Great Plains Population and Environment Data: Agricultural Data, 1870-1997 [United States] (ICPSR 4254)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Sample View help for Sample

The first comprehensive United States Census of Agriculture in 1850 imposed a limit below which no tract of land was to be considered a farm. If the value of produce from small farm lots fell below $100 in value, they were to be excluded from the count of census farms. In 1870, the minimums rose to reflect economic change and to help sort out some of the confusion that emerged in the South with emancipation, including the subdivision of plantations and attendant uncertainties in proprietorship. The preamble to the 1870 census declared that "[n]o farm will be reported of less than three acres, unless five hundred dollars' worth of produce has actually been sold off from it during the year." This same definition was applied in 1880 and 1890.

In the 1900 preamble, census officials explained a return to the original standard of 1850 and offered some cautionary advice. In reviewing the changing definition of census farms they noted that in "no [agricultural] census of the country had one-half of the farms reported products of a value of $500, and the proportion that had sold products of that value was much smaller. The land occupied and the products secured by very many persons devoting their entire time to caring for small dairies, apiaries, florists' establishments, and kindred agricultural establishments were omitted from reports, although those persons were properly included in the occupation tables as dairymen, apiarists, florists, etc." Because of these omissions from tabulated results in 1870, 1880, and 1890, census marshals issued revised instructions to enumerators in 1900 that were far more inclusive. This occupational and self-declared basis for defining farms is further clarified by the explicit inclusion of market-gardens, orchards, nurseries, cranberry marshes, greenhouses and city dairies, provided, [emphasis in the original] the entire time of at least one individual is devoted to their care.

By 1910, size and production minimums returned in a definition that retained language about farms as the agricultural operations of self-described farmers, with the added provision that they again be tracts of farmland at least three acres in size, no matter what the value of their produce, or if they fell below the size minimum, that they employ the continuous services of at least one person or produce at least $250 worth of farm products. This definition was applied again in 1920, 1925, 1930, 1935, 1940, and 1945. However the value of production for farms under three acres was not adjusted for inflation or deflation. The inclusion or exclusion of very small farms in the census during these years relied, in part, on changes in the price of agricultural commodities.

In 1950, farms of three or more acres were included in the census only if the value of their produce from the previous crop year, in 1949, amounted to $150, and similarly operations under three acres in size were included only if they met the minimum of $150 in sales from the previous year.

In 1959, the farm size and product value minimums were adjusted again. The farm size minimum was raised to ten acres, with sales of at least $50, and if farms fell below ten acres, they were required to have estimated sales of at least $250. Farms that fell below either of these product value minimums were included in the census if they could be expected to meet them under normal conditions. The same census definition of census farms was applied in 1964.

The biggest change in process came in 1969 with the switch from face-to-face enumeration to enumeration by mail. Nonsample "short" forms were mailed to every respondent on the Census Bureau's mailing list, a list furnished by the United States Department of Agriculture. By law, everyone on the list was required to respond to the nonsample form. Respondents who reported sales or acreage above specified levels on nonsample forms were then sent correspondence requesting additional sample data. To limit respondent burden, the nonsample form included only questions to be reported at the county level. All county-level data are based on the nonsample form.

The biggest change in definition occurred in 1974 when the acreage criterion was dropped and a farm was defined as a place from which $1,000 in agricultural products was produced and sold, or normally would have been sold. The use of sales as the basic criterion meant that operations that focused on household use or non-commercial distribution were no longer regarded as farms. It is estimated that this change defined 300,000 farms out of existence nationally, an estimate made more precise by the count of farms using both definitions in the 1974 Agricultural Census and the 1980 Population Census (Bruce L. Gardner, American Agriculture in the Twentieth Century: How It Flourished and What It Cost. Cambridge: Harvard University Press. 2002: 51-52). The $1,000 sales criterion persisted until the end of the twentieth century.

Adjustments were made to enumeration by mail in 1978 to improve coverage, including a revised mail list and the adoption of a direct area enumeration sample, including the use of aerial photography, as a data checking method.

In 1982, 3,653,000 nonsample forms were mailed in late December to individuals, businesses and organizations on the mail list. Data collection included a reminder card and five follow-up letters. Further follow-up was conducted by mail and telephone for nonrespondents in 14 states, and adjustments made to the final results. All new farm successors reported by former operators were researched to see if they had been included in the census mailing. Report forms mailed to successor addresses not previously on the mail file improved the coverage of the results.

Since 1987, coverage evaluation reports have accompanied the publication of the agricultural census to provide estimates of the completeness of the enumeration. In 1997, the use of post-census sampling by the USDA increased the count of United States farms from 1.97 to 2.06 million.

Universe View help for Universe

All the counties in the 12 Great Plains states of the United States (Colorado, Iowa, Kansas, Minnesota, Montana, Nebraska, New Mexico, North Dakota, Oklahoma, South Dakota, Texas, and Wyoming).

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Original Release Date View help for Original Release Date

2005-06-22

Version History View help for Version History