United States Department of Agriculture

2017 Congressional District Assignment Methodology

The census of agriculture, conducted every five years, is the leading source of facts and statistics about the Nation's farms and ranches and the only source of uniform, comprehensive agriculture data for every State and county, or county equivalent, in the U.S. Following each census, reporting farms and ranches are assigned to congressional districts (CD) and two products are prepared, CD profiles and CD rankings. In publishing agricultural summary data by CD for the 2017 Census of Agriculture, the CDs pertain to the 116th Congress.

Purpose and Tabular Presentation

Congressional District profiles provide data on selected farm, economic, and operator characteristics for the farms and ranches assigned to the CD. Seven States have only one CD - Alaska, Delaware, Montana, North Dakota, South Dakota, Vermont, and Wyoming. For these "at large" States, the State total is presented for the district. Profiles are not available for those districts with very little agriculture reported, if any, to protect the confidentiality of individual reports.
Because redistricting occurred between the 2012 and 2017 censuses, the physical composition of many CDs changed significantly. In many cases, CDs are not comparable – i.e., 116th Congress vs. 113th Congress – and no comparisons are included in these CD profiles for current census to previous census.

The statistics presented in this report include:

• Producer characteristics
• Farm characteristics
• Selected value of agricultural products sold
• Selected livestock and poultry inventories; and
• Selected crops area harvested

How Farms were Assigned to Congressional Districts

The 116th Congress CD tabulations of 2017 Census of Agriculture data are based on operation location. Census respondents were asked to declare their principal county of operation as the county with the greatest share of their total value of production. Additionally, the respondent’s mail ZIP Code was known; however, this ZIP Code did not necessarily correspond to the actual location of the farm or ranch. CD assignments were based on map boundary files produced by the National Agricultural Statistics Service (NASS), as well as other files produced outside of NASS. Digitized county and CD boundary files were provided by the U.S. Census Bureau, which is responsible for congressional apportionment among the States, as well as for mapping the resulting CDs in accordance with State requests. Digitized ZIP Code boundary files for 2017 were obtained from Environmental Systems Research Institute, Inc.’s (ESRI) data and maps. In addition, digitized agricultural/non-agricultural land boundaries were produced by NASS.

Of the reports returned for the 2017 census that were determined to be from in-scope farms or ranches, approximately 57.6 percent were respondents to the 2012 Census of Agriculture in States where CD boundaries did not change, and thus were assigned to the same CD. Another 26.7 percent were deterministically assigned to a CD because geocoding software was available that enabled NASS to geocode those operations’ locations to the appropriate CD. (I.e., for records where the ZIP Code was within the reported principal county, NASS was able to calculate geographic coordinates for those operations based on address, and spatially determine in which CD each operation’s coordinates were located.) Another 13.6 percent of records were deterministically assigned to CDs because the principal county reported fell entirely within a single CD. Another 0.8 percent of all records were assigned deterministically to CDs because their mail ZIP Code was located in the operation’s principal county, and the Zip Code fell entirely within a single CD. The remaining 1.3 percent of census farm and ranch records were assigned probabilistically following a statistical model. These records involved farm operations where either the county or the ZIP Code contained multiple CDs. For these cases, two models for multiple CD probabilistic assignments were developed. The first model was used if the mail ZIP Code was located in the operation’s principal county. Then probabilities of assignment to each CD within the ZIP Code were equal to the proportion of that ZIP Code’s agricultural land contained in the CD, as calculated by NASS. The second model was used if the mail ZIP Code was not located in the operation’s principal county. In that case NASS used the principal county in carrying out the CD assignment. The probabilities of assignment to each CD within the principal county were equal to the proportion of the county’s agricultural land contained in the CD, as calculated by NASS.

Probabilistically assigned farms and ranches were reviewed by NASS Regional Field Offices to verify their assignments and, in some cases, corrections were made. The farms and ranches that were reviewed were typically large operations (with respect to their total value of production) or operations with a rare commodity. All published farm counts and totals were fully adjusted for nonresponse, misclassification, and coverage.

back to Congressional District Profiles