NPC) in Jeffersonville, IN was contracted to perform
mail packet preparation, initial mailout, and two
follow-up mailings to nonrespondents.
The initial mailout was followed by a thank-you
reminder postcard that was delivered in January 2018
to all operations that received mail packets. First
follow-up mail packets were mailed in mid-February
2018 to approximately 1.5 million nonrespondents.
Second follow-up mail packets were mailed in midMarch 2018 to approximately 1 million
nonrespondents.
Nonresponse Follow-up
Operating concurrently with NPC’s mail data
collection efforts, NASS Data Collection Centers
targeted selected groups of census nonrespondents for
telephone enumeration. NASS regional field offices
targeted selected groups of census nonrespondents for
in-person enumeration. These efforts were referred to
as:
 Must Case Follow-up
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

 American Indian Producer Follow-up
 National Nonresponse Follow-up
 Not on Mail List (NML) Follow-up
Must Case Follow-up. Must cases are known large
or unique operations, the absence of which could have
significantly affected the accuracy of census results.
For the 2017 Census of Agriculture, 125,697 records
were categorized as Must cases. Each active Must
operation was accounted for by mail receipt, phone
interview, or personal enumeration; if an operation
was no longer in business, its nonfarm status was
documented. Call centers conducted CATI calling of
nonrespondent Must cases from March 2018 through
May 2018, after the initial and first follow-up
mailings. Following the CATI calling, the remaining
nonresponse Must cases were assigned to regional
field offices for personal enumeration. Because of the
potential importance of Must cases, they were all
accounted for and therefore not eligible for
nonresponse weighting adjustment.
American Indian Producer Follow-up. The
American Indian report form (17-A300) was mailed
to all operations in Arizona, New Mexico and Utah
thought to have an American Indian producer. It was
included in the initial mailout, but due to poor mail
response, a personal enumeration data collection
strategy was utilized with no additional mail followup. A concerted effort was made to get individual
reports from every American Indian farm producer in
the country. If this was not possible within a
reservation, a single reservation-level census report
was obtained from knowledgeable reservation
officials. These reports covered agricultural activity
on the entire reservation. NASS staff reviewed these
data and removed any duplicate data reported by
American Indian producers from that reservation who
responded on an individual census report form.
Additionally NASS obtained, from knowledgeable
reservation officials, the count of American Indian
farm producers (on the reservations) who were not
counted through individual census report forms, but
whose agricultural activity was included in the
reservation-level report form.
National Nonresponse Follow-up (Excludes Must
Records). The National Nonresponse follow-up
activity was designed to focus nonresponse follow-up
in a manner that would both reflect the characteristics
Appendix A A - 5

of the nonresponders and increase response rates. In
April 2018, a sample of 249,521 nonrespondents was
selected from the remaining 864,260 nonrespondents
using a stratified random design. The strata were
based on State, county, size of farm, type of farm,
producer race, and propensity to respond. Beginning
in mid-April 2018 and continuing through July 2018,
extensive efforts were made to collect data for the
sampled records, including an additional CASI push,
autodial calls, CATI, and CAPI. Records in the same
stratum received the same set of collection methods.
Of the 80,504 responses, 51,846 records were
identified as being in-scope, resulting in a weighted
farm count of 143,847 from the sample.
Not-on-the-Mail List (NML) Follow-up. To account
for farming operations not on the CML, NASS used
its 2017 JAS sample from the NASS area frame,
augmented with the ACES segments. Because the
NASS area frame covers all land in the U.S. with the
exception of Alaska, it includes all farms. As
previously described, NASS conducted a record
linkage operation between the CML records and the
records from the 2017 JAS/ACES. Those 2017 JAS
records that did not match records on the CML were
designated as “Not-on-the-Mail List” (NML) records.
These records were mailed a yellow census form so
that it could be differentiated from the green forms
mailed to CML records. The NML records were
mailed at the same time as the census mailing and
received the same follow-up procedures as the census
mailing through the first follow-up in mid-February
2018. Beginning in March 2018, CATI was used for
nonresponse follow-up for NML nonrespondents.
REPORT FORM PROCESSING
Data Capture
The Census Bureau’s National Processing Center
(NPC) in Jeffersonville, IN was contracted to process
returned mail packets. NASS staff on site at the NPC
provided technical guidance and monitored NPC
processing activities. All report forms returned to the
NPC were immediately checked in, using bar codes
printed on the mailing label, and removed from
follow-up report form mailings. All forms with any
data were scanned and an image was made of each
page of a report form. Optical Mark Recognition
(OMR) was used to capture categorical responses and
to identify the other answer zones in which some type
A - 6 Appendix A

of mark was present.
Data entry operators keyed data from the scanned
images using OMR results that highlighted the areas
of the report forms with respondent entries. The keyer
evaluated the contents and captured pertinent
responses. Ten percent of the captured data were
keyed a second time for quality control. If differences
existed between the first keyed value and the second,
an adjudicator handled resolution. The decision of the
adjudicator was used to grade the performance of the
keyers, who were required to maintain a certain
accuracy level.
The images and the captured data were transferred to
NASS’s centralized network and became available to
NASS analysts on a flow basis. The images were
available for use in all stages of review.
Editing Data
Captured data were processed through a computer
formatting program that verified that records were
valid – that the record ID number was on the list of
census records, that the reported counties of operation
and production were valid, and other related criteria.
Rejected records were referred to analysts for
correction. Accepted records were sent to a complex
computer batch edit process. Each execution of the
computer edit in batch mode consisted of records
from only one State and flowed as the data were
received from NPC, the NASS Computer-Assisted
Self Interview (CASI), or the Computer-Assisted
Telephone Interview (CATI) applications.
The computer edit determined whether a reporting
operation met the qualifying criteria to be counted as
a farm (in-scope). The edit examined each in-scope
record for reasonableness and completeness and
determined whether to accept the recorded value for
each data item or take corrective action. Such
corrective actions included removing erroneously
reported values, replacing an unreasonable value with
one consistent with other reported data, or providing
a value for an item omitted by the respondent. To the
extent possible, the computer edit determined a
replacement value. Strategies for determining
replacement values are discussed in the next section.
Operations failing to meet the qualifying criteria for
being classified as a farm were categorized as out-ofscope for the census. Records that NASS had reason
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

to believe might have been erroneously classified as
out-of-scope (indications of recent and/or significant
agricultural activity reported on NASS surveys, for
example) were referred to analysts for verification.
The edit systematically checked reported data sectionby-section with the overall objective of achieving an
internally consistent and complete report. NASS
subject-matter experts had previously defined the
criteria for acceptable data. Problems that could not
be resolved within the edit were referred to an analyst
for intervention. Prior to the census mail-out, NASS
established a group of analysts in a Census Editing
Unit in the National Operations Center in St. Louis,
MO who examined the scanned images, consulted
additional sources of information, and determined an
appropriate action. Regional field office analysts also
participated using an interactive version of the edit
program to submit corrected data and immediately reedit the record to ensure a satisfactory solution.
Short Form Editing
From the CML, 400,000 records were selected to
receive a short form; this short form was derived from
the full census report form by reducing a number of
sections to a ‘total’ question – for example, instead of
asking the respondent to report the acreage for each
specific type of fruit or vegetable, the short form only
asked for total fruit acreage or total vegetable acreage.
In some cases, the same questions were asked on the
general form, in which case the edit treated the short
form responses as though they were incomplete
general forms, as described in the previous
paragraphs. In other cases, several items on the
general form were collapsed – for example, total acres
of Christmas trees and short rotation woody crops
were asked as a single item on the short form, instead
of separately as on the general form. In such cases,
different approaches were taken in the edit to create a
general form item or items from the short-form
specific items. Any short form record that reported
values above a certain threshold (in practice this
threshold was 0 for almost all items) for these shortform-specific questions was ‘flagged’ by the edit;
these records were later called back and the
respondent asked for additional information about the
items reported – for example, a producer reporting 10
acres of fruit on the short form was called back and
asked for the total, bearing, and nonbearing acres for
each type of fruit grown, as was asked on the general
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

form. If the producer was successfully contacted and
these additional data collected, the information was
added to the record as additional reported data, and
the edit was ‘reset to original’ – that is, the effects of
the previous edit were undone – and the record was
reedited with the new additional information. A flag
was passed to the edit so that the short form record
was not flagged for callback in such cases. In many
cases, of course, it was not possible to recontact the
respondent. In such cases, a flag was passed to the edit
system, and the record was unlocked and available for
review.
Imputing Data
The edit determined the best value to impute for
reported responses that were deemed unreasonable
and for required responses that were absent. If an item
could not be calculated directly from other current
responses, the edit determined whether acreage,
production, or inventory items had been reported for
that farm on a recent NASS crop or livestock survey.
For producers who had not changed in five years,
demographics such as race and gender were taken
from the previous census. Administrative data from
the Farm Service Agency were used for a few items,
such as Conservation Reserve Program acreage.
When deterministic edit logic and previouslyreported data sources were unable to provide a current
value, data from a reporting farm of similar type, size,
and location were considered. In cases where
automated imputation was unable to provide a
consistent report, the record was referred to an analyst
for resolution.
Separate system processes were established to
efficiently provide data from a similar farm to the edit
when donor imputation was required. The farm
characteristics used to define similarity between a
recipient record and its donor record were determined
dynamically by the edit logic. Euclidean distance was
used for similarity computations, with each
contributing
similarity
characteristic
scaled
appropriately. The most similar farm based on this
criterion (the “nearest neighbor”) was identified and
returned to the edit for use as a donor. The calculated
distance between the centroids of the principal
counties of production of the donor and recipient was
always included as one of the measures of similarity.

Appendix A A - 7

To provide donors to the automated edit, a pool of
successfully edited records was maintained for each
section of the report form. These donor pools began
with 2012 census data, reconfigured to emulate 2017
data and then edited using 2017 logic. Data from the
2015 Census Content Test were similarly remapped
and edited before being added to the original donor
pools. As 2017 records were successfully processed,
they were added to the donor pools, which maintained
the most recent data for each farm. Donor pools were
updated approximately every other week, as
determined by edit processing schedules. After
several updates, all initial data records were dropped,
leaving only 2017 records in the donor pools. After
each update, donor pool records were grouped into
strata containing farms in the same State of similar
type and size, using a data-driven algorithm to define
strata. Certain American Indian farms were treated as
a separate group, effectively having their own donor
pool.
In response to each donor request issued by the edit, a
dedicated system process would search the
appropriate stratum and respond with the most similar
donor, while giving preference to more recent donors.
In relatively rare instances where it was unable to
provide a donor, the donor selection process issued an
appropriate failure message to the edit. Imputation
failures occurred for several different reasons. The
requirement that an imputed value be positive could
have ruled out all available donors, as could have the
necessity for the donor record to satisfy a particular
constraint – say, that the donor record has cattle, but
no milk cows. In general, an imputation failure
occurred if there were no satisfactory donors in the
same profile as the report being edited. Records with
imputation failures were either held until more
records were available in the donor pool or referred to
an analyst. In addition, when such a failure occurred
in finding a donor for expenditure data, donor pool
averages were provided in lieu of an individual donor,
wherever possible. This “failover” utility was first
introduced for the 2012 census imputation process,
and significantly reduced the number of imputation
failures among the expenditure and labor variables.
During the early stages of editing, records requiring
imputation for production (and hence yields) of field
crops or hay, land values, or certain expenditure
variables, were set aside or “parked.” These records
were edited when the donor pools contained only
2017 records, ensuring that 2017 data were used in the
A - 8 Appendix A

imputations for the variables.
After receiving a donor's data, the edit substituted the
values into the edited record. In many cases, the donor
record's data value was scaled using another data field
specified in the edit logic. In such cases, the size of
the auxiliary field's value in the edited record, relative
to its value in the donor record, was used to
appropriately scale the donor record's value for the
field to be imputed. The imputed data were then
validated by the same edit logic to which reported data
were subject. Since imputation was conducted
independently for each occurrence, reports requiring
multiple imputations may have drawn from multiple
donors.
Substantial changes were introduced to the Personal
Characteristics section of the form in 2017.
Information on an additional (fourth) producer was
collected, and several new questions were added for
each producer – specifically, whether or not the
person was considered a “principal producer,”
whether the person was a spouse of a principal
producer, and whether the person was involved in any
of five types of decisions with respect to the
operation. These changes necessitated a new
imputation process for records reporting three or more
persons as producers. Records with one or two
persons reported as producers had these data edited
and imputed using the decision logic table edit and
donor pool imputation process. Records with three or
more persons reported as producers, and for which it
was determined that these data were inconsistent or
missing, had these data imputed using a fully
conditional specification method. During the edit for
records reporting three or more producers, the items
needing imputation were marked, and the record was
flagged. Periodically the data for these records (both
the items needing to be imputed and the other
variables needed by the model) were pulled and run
through the imputation program. The resulting
imputed values were loaded back to the records, and
the records were made available for review. This
process was conducted 19 times for the CML, and 6
times for the NML, during census production editing.
Data Analysis
The complex edit ensured the full internal consistency
of the record. Successfully completing the edit did not
provide insight as to whether the report was
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

reasonable compared to other reports in the county.
Analysts were provided an additional set of tools, in
the form of listings and graphs, to review record-level
data across farms. These examinations revealed
extreme outliers, large and small, or unique data
distribution patterns that were possibly a result of
reporting, recording, or handling errors. Potential
problems were investigated and, when necessary,
corrections were made and the record interactively
edited again.
When NASS summarizes data from the census of
agriculture, each individual report is typically
assigned to a single “principal” county. The principal
county is the county in which the majority of an
operation’s agricultural products are produced, as
reported by the producer. For large operations that
have significant production in multiple counties, their
reports may be broken up into multiple source
counties to more accurately summarize the data.
Similarly, for large farms operating in more than one
State, separate report forms are completed by State in
order to assign the proper portion of the farm’s total
agricultural production to each State in which the
farm operates.
ACCOUNTING FOR UNDERCOVERAGE,
NONRESPONSE, AND MISCLASSIFICATION
Although much effort was expended making the CML
as complete as possible, the CML did not include all
U.S. farms, resulting in list undercoverage. Some
farm producers who were on the CML did not respond
to the census, despite numerous attempts to contact
them. In addition, although each operation was
classified as a farm or a nonfarm based on the
responses to the census report form, some were
misclassified; that is, some nonfarms were classified
as farms and some farms were classified as nonfarms.
NASS’s goal was to produce agricultural census totals
for publication at the county level that were fully
adjusted for list undercoverage, nonresponse, and
misclassification.
In 2012 NASS used capture-recapture methodology
to adjust for undercoverage, nonresponse, and
misclassification. This same methodology was
implemented for the 2017 Census of Agriculture. To
implement
capture-recapture
methods,
two
independent surveys were required. The 2017 Census
of Agriculture (based on the CML) and the 2017 JAS
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

(based on the area frame) were those two surveys.
Historically, NASS has been careful to maintain the
independence of these two surveys.
A second assumption was that the proportion of JAS
farms with a given set of characteristics captured by
the census was equal to the proportion of U.S. farms
with those same characteristics captured by the
census.
For a farm to be identified as a farm, and thus captured
by the census, it must be on the CML, respond to the
census report form and, based on the census response,
be classified as a farm. Only those nonrespondents
included in the nonresponse sample had an
opportunity to be captured and had a probability πS of
being included in the sample; respondents prior to
drawing the nonresponse sample had πS = 1. Thus, the
capture probability πC is of interest:

πC = π(CML, Responded, Farm on Census|Farm) πS
Two types of classification error can occur. First, a
farm can be misclassified as a nonfarm. This type of
misclassification is accounted for in determining the
probability of capture πC. The second type of
classification error results when a response to the
census is classified as a farm operation when it does
not meet the definition of a farm. That is, some farms
on the CML may be misclassified from their census
report response and may be nonfarms. To account for
the misclassification of nonfarms as farms, the
probability of a farm on the census being classified
correctly must be estimated; that is,

πCCFC = π(Farm | Farm on Census)
where CCFC represents Correct Census Farm
Classification. To adjust for undercoverage,
nonresponse, and misclassification, each CML record
classified as a farm based on its response to the census
report form was given a weight of the ratio of the
estimated probability of correct classification of a
farm on the census and the estimated probability of
capture
where the hat symbol (^) denotes an
estimate). To estimate the number of farms with a
given set of characteristics, the weights of CML
records responding as farms on the census and having
that set of characteristics were summed. This
Appendix A A - 9

estimator is referred to as the capture-recapture
estimator (CR):

where F is the set of all CML records classified as
farms based on their responses to the census report
form.
To estimate the capture and correct census farm
classification probabilities, a matched dataset
consisting of JAS records and census records was
created. Records in the 2017 JAS sample were
matched to the 2017 census using probabilistic record
linkage. The CML records that matched with JAS
tracts represent the Census Sample.
Note: The Census Sample is a subset of the CML
records and includes only those records matching a
JAS tract. Both agricultural and non-agricultural
tracts were included in the matched dataset.

Resolving Farm Status
The farm status based on census responses to either
the CML or NML census data collection and the JAS
agreed in most cases; these records are referred to as
having resolved farm status. However, in other cases,
a record was identified as a farm (nonfarm) on the
JAS and as a nonfarm (farm) by the census through
either the CML or the NML. Such records are said to
have conflicting or unresolved farm status. An
operation identified as a farm is referred to as inscope; an operation identified as a nonfarm is referred
to as out-of-scope. From the set of matched records,
two groups with conflicting farm status were
identified: 1) in-scope JAS records that were out-ofscope on the census and 2) census in-scope and JAS
out-of-scope records. The records with conflicting
farm status were sent to NASS regional field offices
for review. In each case, efforts were made to
determine whether (1) the status had changed between
June and December when the census was conducted,
(2) the JAS farm status was correct, (3) the census
farm status was correct, (4) the records were
incorrectly matched, or (5) the farm status could not
be resolved. Not all of the records with conflicting
farm status could be resolved. In 2017, 8.1 percent of
A - 10 Appendix A

the records in the Census Sample had unresolved farm
status.
The probability an operation is a farm was estimated
for the records with unresolved farm status. Using the
2017 matched dataset, a logistic model of the
probability an operation is a farm based on the records
with resolved farm status was developed; that is, the
operations where the farm (or nonfarm) status agreed
between the JAS and the census were used to develop
a missing data model, which was then used to resolve
farm status. The final missing data model was used to
impute the probability that each of the agricultural
operations with unresolved farm status is a farm. For
the resolved farms and nonfarms, the probability of
the operation being a farm was 1 and 0, respectively.
Five-fold cross-validation was used to develop and to
compare competing models. The accuracy of the
model was thereby not overstated due to fitting and
evaluating the model on the same set of data. To
ensure that each of the cross-validation samples
covered the U.S., the five cross-validation samples of
JAS segments were drawn within State-stratum
combinations. Characteristics of the JAS tracts were
considered as potential covariates in the model.
Because limited information is available for JAS
nonfarm tracts, other covariates considered included
county-level socio-demographic variables from the
most recent U.S. population census, segment-level
data from the Cropland Data Layer, the county-level
rural-urban code, state-level response rates, an
indicator for records that are thought to be out-ofbusiness, and an indicator for records in the national
nonresponse sample. The sample weight associated
with each JAS tract was multiplied by the probability
of being a farm. This adjusted weight was used in all
subsequent modeling.
Capture Probabilities
Recall that, for a farm to be identified as a farm, and
thus captured, by the census, it must be on the CML,
respond to the census report form and, based on the
census response, be classified as a farm. These
adjustments are dependent. Further, those
nonrespondents at the time the nonresponse sample
was drawn had a known probability πS of being
included in the sample; respondents before the sample
was drawn had πS = 1. Therefore, the probability of
capture πC may be written as
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

πC = π(CML, Responded, Farm on Census|Farm) πS
= π(CML|Farm)π(Responded|CML, Farm)π(Farm
on Census|CML, Responded, Farm) πS
The probability of being included in the sample πS is
known for all responding farms. The other terms in
the probability of capturing a farm depend on the
characteristics of the farm. Using five-fold crossvalidation, three logistic models were developed
based on the matched dataset. The first model
estimated the probability of a farm being on the CML.
The second model estimated the probability that a
farm on the CML responded to the census report form.
The final model estimated the probability that a farm
that was on the CML and responded to the census was
identified as a farm based on its response. The
probability that a farm is captured by the census of
agriculture is then the product of the three conditional
probabilities that a farm is on the CML, responds, and
is identified as a farm.
Note 1: Responses were required for Must cases.
These operations were only excluded in modeling the
probability of a farm responding given that it was on
the CML.
Note 2: Because Alaska is not included in the JAS and
thus has no area frame, the Alaskan agricultural
operations were not included in the capture-recapture
process. No adjustments were made for
undercoverage or misclassification. To account for
nonresponse, the CML records were divided into
three groups: (1) the Must records, (2) the Criteria
Records, and (3) the remaining CML records. The
must records received a weight of one, thereby
receiving no adjustment for nonresponse. The
probability of response for each of the other two
groups was the proportion of responders within the
group. Each record within the group was then given a
weight equal to the reciprocal of the probability of
response.
Misclassification
An operation is misclassified if: (1) it meets the
definition of a farm, but is classified as a nonfarm on
the census or (2) it does not meet the definition of a
farm, but is classified as a farm on the census. The
first type of misclassification is accounted for when
modeling the probability of capture. An adjustment is
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

still needed for the misclassification of nonfarms as
farms. As with farm status and capture, the probability
of this misclassification depends on an operation’s
characteristics. Thus, a final logistic model was
developed. Given that an operation was classified as
a farm on the CML, the probability of its being a farm
was modeled based on its characteristics. Five-fold
cross-validation was used to ensure that the model
was not over-fitted.
CALIBRATION
Each operation identified as being in-scope on the
CML was given a weight equal to the probability of
misclassification divided by the probability of
capture. This weight accounted for undercoverage,
nonresponse, both types of misclassification, and the
nonresponse sample.
The record weighting processes were initially applied
at the State level to produce adjusted estimates of farm
numbers and land in farms for 63 different categories
of 8 characteristics of the farm operation or the farm
producer -- value of agricultural sales (9); age (2);
female; race (3); Hispanic origin of principal farm
producer; 4 sales categories for each of 10 major
commodities (40); and farm type groups (7). The
State-level number of farms and land in farms were
two additional adjusted estimates, resulting in 65
categories. To reduce the intercensal variation at the
State level, the State targets were smoothed by
averaging the 2017 estimates from capture-recapture
and the published 2012 State estimates with the
restrictions that the smoothed targets were within two
standard errors of the capture-recapture estimates.
The smoothed State targets were rescaled so that they
summed to the national capture-recapture estimates.
These State estimates were general purpose in that
they did not provide any control over expected levels
of commodity production of the individual farm
operation. As a result of this limitation, the procedures
could have over-adjusted or under-adjusted for
commodity production. To address this, a second set
of variables, known as commodity targets, was added
to the calibration algorithm. These targets were
commodity totals from administrative sources or from
NASS surveys of nonfarm populations (e.g. USDA
Farm Service Agency program data, Agricultural
Marketing Service market orders, livestock slaughter
data, cotton ginning data). The introduction of these
Appendix A A - 11

commodity coverage targets strengthened the overall
adjustment procedure by ensuring that major
commodity totals remained within reasonable bounds
of established benchmarks.
Each State was calibrated separately. The calibration
algorithm addressed commodity coverage. The
algorithm was controlled by the 65 State farm
operation coverage targets and the State commodity
coverage targets. Because calibration targets are
estimates subject to uncertainty, NASS allowed some
tolerance in the determination of the adjusted weights.
Rather than forcing the total for each calibration
variable computed using the adjusted weights to equal
a specific amount, NASS allowed the estimated total
to fall within a tolerance range.
Tolerance ranges for the farm operation coverage
targets were determined differently from the
commodity targets. The tolerance range for the 65
State farm operation coverage targets was the
estimated smoothed State total for the variable plus or
minus one standard error of the capture-recapture
estimate. This choice limited the cumulative deviation
from the estimated total for a variable when State
totals were summed to a U.S. total. Commodity
coverage targets with acceptable ranges were
established based on the administrative source for
each State. Ranges were not necessarily symmetric
around the target value.
To ensure that all subdomains for which NASS
publishes summed to their grand total, integer weights
were produced by a discrete calibration algorithm.
This eliminated the need for rounding individual cell
values and ensured that marginal totals always added
correctly to the grand total. If a weight was initially
not in the interval [1,6], it was trimmed so that in was
in that interval. That is, adjusted weights less than 1
were set to 1, and those greater than 6 were set to 6.
The remaining non-integer weights were then
rounded sequentially to reduce the distance of the
estimated totals from the targets.
Calibration adjustments began with the computation
of a priority index for each record. The priority index
was the absolute value of the gradient of the relative
error associated with increasing or decreasing a
record’s weight by one. The record with the highest
priority index was then selected as a candidate to
increase or decrease its weight by one to reduce the
A - 12 Appendix A

cumulative distance from the targets as measured by
the relative error. If the new value produced an
improvement and satisfied the range restrictions, the
weight was updated and new priorities were assigned;
otherwise, the record with the next highest priority
index was processed. This process was iteratively
performed until convergence was attained. Because
census data collection was assumed to be complete for
very large and unique farms, their weights were
controlled to 1 during the calibration adjustment
process. For all other farms, the final census record
weights were forced to be an integer number in the
interval [1, 6]. The calibration process considered all
targets simultaneously through the priority index.
Although calibration was seldom able to adjust
weights so that all State targets were met, all targets
were brought collectively as close to the targets as
possible.
The proportions of selected census data items that
were due to coverage, response, and classification
adjustments are displayed in Tables A and C.
DISCLOSURE REVIEW
After tabulation and review of the aggregates, a
comprehensive disclosure review was conducted.
NASS is obligated to withhold, under Title 7, U.S.
Code, any total that would reveal an individual’s
information or allow it to be closely estimated by the
public. Farm counts are not considered sensitive and
are not subject to disclosure controls. Cell suppression
was used to protect the cells that were determined to
be sensitive to a disclosure of information.
Based on agency standards, data cells were
determined to be sensitive to a disclosure of
information if they failed either of two rules. The
threshold rule failed if the data cell contained less than
three operations. For example, if only one farmer
produced turkeys in a county, NASS could not
publish the county total for turkey inventory without
disclosing that individual’s information. The
dominance rule failed if the distribution of the data
within the cell allowed a data user to estimate any
respondent’s data too closely. For example, if there
are many farmers producing turkeys in a county and
some of them were large enough to dominate the cell
total, NASS could not publish the county total for
turkey inventory without risking disclosing an
individual respondent’s data. In both of these
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

situations, the data were suppressed and a “(D)” was
placed in the cell in the census publication table.
These data cells are referred to as primary
suppressions.
Since most items were summed to marginal totals,
primary suppressions within these summation
relationships were protected by ensuring that there
were additional suppressions within the linear
relationship that provided adequate protection for the
primary. A detailed computer routine selected
additional data cells for suppression to ensure all
primary suppressions were properly protected. These
data cells are referred to as complementary
suppressions. These cells are not themselves sensitive
to a disclosure of information but were suppressed to
protect other primary suppressions. A “(D)” was also
placed in the cell of the census publication table to
indicate a complementary suppression. A data user
cannot determine whether a cell with a (D) represents
a primary or a complementary suppression.
Regional field office analysts reviewed all
complementary suppressions to ensure no cells had
been withheld that were vital to the data users. In
instances where complementary suppressions were
deemed critically important to a State or county,
analysts requested an override and a different
complementary cell was chosen.

high level of quality. The quality of a census may be
measured in many ways. One of the first indicators
used is a measure of the response to the census data
collection as it has generally been thought that a high
response rate indicates more complete coverage of the
population of interest. This is a valid assumption if the
enumeration list, the CML here, has complete
coverage of the population of interest. In the case of
the census of agriculture, the definition requiring
advance knowledge of sales makes achieving a high
level of coverage difficult. To ensure that the census
of agriculture is as complete as possible, records are
included that might not meet the census definition of
a farm – in fact, almost 50 percent more records than
the anticipated number of qualifying farm operations
were included in the 2017 CML. A second indicator
of quality then is the coverage of the farm population
by the CML. Other indicators of quality relate to the
accuracy and completeness of the data, and the
validity of the procedures used in processing the data.
In some cases, NASS was able to produce measures
of quality – such as the response rate to the data
collection, the coverage of the census mail list, and
the variability of the final adjusted estimates. In other
cases, measures were not produced but descriptions of
procedures that NASS used to reduce errors from the
procedures were subsequently provided.
Census Response Rate

CENSUS QUALITY
The purpose of the census of agriculture is to account
for “any place from which $1,000 or more of
agricultural products were produced and sold, or
normally would have been sold, during the census
year.” To accomplish this, NASS develops a CML
that contains identifying information for operations
that have an indication of meeting the census
definition, develops procedures to collect agricultural
information from those records, establishes criteria
for analyst review of the data, creates computer
routines to correct or complete the requested
information, and provides census estimates of the
characteristics of farms and farm producers with
associated measures of uncertainty.

The response rate is one indicator of the quality of a
data collection. It is generally assumed that if a
response rate is close to a full participation level of
100 percent, the potential for nonresponse bias is
small, although this has been questioned in the
literature. The response rate for the 2017 Census of
Agriculture CML was 71.8 percent, as compared with
the 2012 Census of Agriculture’s response rate of 74.6
percent and 78.2 percent for the 2007 Census of
Agriculture.
The 2017 Census of Agriculture’s response rate used
the fourth response rate formula (RR4) from the
American Association of Public Opinion Research’s
Response Rate Standard Definitions manual:

It is not likely that either the CML includes all
operations that meet the definition of a farm or that all
those that do meet the definition of a farm respond to
the census inquiry. The goal is to publish data with a
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Appendix A A - 13

where

Census Coverage

Cadj = number of fully and partially completed
records, excluding replicated records
R = number of explicit refusals
NC = number of non-contacted operations known to
be eligible
O = number of other types of nonrespondents
Replicated = number of replicated records
U = number of operations of unknown eligibility
e(U) = estimated number of operations of unknown
eligibility assumed to be eligible

As a side-product of the statistical adjustment used to
account for undercoverage, nonresponse of farms on
the CML, and misclassification of responses to the
census, the proportion of the adjustments due to each
of those factors can be derived. The percentages of
final census estimates due to adjustments for
undercoverage, nonresponse, and misclassification as
well as the total percent adjustment for selected items
are displayed in Tables A and C.

Records were classified into the above variables
based on the combination of their active status (AS)
codes, in-scope status, and replication status. Active
status refers to the eligibility status of records for
selection on the CML. All replicated records were
considered to be a form of nonresponse and were
classified into other nonrespondents; in-scope status
was considered immaterial.
Certain active status classifications indicated records
of unknown agricultural status. These classifications
included records to be removed from the CML but had
data from outside sources indicating agricultural
activity, new records from outside data sources,
nonrespondents and refusals to the NACS, records for
regional office handling only, and records with Farm
Service Agency or Conservation Reserve Program
data on operations that are not owned by the principal
producer. These records were stratified (grouped)
based on their probabilities of being in-scope had they
responded. The estimated number of in-scope
nonrespondents was calculated for the hth stratum
(group) by the following formula:

MEASURED ERRORS IN THE CENSUS
PROCESS
Although the census of agriculture does not inherently
rely on a sample, NASS used a national nonresponse
sample as part of its follow-up efforts in 2017. In
addition to the uncertainty introduced by the
nonresponse sample, NASS uses statistical
procedures in compiling the CML, in its data
collection procedures, in data editing and processing,
and in compiling the final data. Additionally, it uses
statistical procedures to both measure errors in the
various processes and in making adjustments for
those errors in the final data. One example is the
statistical process used to account for undercoverage,
nonresponse of farms on the CML, and
misclassification of responses to the census. The basis
of the undercoverage adjustment is the capturerecapture procedure that uses the area sample
enumeration from the JAS. The largest contributors to
error in the census estimates are due to the
adjustments for nonresponse, undercoverage,
misclassification, calibration, and integerization.
Variability in Census Estimates due to
Statistical Adjustment

where
e(Uh) = estimated number of operations of unknown
eligibility assumed to be eligible in the hth group
Cin-scope,h = the number of completed and in-scope
census records in the hth group
Ch = the number of completed census records in the
hth group
Uh = number of operations of unknown eligibility in
the hth group
A - 14 Appendix A

In conducting the 2017 Census of Agriculture, efforts
were initiated to measure error associated with the
adjustments for farm operations that were not on the
CML, for farm operations that were on the CML but
did not respond to the census report form, and for
farms and nonfarms that were misclassified as
nonfarms and farms, respectively, for calibration.
These error measurements were developed from the
standard error of the estimates at the national, State,
and county levels and were expressed as coefficients
of variation (CVs) at the national and State levels and
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

as generalized coefficients of variation (GCVs) at the
county levels.
The standard error of an estimate is an estimate of the
standard deviation of the sampling distribution of the
estimator. Because Alaska was modeled separately
from the other States, the variances of a national-level
data item for this State was computed separately and
added to the variance of that data item for the rest of
the U.S. The standard error was then the square root
of the total variance. In each case, standard errors
were computed using an approach based on a
combination of group jackknife and bootstrap
methodologies. To conduct the jackknifing, k = 10
mutually exclusive and exhaustive groups of JAS
segments were formed. The groups were selected
using a stratified random design so that each group
reflected the survey design, including State and
agricultural strata within a State. The weight of record
i in jackknife group j is CRi(j )for j = 1, 2, …, k. Based
on these weights, a group jackknife estimator to
estimate the variance would account for the
uncertainty associated with modeling the capturerecapture probabilities. To account for the additional
uncertainty due to calibration, the weights within each
jackknife group were transformed through bootstrap
simulation; these transformed weights are called
calibration-adjusted-jackknife weights. The full
dataset, which is composed of the records of all
responding farms on the CML, is calibrated as
described in the Calibration section, and the final
calibration-adjusted weight of record i is denoted by
ŵi. For each record i in jackknife group k, the
calibration-adjusted-jackknife weights of that record
can be approximated as wi(j)=ai(j)CRi(j) where ai(j) ~
N(1,( ŵi – 1) / ŵi). The bootstrap process simulated the
value of the adjustment ai(j) for each record on the
CML to obtain the calibration-adjusted-jackknife
weights. For a given data item, such as the number of
farms, the estimate T(j) was computed at the specified
geographical level, such as nation, State, or county,
using the (k – 1) groups remaining after deleting the
calibration-adjusted jackknife group j. Estimates of
the variance and standard error associated with the
estimator Ti are then, respectively,
l
k  1 k   j  k Ti   
2
 
 Ti  
 ; SE Ti    i

k j 1 
l 1 k 
2

2
i

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Increasing k improves the estimate of the variance but,
as k increases, the observations become too sparse to
reflect the survey design and to provide countrywide
coverage. Ten (10) calibration-adjusted jackknife
groups were used to provide standard errors for 2017
State and national estimates. For the estimate of the
number of farms with a given set of characteristics,
only the CML records with those characteristics were
used to obtain the overall estimate as well as the
estimates from each calibration-adjusted jackknife
group.
Note that the calibrated jackknife groups were only
constructed once, and different subsets of the records
were used to compute estimates and standard errors
for the data items.
The CV is a measure of the relative amount of error
associated with the sample estimate:
SE Ti 
100%
Ti
where SE(Ti) is the standard error of the capturerecapture estimate for data item i. This relative
measure allows the reliability of a range of estimates
to be compared. For example, the standard error is
often larger for large population estimates than for
small population estimates, but the large population
estimates may have a smaller CV, indicating a more
reliable estimate. For county-level estimates, a
generalized coefficient of variation (GCV) was
determined for each estimate within a State. A
generalized variance function relates a function of the
variance of an estimator to a function of the estimator.
Within a State, the standard error of an estimate for a
data item was often found to be linearly related to the
estimate of that item with an intercept of zero. Based
on this modeled relationship, the GCV is the slope of
the line relating the standard error to the estimate,
multiplied times 100 to represent the GCV as a
percentage.
CVi 

The standard error is the product of the CV (or GCV
for county estimates) and the estimate divided by 100.
As an example, if the GCV for a State is 25 percent
and a county’s estimate is 4, then the standard error is
25(4)/100 = 1. The standard error of an estimated data
item from the census provides a measure of the error
variation in the value of that estimated data item based
on the possible outcomes of the census collection,
Appendix A A - 15

including variants as to who was on the CML, who
returned a census form, who was misclassified either
as a farm or as a nonfarm, and the uncertainty
associated with calibration and integerization. With
95 percent confidence, an estimate is within two
standard errors of the true value being estimated. For
this example, with 95 percent confidence, the estimate
of 4 is within 2(1) = 2 of the true county value.
Table B presents the fully adjusted estimates with the
coefficient of variation for selected items.
NONMEASURED ERRORS IN THE CENSUS
PROCESS
As noted in the previous section, sampling errors can
be introduced from the coverage, nonresponse and
misclassification adjustment procedures. This error is
measureable. However, nonsampling errors are
imbedded in the census process that cannot be directly
measured as part of the design of the census but must
be contained to ensure an accurate count. Extensive
efforts were made to compile a complete and accurate
mail list for the census, to elicit response to the
census, to design an understandable report form with
clear instructions, to minimize processing errors
through the use of quality control measures, to reduce
matching error associated with the capture-recapture
estimation process, and to minimize error associated
with identification of a respondent as a farm operation
(referred to as classification error). The weight
adjustment and tabulation processes recognize the
presence of nonsampling errors; however, it is
assumed that these errors are small and that, in total,
the net effect is zero. In other words, the positive
errors cancel the negative errors.
Respondent and Enumerator Error
Incorrect or incomplete responses to the census report
form or to the questions posed by an enumerator can
introduce error into the census data. Steps were taken
in the design and execution of the census of
agriculture to reduce errors from respondent
reporting. Poor instructions and ambiguous
definitions lead to misreporting. Respondents may not
remember accurately, may estimate responses, or may
record an item in the wrong cell. To reduce reporting
and recording errors, the report form was tested prior
to the census using industry accepted cognitive testing
procedures. Detailed instructions for completing the
A - 16 Appendix A

report form were provided to each respondent.
Questions were phrased as clearly as possible based
on previous tests of the report form. Computerassisted telephone interviewing software included
immediate integrity checks of recorded responses so
suspect data could be verified or corrected. In
addition, each respondent’s answers were checked for
completeness and consistency by the complex edit
and imputation system.
Processing Error
Processing of each census report form was another
potential source of nonsampling error. All mail
returns that included multiple reports, respondent
remarks, or that were marked out of business and
report forms with no reported data were sent to an
analyst for verification and appropriate action.
Integrity checks were performed by the imaging
system and data transfer functions. Standard quality
control procedures were in place that required that
randomly selected batches of data keyed from image
be re-entered by a different operator to verify the work
and evaluate key entry operators. All systems and
programs were thoroughly tested before going on-line
and were monitored throughout the processing period.
Developing accurate processing methods is
complicated by the complex structure of agriculture.
Among the complexities are the many places to be
included, the variety of arrangements under which
farms are operated, the continuing changes in the
relationship of producers to the farm operated, the
expiration of leases and the initiation or renewal of
leases, the problem of obtaining a complete list of
agriculture operations, the difficulty of contacting and
identifying some types of contractor/contractee
relationships, the producer’s absence from the farm
during the data collection period, and the producer’s
opinion that part or all of the operation does not
qualify and should not be included in the census.
During data collection and processing of the census,
all operations underwent a number of quality control
checks to ensure results were as accurate as possible.
Item Nonresponse
All item nonresponse actions provide another
opportunity to introduce measurement errors.
Regardless of whether it was previously reported data,
administrative data, the nearest neighbor algorithm,
2017 Census of Agriculture
USDA, National Agricultural Statistics Service

the fully conditional specification method, or
manually imputed by an analyst, some risk exists that
the imputed value does not equal the actual value.
Previously reported and administrative data were used
only when they related to the census reference period.
A new nearest neighbor was randomly selected for
each incident to eliminate the chance of a consistent
bias.
Record Matching Error
The process of building and expanding the CML
involves finding new list sources and checking for
names not on the list. An automated processing
system compared each new name to the existing CML
names and “linked” like records for the purpose of
preventing duplication. New names with strong links
to a CML name were discarded and those with no
links were added as potential farms. Names with weak
links, possible matches, were reviewed by staff to
determine whether the new name should be added.
Despite this thorough review, some new names may
have been erroneously added or deleted. Additions
could contribute to duplication (overcoverage)
whereas deletions could contribute to undercoverage.
As a result, some names received more than one
report form, and some farm producers did not receive
a report form. Respondents were instructed to

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

complete one form and return all forms so the
duplication could be removed.
Another chance for error came when comparing June
Area Survey tract producer names to the CML. Area
producers whose names were not found on the CML
were part of the measure of list incompleteness, or
NML. Mistakes in determining overlap status resulted
in overcounts (including a tract whose producer was
on the CML) or undercounts (excluding a tract whose
producer was not on the CML). All tracts determined
to not be on the list were triple checked to eliminate,
or at least minimize, any error. NML tract producers
were mailed a report form printed in a different color.
In order to attempt to identify duplication, all
respondents who received multiple report forms were
instructed to complete the CML version and return all
forms so duplication could be removed.
Records in the 2017 JAS were matched to the 2017
census using probabilistic record linkage. The records
of operations with differing farm status were sent out
to be reviewed by NASS regional field offices. If farm
status could not be resolved, the probability of an
operation being a farm was imputed using a missing
data model. The uncertainty associated with this
estimate, with the exception of model uncertainty,
was accounted for, but errors not found through this
process were not.

Appendix A A - 17

Table A. Summary of U.S. Coverage, Nonresponse, and Misclassification Adjustments: 2017
[For meaning of abbreviations and symbols, see introductory text.]
Item

Total

Standard
error

Adjustment
as percent
of total

Percent of total
adjustment
from coverage

Percent of total
adjustment from
nonresponse

Percent of total
adjustment from
misclassification

Farms .......................................................................................... number
Land in farms .................................................................................. acres

2,042,220
900,217,576

43,278
15,031,334

37.6
22.7

15.1
4.4

13.9
11.8

8.6
6.5

Farms by size:
1 to 9 acres ................................................................................. farms
acres
10 to 49 acres ............................................................................. farms
acres
50 to 69 acres ............................................................................. farms
acres
70 to 99 acres ............................................................................. farms
acres
100 to 139 acres ......................................................................... farms
acres
140 to 179 acres ......................................................................... farms
acres
180 to 219 acres ......................................................................... farms
acres
220 to 259 acres ......................................................................... farms
acres
260 to 499 acres ......................................................................... farms
acres
500 to 999 acres ......................................................................... farms
acres
1,000 to 1,999 acres ................................................................... farms
acres
2,000 acres or more .................................................................... farms
acres

273,325
1,302,208
583,001
14,787,940
135,126
7,845,508
163,251
13,414,191
149,478
17,343,842
116,908
18,399,918
74,086
14,645,228
57,096
13,586,644
183,835
65,775,717
133,321
92,872,530
87,666
120,680,141
85,127
519,563,709

23,216
119,480
27,053
728,067
2,902
169,522
1,480
117,725
2,564
312,035
4,263
669,574
1,276
255,063
1,560
372,774
3,483
1,244,202
1,651
1,164,377
2,592
3,602,322
2,002
16,173,465

57.2
57.8
42.4
41.3
34.6
34.6
33.3
33.2
32.1
32.1
31.5
31.5
28.2
28.2
27.9
27.9
29.3
29.4
30.1
30.3
30.2
30.2
24.6
16.8

25.9
25.1
19.1
18.0
14.3
14.2
12.3
12.2
10.9
10.9
9.7
9.7
9.9
9.9
9.5
9.5
8.4
8.3
8.5
8.3
4.2
4.2
2.8
1.9

17.6
17.0
13.8
13.2
13.4
13.3
13.1
13.1
12.3
12.2
11.5
11.5
13.5
13.6
13.4
13.4
14.9
15.0
17.6
17.9
18.1
18.0
18.2
8.9

13.8
15.7
9.5
10.1
7.0
7.0
7.9
7.9
8.9
9.0
10.2
10.2
4.8
4.8
5.0
5.0
6.0
6.1
4.1
4.1
7.8
8.0
3.5
6.0

Irrigated land use:
Harvested cropland ..................................................................... farms
acres
Pastureland and other land ......................................................... farms
acres

255,348
53,959,077
64,450
4,054,830

7,306
1,097,813
3,314
170,744

35.2
22.5
44.9
24.8

13.6
2.7
19.6
6.0

15.1
16.0
15.0
12.5

6.6
3.8
10.2
6.3

Market value of agricultural products
sold (see text) .............................................................................. $1,000

388,522,695

2,923,858

18.1

3.4

11.1

3.6

Farms by value of sales:
Less than $1,000 (see text) ......................................................... farms
$1,000
$1,000 to $2,499 ......................................................................... farms
$1,000
$2,500 to $4,999 ......................................................................... farms
$1,000
$5,000 to $9,999 ......................................................................... farms
$1,000
$10,000 to $19,999 ..................................................................... farms
$1,000
$20,000 to $24,999 ..................................................................... farms
$1,000
$25,000 to $39,999 ..................................................................... farms
$1,000
$40,000 to $49,999 ..................................................................... farms
$1,000
$50,000 to $99,999 ..................................................................... farms
$1,000
$100,000 to $249,999 ................................................................. farms
$1,000
$250,000 to $499,999 ................................................................. farms
$1,000
$500,000 to $999,999 ................................................................. farms
$1,000
$1,000,000 or more ..................................................................... farms
$1,000

603,752
93,210
187,949
310,520
185,341
662,980
208,074
1,477,595
174,780
2,468,212
53,438
1,181,954
100,490
3,162,749
43,623
1,937,293
119,434
8,477,635
130,932
21,171,316
87,839
31,318,548
69,703
49,338,998
76,865
266,921,684

26,259
5,223
8,512
13,915
4,777
17,382
5,255
36,707
4,230
57,415
864
18,406
2,235
65,497
645
29,399
2,473
170,770
1,810
275,035
1,376
505,169
1,012
761,022
922
2,573,412

51.2
60.6
42.2
42.1
38.5
38.3
37.1
36.9
25.7
25.7
26.4
26.3
26.3
26.3
27.5
27.5
28.0
28.1
27.6
27.9
30.9
31.1
30.9
31.1
20.7
12.6

22.1
26.5
20.3
20.2
18.4
18.3
17.0
16.8
9.3
9.3
9.3
9.3
7.4
7.4
7.8
7.8
7.3
7.2
3.8
3.7
3.2
3.1
2.2
2.3
2.9
3.0

15.1
18.1
14.0
14.0
12.8
12.8
12.9
12.9
11.1
11.1
11.6
11.6
13.6
13.6
14.2
14.2
15.2
15.4
17.9
18.3
22.9
23.3
25.9
26.0
14.7
6.5

14.0
16.1
7.8
7.9
7.2
7.2
7.2
7.2
5.3
5.3
5.4
5.4
5.3
5.3
5.5
5.5
5.5
5.5
5.9
5.9
4.8
4.8
2.8
2.8
3.2
3.2

1,751,126
541,071,476
130,173
158,051,459

39,037
6,917,125
3,658
3,722,094

38.5
27.3
31.6
17.7

15.8
6.1
9.4
2.4

13.9
15.0
14.9
9.3

8.8
6.2
7.3
6.0

104,155
126,671,963
12,685
12,889,821

1,350
6,438,831
626
1,116,353

31.3
17.1
34.2
11.8

10.2
2.1
12.3
1.7

14.1
8.0
14.3
4.4

7.0
7.0
7.5
5.7

44,081
61,532,857

2,263
3,091,251

33.7
8.9

13.6
2.0

11.4
2.5

8.7
4.4

Tenure:
Full owners .................................................................................. farms
acres
Part owners ................................................................................. farms
acres
Tenants ....................................................................................... farms
acres

1,408,961
310,218,983
493,137
503,138,279
140,122
86,860,314

33,785
6,470,335
7,102
8,927,625
6,570
1,990,706

39.3
21.9
31.5
22.8
41.5
24.8

16.9
5.9
8.9
2.9
14.7
5.5

12.8
8.2
16.8
14.0
20.2
15.6

9.6
7.8
5.9
5.9
6.7
3.6

All principal producer characteristics by 1Sex of operator:
Male ........................................................................................ farms
acres
Female .................................................................................... farms
acres

1,787,998
847,232,627
766,474
238,157,861

39,842
14,008,443
21,918
6,484,070

36.2
22.4
43.0
25.1

14.1
4.0
17.0
5.2

14.2
12.1
15.0
12.8

7.9
6.3
10.9
7.2

Primary occupation:
Farming ................................................................................... farms
Other ....................................................................................... farms

1,207,375
1,533,078

18,039
55,236

34.1
41.3

11.5
16.2

14.4
15.5

8.1
9.7

Legal status for tax purposes (see text):
Family or individual ..................................................................... farms
acres
Partnership .................................................................................. farms
acres
Corporation:
Family held .............................................................................. farms
acres
Other than family held ............................................................. farms
acres
Other - estate or trust, prison farm, grazing association,
American Indian Reservation, etc ............................................. farms
acres

See footnote(s) at end of table.

A - 18 Appendix A

--continued

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Table A. Summary of U.S. Coverage, Nonresponse, and Misclassification Adjustments: 2017 (continued)
[For meaning of abbreviations and symbols, see introductory text.]
Item

Total

Standard
error

Adjustment
as percent
of total

Percent of total
adjustment
from coverage

Percent of total
adjustment from
nonresponse

Percent of total
adjustment from
misclassification

All principal producer characteristics by 1- - Con.
Hispanic, Latino, or
Spanish origin (see text) ........................................................... farms
acres

77,416
26,041,600

7,488
1,211,639

55.9
26.7

23.6
7.5

21.5
11.1

10.7
8.1

Race:
American Indian or
Alaska Native ........................................................................ farms
acres
Asian ....................................................................................... farms
acres
Black or African American ....................................................... farms
acres
Native Hawaiian or
Other Pacific Islander ........................................................... farms
acres
White ...................................................................................... farms
acres
More than one race reported .................................................. farms
acres

39,632
51,095,994
13,904
1,831,229
32,052
3,862,936

4,690
1,971,823
1,191
141,253
2,720
288,775

52.3
14.6
47.4
27.8
59.9
52.5

17.2
3.8
14.7
6.0
12.9
7.5

22.4
6.5
21.6
14.9
31.2
32.6

12.7
4.3
11.1
6.9
15.7
12.4

2,092
426,068
1,955,737
843,497,615
19,773
6,712,435

434
228,386
40,615
14,192,177
1,469
319,035

47.9
39.9
36.9
23.0
47.7
24.5

15.3
7.8
15.1
4.4
17.9
4.4

20.5
23.4
13.4
12.0
19.1
12.8

12.1
8.7
8.4
6.6
10.8
7.3

Military service (see text):
Never served ................................................................... producers
Served ............................................................................. producers

2,402,342
338,111

63,141
9,391

38.4
36.1

14.2
14.5

15.3
11.9

8.9
9.7

All producers by age group 1:
Under 25 years ........................................................................... farms
25 to 34 years ............................................................................. farms
35 to 44 years ............................................................................. farms
45 to 54 years ............................................................................. farms
55 to 64 years ............................................................................. farms
65 to 74 years ............................................................................. farms
75 years and over ....................................................................... farms

50,943
234,496
390,345
614,654
955,354
757,936
396,106

8,438
27,511
18,472
27,200
12,836
13,931
7,027

49.4
50.5
44.0
40.4
35.9
34.4
31.4

13.2
16.8
16.8
13.9
15.0
14.3
12.3

24.3
22.1
20.4
17.7
12.8
9.6
9.2

11.9
11.6
6.8
8.8
8.0
10.5
10.0

Net cash farm income of operations (see text):
Farms with gains of 2Less than $1,000 .................................................................... farms
$1,000
$1,000 to $4,999 ..................................................................... farms
$1,000
$5,000 to $9,999 ..................................................................... farms
$1,000
$10,000 to $24,999 ................................................................. farms
$1,000
$25,000 to $49,999 ................................................................. farms
$1,000
$50,000 or more ..................................................................... farms
$1,000

66,633
31,436
156,683
431,683
103,942
756,426
153,619
2,525,811
114,269
4,097,569
296,183
104,245,583

1,074
699
1,875
5,726
2,032
15,498
2,497
39,809
2,015
73,458
3,414
1,009,852

35.7
34.5
30.7
30.3
27.1
27.0
26.1
26.1
26.5
26.5
26.9
20.0

16.6
16.0
13.7
13.4
10.6
10.5
8.4
8.3
6.9
6.8
4.0
3.3

11.3
11.0
10.3
10.3
10.4
10.5
11.8
11.9
13.7
13.9
18.2
12.9

7.8
7.5
6.7
6.6
6.0
6.0
5.9
5.9
5.8
5.8
4.7
3.8

Farms with losses of Less than $1,000 .................................................................... farms
$1,000
$1,000 to $4,999 ..................................................................... farms
$1,000
$5,000 to $9,999 ..................................................................... farms
$1,000
$10,000 to $24,999 ................................................................. farms
$1,000
$25,000 to $49,999 ................................................................. farms
$1,000
$50,000 or more ..................................................................... farms
$1,000

89,302
45,846
342,608
988,554
256,919
1,854,855
272,079
4,266,566
104,865
3,629,228
85,118
13,380,008

3,454
1,742
13,270
41,411
9,567
71,776
10,940
188,227
3,872
133,023
2,038
254,975

42.8
43.4
46.6
47.0
47.3
47.2
45.5
45.3
42.1
42.0
35.7
30.2

19.3
19.4
20.9
21.0
20.7
20.6
19.1
18.8
16.5
16.3
11.7
8.6

13.2
13.5
14.5
14.8
15.6
15.6
15.5
15.6
15.7
15.7
16.4
15.3

10.3
10.5
11.2
11.2
11.0
11.0
11.0
11.0
10.0
10.0
7.6
6.3

Livestock and poultry:
Cattle and calves inventory ......................................................... farms
number
Beef cows inventory ................................................................ farms
number
Milk cows inventory ................................................................. farms
number
Hog and pigs inventory ............................................................... farms
number
Layers inventory ........................................................................ farms
number
Broilers sold ................................................................................ farms
number
Aquaculture sold ......................................................................... farms
$1,000

882,692
93,648,041
729,046
31,722,039
54,599
9,539,631
66,439
72,381,007
232,500
368,241,393
32,751
8,889,759,283
5,350
1,778,587

19,877
1,983,371
14,946
809,066
1,722
161,118
3,424
1,322,671
10,221
10,596,560
1,582
248,694,312
201
88,052

36.1
22.8
34.4
24.7
32.8
11.3
42.7
24.5
50.0
1.4
41.8
27.2
28.4
6.1

13.5
4.0
12.6
4.2
9.3
2.3
17.2
7.1
21.4
0.5
16.8
7.9
14.1
2.8

16.3
13.7
15.6
15.1
20.5
7.7
17.3
8.8
18.0
0.5
17.3
13.0
9.3
1.5

6.3
5.1
6.2
5.4
3.0
1.3
8.2
8.5
10.5
0.4
7.8
6.3
4.9
1.7

Selected crops harvested:
Corn for grain .............................................................................. farms
acres
Durum wheat for grain ................................................................ farms
acres
Other spring wheat for grain (see text) ....................................... farms
acres
Winter wheat for grain ................................................................. farms
acres
Sorghum for grain ....................................................................... farms
acres
Soybeans for beans .................................................................... farms
acres
Rice ............................................................................................ farms
acres
Cotton ......................................................................................... farms
acres
Peanuts ...................................................................................... farms
acres

304,801
84,738,562
3,093
2,206,169
20,076
10,419,033
86,596
26,186,417
15,339
5,070,159
303,191
90,149,480
4,637
2,395,054
16,149
11,401,965
6,379
1,786,767

4,815
1,097,857
161
117,367
517
511,562
1,103
210,542
345
155,826
3,615
1,746,145
466
638,071
610
316,506
400
134,399

27.0
24.2
23.2
19.7
27.9
25.5
25.4
21.9
27.0
25.5
27.1
25.3
27.8
19.9
28.0
25.6
33.2
28.1

5.4
2.3
2.9
1.9
3.8
2.3
5.0
2.7
4.7
2.7
5.6
2.5
2.4
1.1
4.0
2.4
4.8
2.4

16.9
18.8
17.0
14.7
20.3
19.4
15.9
15.3
17.9
19.1
16.8
19.5
21.3
15.2
20.6
20.4
23.5
22.2

4.7
3.2
3.3
3.1
3.8
3.8
4.5
3.9
4.4
3.8
4.7
3.3
4.1
3.7
3.4
2.9
5.0
3.5

See footnote(s) at end of table.

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

--continued

Appendix A A - 19

Table A. Summary of U.S. Coverage, Nonresponse, and Misclassification Adjustments: 2017 (continued)
[For meaning of abbreviations and symbols, see introductory text.]
Item

Total

Standard
error

Adjustment
as percent
of total

Percent of total
adjustment
from coverage

Percent of total
adjustment from
nonresponse

Percent of total
adjustment from
misclassification

Selected crops harvested: - Con.

1
2

Barley .......................................................................................... farms
acres
Oats ............................................................................................ farms
acres

11,188
2,206,808
19,842
814,140

287
111,622
450
23,504

27.0
22.1
30.9
28.3

4.7
2.1
7.6
4.7

18.0
16.7
17.5
19.0

4.2
3.4
5.8
4.7

Forage - land used for all hay and all
haylage, grass silage, and
greenchop (see text) ................................................................. farms
acres
Land in vegetables (see text) ...................................................... farms
acres
Potatoes .................................................................................. farms
acres
Tomatoes in the open ............................................................. farms
acres
Sweet corn .............................................................................. farms
acres
Lettuce .................................................................................... farms
acres
Land in orchards (see text) ......................................................... farms
acres
Apples ..................................................................................... farms
acres
Grapes .................................................................................... farms
acres
Oranges .................................................................................. farms
acres
Almonds .................................................................................. farms
acres
Land in berries ............................................................................ farms
acres

799,627
56,858,622
74,276
3,965,622
16,554
1,133,128
28,673
335,348
20,784
496,096
10,869
342,965
111,955
5,665,600
26,408
381,718
28,387
1,136,155
7,973
602,830
7,954
1,266,160
33,919
302,199

15,837
729,705
4,298
102,015
1,099
36,604
1,806
21,300
1,179
16,260
949
7,049
3,892
210,257
1,296
18,529
911
69,903
267
29,323
348
35,893
1,472
6,508

34.5
28.2
37.0
10.6
33.7
6.6
37.2
6.2
32.5
13.2
39.3
6.5
36.0
20.1
35.6
14.2
34.3
22.5
36.2
11.3
33.2
20.7
36.4
13.3

12.9
6.2
14.6
1.8
13.6
1.1
15.2
1.3
11.9
2.3
16.6
2.3
17.3
4.5
17.0
4.2
16.9
4.0
16.7
2.4
10.9
3.9
16.9
4.6

12.9
16.2
17.2
6.9
15.5
4.4
17.0
3.3
16.3
8.1
17.4
2.0
13.1
11.9
12.9
7.2
12.0
15.0
14.0
6.2
17.4
12.7
14.0
5.9

8.7
5.9
5.2
1.9
4.6
1.2
5.0
1.6
4.3
2.8
5.2
2.1
5.5
3.8
5.7
2.7
5.3
3.5
5.5
2.7
5.0
4.1
5.6
2.7

Data were collected for a maximum of four producers per farm.
Farms with total production expenses equal to market value of agricultural products sold, government payments, and farm-related income are included as farms with gains of less than $1,000.

A - 20 Appendix A

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Table B. Reliability Estimates of U.S. Totals: 2017
[For meaning of abbreviations and symbols, see introductory text.]
Item

Total

Coefficient
of variation
(percent)

Farms ................................................................................. number
Land in farms .......................................................................... acres

2,042,220
900,217,576

2.1
1.7

Farms by size:
1 to 9 acres .........................................................................farms
acres
10 to 49 acres .....................................................................farms
acres
50 to 69 acres .....................................................................farms
acres
70 to 99 acres .....................................................................farms
acres
100 to 139 acres .................................................................farms
acres
140 to 179 acres .................................................................farms
acres
180 to 219 acres .................................................................farms
acres
220 to 259 acres .................................................................farms
acres
260 to 499 acres .................................................................farms
acres
500 to 999 acres .................................................................farms
acres
1,000 to 1,999 acres ...........................................................farms
acres
2,000 acres or more ............................................................farms
acres

273,325
1,302,208
583,001
14,787,940
135,126
7,845,508
163,251
13,414,191
149,478
17,343,842
116,908
18,399,918
74,086
14,645,228
57,096
13,586,644
183,835
65,775,717
133,321
92,872,530
87,666
120,680,141
85,127
519,563,709

8.5
9.2
4.6
4.9
2.1
2.2
0.9
0.9
1.7
1.8
3.6
3.6
1.7
1.7
2.7
2.7
1.9
1.9
1.2
1.3
3.0
3.0
2.4
3.1

Irrigated land use:
Harvested cropland .............................................................farms
acres
Pastureland and other land .................................................farms
acres

255,348
53,959,077
64,450
4,054,830

2.9
2.0
5.1
4.2

Market value of agricultural products
sold (see text) .....................................................................$1,000

388,522,695

0.8

Farms by value of sales:
Less than $1,000 (see text) ................................................farms
$1,000
$1,000 to $2,499 .................................................................farms
$1,000
$2,500 to $4,999 .................................................................farms
$1,000
$5,000 to $9,999 .................................................................farms
$1,000
$10,000 to $19,999 .............................................................farms
$1,000
$20,000 to $24,999 .............................................................farms
$1,000
$25,000 to $39,999 .............................................................farms
$1,000
$40,000 to $49,999 .............................................................farms
$1,000
$50,000 to $99,999 .............................................................farms
$1,000
$100,000 to $249,999 .........................................................farms
$1,000
$250,000 to $499,999 .........................................................farms
$1,000
$500,000 to $999,999 .........................................................farms
$1,000
$1,000,000 or more ............................................................farms
$1,000

603,752
93,210
187,949
310,520
185,341
662,980
208,074
1,477,595
174,780
2,468,212
53,438
1,181,954
100,490
3,162,749
43,623
1,937,293
119,434
8,477,635
130,932
21,171,316
87,839
31,318,548
69,703
49,338,998
76,865
266,921,684

4.3
5.6
4.5
4.5
2.6
2.6
2.5
2.5
2.4
2.3
1.6
1.6
2.2
2.1
1.5
1.5
2.1
2.0
1.4
1.3
1.6
1.6
1.5
1.5
1.2
1.0

1,751,126
541,071,476
130,173
158,051,459

2.2
1.3
2.8
2.4

104,155
126,671,963
12,685
12,889,821

1.3
5.1
4.9
8.7

44,081
61,532,857

5.1
5.0

Tenure:
Full owners .........................................................................farms
acres
Part owners .........................................................................farms
acres
Tenants ...............................................................................farms
acres

1,408,961
310,218,983
493,137
503,138,279
140,122
86,860,314

2.4
2.1
1.4
1.8
4.7
2.3

All principal producer characteristics by 1Sex of operator:
Male ................................................................................farms
acres
Female ............................................................................farms
acres

1,787,998
847,232,627
766,474
238,157,861

2.2
1.7
2.9
2.7

Primary occupation:
Farming ..........................................................................farms
Other ...............................................................................farms

1,207,375
1,533,078

1.5
3.6

Legal status for tax purposes (see text):
Family or individual .............................................................farms
acres
Partnership .........................................................................farms
acres
Corporation:
Family held .....................................................................farms
acres
Other than family held .....................................................farms
acres
Other - estate or trust, prison farm, grazing association,
American Indian Reservation, etc .....................................farms
acres

See footnote(s) at end of table.

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Item

Total

Coefficient
of variation
(percent)

All principal producer characteristics by 1- - Con.
Hispanic, Latino, or
Spanish origin (see text) .................................................. farms
acres

77,416
26,041,600

9.7
4.7

Race:
American Indian or
Alaska Native ............................................................... farms
acres
Asian .............................................................................. farms
acres
Black or African American .............................................. farms
acres
Native Hawaiian or
Other Pacific Islander ................................................... farms
acres
White .............................................................................. farms
acres
More than one race reported .......................................... farms
acres

39,632
51,095,994
13,904
1,831,229
32,052
3,862,936

11.8
3.9
8.6
7.7
8.5
7.5

2,092
426,068
1,955,737
843,497,615
19,773
6,712,435

20.7
53.6
2.1
1.7
7.4
4.8

Military service (see text):
Never served ........................................................... producers
Served ..................................................................... producers

2,402,342
338,111

2.6
2.8

All producers by age group 1:
Under 25 years .................................................................. farms
25 to 34 years .................................................................... farms
35 to 44 years .................................................................... farms
45 to 54 years .................................................................... farms
55 to 64 years .................................................................... farms
65 to 74 years .................................................................... farms
75 years and over .............................................................. farms

50,943
234,496
390,345
614,654
955,354
757,936
396,106

16.6
11.7
4.7
4.4
1.3
1.8
1.8

Net cash farm income of operations (see text):
Farms with gains of 2Less than $1,000 ........................................................... farms
$1,000
$1,000 to $4,999 ............................................................ farms
$1,000
$5,000 to $9,999 ............................................................ farms
$1,000
$10,000 to $24,999 ........................................................ farms
$1,000
$25,000 to $49,999 ........................................................ farms
$1,000
$50,000 or more ............................................................. farms
$1,000

66,633
31,436
156,683
431,683
103,942
756,426
153,619
2,525,811
114,269
4,097,569
296,183
104,245,583

1.6
2.2
1.2
1.3
2.0
2.0
1.6
1.6
1.8
1.8
1.2
1.0

Farms with losses of Less than $1,000 ........................................................... farms
$1,000
$1,000 to $4,999 ............................................................ farms
$1,000
$5,000 to $9,999 ............................................................ farms
$1,000
$10,000 to $24,999 ........................................................ farms
$1,000
$25,000 to $49,999 ........................................................ farms
$1,000
$50,000 or more ............................................................. farms
$1,000

89,302
45,846
342,608
988,554
256,919
1,854,855
272,079
4,266,566
104,865
3,629,228
85,118
13,380,008

3.9
3.8
3.9
4.2
3.7
3.9
4.0
4.4
3.7
3.7
2.4
1.9

Livestock and poultry:
Cattle and calves inventory ................................................ farms
number
Beef cows inventory ....................................................... farms
number
Milk cows inventory ........................................................ farms
number
Hog and pigs inventory ...................................................... farms
number
Layers inventory ................................................................ farms
number
Broilers sold ....................................................................... farms
number
Aquaculture sold ................................................................ farms
$1,000

882,692
93,648,041
729,046
31,722,039
54,599
9,539,631
66,439
72,381,007
232,500
368,241,393
32,751
8,889,759,283
5,350
1,778,587

2.3
2.1
2.1
2.6
3.2
1.7
5.2
1.8
4.4
2.9
4.8
2.8
3.8
5.0

Selected crops harvested:
Corn for grain ..................................................................... farms
acres
Durum wheat for grain ....................................................... farms
acres
Other spring wheat for grain (see text) ............................... farms
acres
Winter wheat for grain ........................................................ farms
acres
Sorghum for grain .............................................................. farms
acres
Soybeans for beans ........................................................... farms
acres
Rice .................................................................................... farms
acres
Cotton ................................................................................ farms
acres

304,801
84,738,562
3,093
2,206,169
20,076
10,419,033
86,596
26,186,417
15,339
5,070,159
303,191
90,149,480
4,637
2,395,054
16,149
11,401,965

1.6
1.3
5.2
5.3
2.6
4.9
1.3
0.8
2.3
3.1
1.2
1.9
10.1
26.6
3.8
2.8
--continued

Appendix A A - 21

Table B. Reliability Estimates of U.S. Totals: 2017 (continued)
[For meaning of abbreviations and symbols, see introductory text.]
Item

Total

Coefficient
of variation
(percent)

Selected crops harvested: - Con.

1
2

Item

Total

Coefficient
of variation
(percent)

Selected crops harvested: - Con.
Land in vegetables (see text) - Con.

Peanuts .............................................................................. farms
acres
Barley ................................................................................. farms
acres
Oats ................................................................................... farms
acres

6,379
1,786,767
11,188
2,206,808
19,842
814,140

6.3
7.5
2.6
5.1
2.3
2.9

Forage - land used for all hay and all
haylage, grass silage, and
greenchop (see text) ........................................................ farms
acres
Land in vegetables (see text) ............................................. farms
acres
Potatoes ......................................................................... farms
acres
Tomatoes in the open .................................................... farms
acres

799,627
56,858,622
74,276
3,965,622
16,554
1,133,128
28,673
335,348

2.0
1.3
5.8
2.6
6.6
3.2
6.3
6.4

Sweet corn ..................................................................... farms
acres
Lettuce ............................................................................ farms
acres
Land in orchards (see text) ................................................. farms
acres
Apples ............................................................................ farms
acres
Grapes ............................................................................ farms
acres
Oranges .......................................................................... farms
acres
Almonds ......................................................................... farms
acres
Land in berries .................................................................... farms
acres

20,784
496,096
10,869
342,965
111,955
5,665,600
26,408
381,718
28,387
1,136,155
7,973
602,830
7,954
1,266,160
33,919
302,199

5.7
3.3
8.7
2.1
3.5
3.7
4.9
4.9
3.2
6.2
3.4
4.9
4.4
2.8
4.3
2.2

Data were collected for a maximum of four producers per farm.
Farms with total production expenses equal to market value of agricultural products sold, government payments, and farm-related income are included as farms with gains of less than $1,000.

A - 22 Appendix A

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Table C. Summary of Coverage, Nonresponse, and Misclassification Adjustments by State: 2017
[For meaning of abbreviations and symbols, see introductory text.]
Geographic area

Total
(number)

Standard
error

Adjustment
as percent
of total

Percent of total
adjustment
from coverage

Percent of total
adjustment from
nonresponse

Percent of total
adjustment from
misclassification

ALL FARMS (NUMBER)
United States Total
United States ............................................................................................

2,042,220

43,278

37.6

15.1

13.9

8.6

Alabama ...................................................................................................
Alaska .......................................................................................................
Arizona .....................................................................................................
Arkansas...................................................................................................
California ..................................................................................................
Colorado ...................................................................................................
Connecticut ...............................................................................................
Delaware ..................................................................................................
Florida.......................................................................................................
Georgia .....................................................................................................

40,592
990
19,086
42,625
70,521
38,893
5,521
2,302
47,590
42,439

1,545
13
2,637
1,661
1,896
3,173
464
204
1,426
1,215

39.5
3.2
56.3
37.3
42.0
39.6
49.6
42.5
47.0
36.1

15.3
(NA)
20.5
14.0
18.8
15.4
22.0
16.3
20.5
14.7

15.1
3.2
22.7
15.1
15.5
14.1
17.2
17.2
16.2
13.2

9.1
(NA)
13.1
8.3
7.7
10.1
10.4
9.1
10.2
8.2

Hawaii .......................................................................................................
Idaho.........................................................................................................
Illinois........................................................................................................
Indiana ......................................................................................................
Iowa ..........................................................................................................
Kansas......................................................................................................
Kentucky ...................................................................................................
Louisiana ..................................................................................................
Maine ........................................................................................................
Maryland ...................................................................................................

7,328
24,996
72,651
56,649
86,104
58,569
75,966
27,386
7,600
12,429

560
1,288
1,894
1,822
1,650
2,763
3,436
1,360
1,065
1,107

47.4
40.0
26.0
33.5
23.3
33.8
38.6
44.7
45.5
32.0

21.1
17.9
9.6
13.2
8.3
10.5
15.7
15.8
20.5
13.3

17.2
12.5
10.4
12.8
9.8
15.6
13.4
18.9
15.4
11.1

9.1
9.6
5.9
7.5
5.2
7.7
9.6
9.9
9.6
7.7

Massachusetts ..........................................................................................
Michigan ...................................................................................................
Minnesota .................................................................................................
Mississippi ................................................................................................
Missouri ....................................................................................................
Montana....................................................................................................
Nebraska ..................................................................................................
Nevada .....................................................................................................
New Hampshire ........................................................................................
New Jersey ...............................................................................................

7,241
47,641
68,822
34,988
95,320
27,048
46,332
3,423
4,123
9,883

723
2,276
1,138
2,117
3,297
2,046
1,383
264
352
1,076

46.0
38.2
30.3
39.4
33.7
37.5
37.8
50.8
50.1
35.4

21.4
16.4
10.9
13.7
13.2
13.7
10.4
22.6
22.8
15.7

14.5
13.3
12.4
16.6
12.7
14.7
20.0
17.1
16.6
11.2

10.2
8.5
7.0
9.1
7.9
9.1
7.4
11.1
10.7
8.5

New Mexico ..............................................................................................
New York ..................................................................................................
North Carolina ..........................................................................................
North Dakota ............................................................................................
Ohio ..........................................................................................................
Oklahoma .................................................................................................
Oregon......................................................................................................
Pennsylvania ............................................................................................
Rhode Island ............................................................................................
South Carolina ..........................................................................................

25,044
33,438
46,418
26,364
77,805
78,531
37,616
53,157
1,043
24,791

2,354
1,263
1,604
787
2,385
3,431
2,687
2,075
186
1,346

49.8
37.0
41.2
37.9
32.2
38.8
40.2
38.3
47.0
44.8

21.2
15.2
16.5
8.7
13.8
14.0
18.2
16.5
25.4
16.6

17.0
13.6
14.7
22.0
10.8
15.8
12.7
13.5
11.6
17.6

11.6
8.2
10.0
7.2
7.5
9.0
9.3
8.3
10.0
10.6

South Dakota ............................................................................................
Tennessee ................................................................................................
Texas ........................................................................................................
Utah ..........................................................................................................
Vermont ....................................................................................................
Virginia ......................................................................................................
Washington ...............................................................................................
West Virginia ............................................................................................
Wisconsin .................................................................................................
Wyoming...................................................................................................

29,968
69,983
248,416
18,409
6,808
43,225
35,793
23,622
64,793
11,938

717
2,066
8,706
1,780
569
1,037
2,013
1,297
1,665
917

36.5
35.5
42.1
42.1
44.3
39.0
41.8
34.7
35.2
40.3

8.5
15.2
17.8
18.1
17.6
17.2
20.4
14.7
14.2
15.1

20.7
11.3
14.5
13.8
17.0
12.2
12.0
10.9
13.2
15.1

7.3
9.0
9.8
10.3
9.7
9.6
9.4
9.1
7.8
10.2

900,217,576

15,031,334

22.7

4.4

11.8

6.5

Alabama ...................................................................................................
Alaska .......................................................................................................
Arizona .....................................................................................................
Arkansas...................................................................................................
California ..................................................................................................
Colorado ...................................................................................................
Connecticut ...............................................................................................
Delaware ..................................................................................................
Florida.......................................................................................................
Georgia .....................................................................................................

8,580,940
849,753
26,125,819
13,888,929
24,522,801
31,820,957
381,539
525,324
9,731,731
9,953,730

199,954
1,199
1,432,761
310,057
3,248,717
1,722,400
27,934
36,130
344,544
496,845

27.7
0.2
11.1
23.5
17.7
21.8
30.0
23.7
19.2
24.4

8.6
(NA)
2.9
5.9
3.7
4.0
11.1
5.6
4.7
7.0

12.1
0.2
4.0
12.3
8.4
11.3
11.3
12.3
7.9
10.9

7.0
(NA)
4.2
5.3
5.7
6.4
7.5
5.8
6.7
6.5

Hawaii .......................................................................................................
Idaho.........................................................................................................
Illinois........................................................................................................
Indiana ......................................................................................................
Iowa ..........................................................................................................
Kansas......................................................................................................
Kentucky ...................................................................................................
Louisiana ..................................................................................................
Maine ........................................................................................................
Maryland ...................................................................................................

1,135,352
11,691,912
27,006,288
14,969,996
30,563,878
45,759,319
12,961,784
7,997,511
1,307,613
1,990,122

42,889
703,194
773,609
297,098
515,336
974,716
377,113
580,578
72,510
72,254

6.3
18.8
20.7
20.6
24.7
26.2
27.3
30.7
27.8
15.7

2.1
3.8
3.1
3.1
3.0
3.6
8.3
5.9
10.3
4.9

1.0
9.8
14.4
14.5
18.4
18.0
12.5
18.4
11.1
7.1

3.2
5.2
3.2
3.0
3.3
4.6
6.5
6.4
6.4
3.7

Massachusetts ..........................................................................................
Michigan ...................................................................................................
Minnesota .................................................................................................
Mississippi ................................................................................................
Missouri ....................................................................................................
Montana....................................................................................................
Nebraska ..................................................................................................

491,653
9,764,090
25,516,982
10,415,136
27,781,883
58,122,878
44,986,821

30,187
371,324
402,371
251,770
657,610
3,130,653
1,155,962

30.5
21.7
25.1
26.6
26.6
19.8
26.9

12.4
5.8
4.1
7.0
6.2
2.3
3.1

10.9
11.3
17.1
13.9
14.5
12.0
18.4

7.2
4.7
3.9
5.6
5.8
5.5
5.4

States

LAND IN FARMS (ACRES)
United States Total
United States ............................................................................................
States

--continued

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Appendix A A - 23

Table C. Summary of Coverage, Nonresponse, and Misclassification Adjustments by State: 2017 (continued)
[For meaning of abbreviations and symbols, see introductory text.]
Geographic area

Total
(number)

Standard
error

Adjustment
as percent
of total

Percent of total
adjustment
from coverage

Percent of total
adjustment from
nonresponse

Percent of total
adjustment from
misclassification

LAND IN FARMS (ACRES) - Con.
States - Con.
Nevada ......................................................................................................
New Hampshire.........................................................................................
New Jersey ...............................................................................................
New Mexico ..............................................................................................
New York ..................................................................................................
North Carolina ...........................................................................................
North Dakota .............................................................................................
Ohio ..........................................................................................................
Oklahoma ..................................................................................................
Oregon ......................................................................................................

6,128,153
425,393
734,084
40,659,836
6,866,171
8,430,522
39,341,591
13,965,295
34,156,290
15,962,322

512,649
25,211
53,129
4,089,251
167,815
262,971
2,073,608
278,984
865,429
522,836

13.2
31.0
18.3
17.7
23.6
24.0
32.9
20.6
25.2
13.2

2.4
14.1
6.3
3.2
7.0
6.6
2.9
5.7
5.8
2.7

2.7
8.5
7.5
5.5
11.8
11.2
24.7
10.6
13.3
6.1

8.0
8.4
4.5
9.0
4.7
6.2
5.3
4.4
6.1
4.5

Pennsylvania .............................................................................................
Rhode Island .............................................................................................
South Carolina ..........................................................................................
South Dakota ............................................................................................
Tennessee ................................................................................................
Texas ........................................................................................................
Utah ..........................................................................................................
Vermont ....................................................................................................
Virginia ......................................................................................................
Washington ...............................................................................................

7,278,668
56,864
4,744,913
43,243,742
10,874,238
127,036,184
10,811,604
1,193,437
7,797,979
14,679,857

228,467
3,981
347,006
1,100,330
219,305
3,219,092
247,332
123,507
145,486
291,923

28.4
25.7
28.6
29.5
26.0
23.0
9.4
28.3
26.5
13.4

8.3
12.0
8.5
2.4
8.3
5.8
2.1
7.1
8.7
3.3

14.4
6.6
13.3
22.0
11.2
8.3
4.3
14.9
10.8
6.3

5.8
7.1
6.8
5.0
6.4
8.9
3.0
6.3
7.1
3.8

West Virginia .............................................................................................
Wisconsin ..................................................................................................
Wyoming ...................................................................................................

3,662,178
14,318,630
29,004,884

156,944
435,791
3,226,928

24.6
24.0
14.0

10.3
5.8
1.4

7.5
13.8
5.7

6.9
4.4
6.9

388,522,695

2,923,858

18.1

3.4

11.1

3.6

Alabama ....................................................................................................
Alaska .......................................................................................................
Arizona ......................................................................................................
Arkansas ...................................................................................................
California ...................................................................................................
Colorado ...................................................................................................
Connecticut ...............................................................................................
Delaware ...................................................................................................
Florida .......................................................................................................
Georgia .....................................................................................................

5,980,595
70,459
3,852,008
9,651,160
45,154,359
7,491,702
580,114
1,465,973
7,357,343
9,573,252

173,995
160
84,786
375,713
1,116,572
288,834
25,685
87,913
275,257
413,404

21.1
0.3
6.4
22.7
15.8
10.3
11.1
37.2
13.2
21.8

5.5
(NA)
1.8
4.1
3.3
2.5
3.3
9.1
3.0
4.8

11.7
0.3
3.1
14.5
9.0
4.5
5.7
19.9
6.7
12.9

4.0
(NA)
1.5
4.1
3.5
3.3
2.1
8.1
3.6
4.1

Hawaii .......................................................................................................
Idaho .........................................................................................................
Illinois ........................................................................................................
Indiana ......................................................................................................
Iowa ..........................................................................................................
Kansas ......................................................................................................
Kentucky ...................................................................................................
Louisiana ...................................................................................................
Maine ........................................................................................................
Maryland ...................................................................................................

563,803
7,567,439
17,009,971
11,107,336
28,956,455
18,782,726
5,737,920
3,172,978
666,962
2,472,805

96,131
555,829
425,802
377,238
363,219
278,393
124,641
93,001
49,050
173,825

10.3
11.3
18.5
17.2
25.1
14.4
13.8
25.6
12.8
19.4

3.4
1.8
2.5
2.4
3.4
2.8
3.2
3.8
3.5
5.8

5.1
7.2
13.1
12.3
17.8
7.9
7.4
18.1
7.4
9.6

1.8
2.3
2.9
2.5
3.8
3.7
3.2
3.7
2.0
4.0

Massachusetts ..........................................................................................
Michigan ....................................................................................................
Minnesota .................................................................................................
Mississippi .................................................................................................
Missouri .....................................................................................................
Montana ....................................................................................................
Nebraska ...................................................................................................
Nevada ......................................................................................................
New Hampshire.........................................................................................
New Jersey ...............................................................................................

475,184
8,220,935
18,395,390
6,195,968
10,525,938
3,520,623
21,983,430
665,758
187,794
1,097,950

44,440
264,720
327,929
235,789
333,340
179,935
591,020
112,662
17,904
28,911

17.6
13.9
25.3
19.9
20.4
24.0
20.5
21.3
10.7
8.5

4.1
2.9
3.5
4.2
3.4
2.3
2.6
2.8
3.3
2.7

10.5
8.2
17.9
12.2
12.9
16.7
13.6
13.4
5.4
3.8

3.0
2.8
3.8
3.5
4.1
4.9
4.3
5.1
2.0
1.9

New Mexico ..............................................................................................
New York ..................................................................................................
North Carolina ...........................................................................................
North Dakota .............................................................................................
Ohio ..........................................................................................................
Oklahoma ..................................................................................................
Oregon ......................................................................................................
Pennsylvania .............................................................................................
Rhode Island .............................................................................................
South Carolina ..........................................................................................

2,582,343
5,369,212
12,900,674
8,234,102
9,341,225
7,465,512
5,006,822
7,758,884
57,998
3,008,739

68,848
95,711
239,004
457,678
147,955
163,532
145,919
227,164
3,127
139,270

8.4
13.1
20.4
30.4
16.9
13.9
10.2
17.0
11.7
12.0

2.2
2.9
5.1
2.0
3.9
3.2
2.3
3.6
4.7
2.6

3.8
7.9
10.1
25.0
9.7
6.8
5.9
10.9
4.4
7.2

2.4
2.3
5.2
3.5
3.4
3.8
2.0
2.6
2.6
2.2

South Dakota ............................................................................................
Tennessee ................................................................................................
Texas ........................................................................................................
Utah ..........................................................................................................
Vermont ....................................................................................................
Virginia ......................................................................................................
Washington ...............................................................................................
West Virginia .............................................................................................
Wisconsin ..................................................................................................
Wyoming ...................................................................................................

9,721,522
3,798,934
24,924,041
1,838,610
780,968
3,960,501
9,634,461
754,279
11,427,423
1,472,113

323,047
146,980
417,770
141,700
72,090
86,812
355,103
19,100
318,424
56,575

29.0
18.5
14.1
12.5
15.7
13.1
10.6
11.5
17.5
17.3

2.2
3.3
3.9
2.7
2.3
3.6
2.8
4.7
2.7
1.9

23.1
11.8
6.1
7.3
11.5
6.4
5.1
3.4
12.5
9.7

3.7
3.4
4.1
2.5
2.0
3.0
2.6
3.3
2.3
5.7

SALES ($1,000)
United States Total
United States.............................................................................................
States

A - 24 Appendix A

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Table D. American Indian or Alaska Native Producers: 2017
[For meaning of abbreviations and symbols, see introductory text.]
American Indian or Alaska Native farm producers
Geographic area

Individually
reported 1

Total

American Indian or Alaska Native farm producers
Geographic area

Other 2

United States Total
United States ......................................

2

Individually
reported 1

Other 2

States - Con.
79,597

79,198

399

Alabama .............................................
Alaska .................................................
Arizona ...............................................
Arkansas.............................................
California ............................................
Colorado .............................................
Connecticut .........................................
Delaware ............................................
Florida.................................................
Georgia ...............................................

1,326
88
19,656
1,326
2,538
1,185
55
8
1,027
524

1,326
88
19,481
1,326
2,537
963
55
8
1,027
524

175
1
222
-

Hawaii .................................................
Idaho...................................................
Illinois..................................................
Indiana ................................................
Iowa ....................................................
Kansas................................................
Kentucky .............................................
Louisiana ............................................
Maine ..................................................
Maryland .............................................

265
461
332
325
229
961
650
523
113
142

265
461
332
325
229
961
650
523
112
142

1
-

Massachusetts ....................................
Michigan .............................................
Minnesota ...........................................

66
777
408

66
777
408

-

States

1

Total

Mississippi...........................................
Missouri...............................................
Montana ..............................................
Nebraska.............................................
Nevada................................................
New Hampshire ..................................
New Jersey .........................................
New Mexico ........................................
New York ............................................
North Carolina .....................................

321
1,544
2,130
210
315
39
91
8,812
278
1,023

321
1,544
2,130
210
315
39
91
8,812
278
1,023

-

North Dakota .......................................
Ohio ....................................................
Oklahoma............................................
Oregon ................................................
Pennsylvania .......................................
Rhode Island .......................................
South Carolina ....................................
South Dakota ......................................
Tennessee ..........................................
Texas ..................................................

470
530
17,102
1,255
302
5
307
1,242
843
5,663

470
530
17,102
1,255
302
5
307
1,242
843
5,663

-

Utah ....................................................
Vermont ..............................................
Virginia ................................................
Washington .........................................
West Virginia .......................................
Wisconsin............................................
Wyoming .............................................

1,467
90
440
1,202
249
293
389

1,467
90
440
1,202
249
293
389

-

Data were collected for a maximum of four producers per farm.
Data represent American Indian or Alaska Native farm or ranch producers on reservations who did not report individually. Data obtained by reservation officials.

2017 Census of Agriculture
USDA, National Agricultural Statistics Service

Appendix A A - 25