Stay Connected

Follow ERS on Twitter
Subscribe to RSS feeds
Subscribe to ERS e-Newsletters.aspx
Listen to ERS podcasts
Read ERS blogs at USDA


Data on farm household income and characteristics come from USDA’s annual Agricultural Resource Management Survey (ARMS). Changes in survey methodology, sampling, or response rates can influence estimates derived from survey data. In 2014, the most recent year, the number of family farm records used to calculate farm household statistics grew to 29,081 farms from 16,607 in 2013 and 20,561 in 2012. Over the past 6 years, the highest response rate (67.4 percent) was in 2012, the year that the Census of Agriculture was conducted.  

Changes Between 2012 and Earlier Years

Beginning in 2012, two major changes affected how ARMS data are collected. First, ARMS switched to an all-mail survey: each survey is mailed to the respondent, though most surveys are eventually completed through follow-up in person or over the phone. The number of records enumerated through the mail has ranged from 14.2 percent in 2011 to 42.6 percent in 2012, averaging 28 percent over the past 6 years. Second, the all-mail "Core" version of ARMS Phase III, which had been introduced in 2003, was discontinued. Together, these survey design changes make farm household statistics from 2012 onward difficult to compare with earlier surveys. These changes can also affect the precision of estimates. The coefficient of variation (which measures the size of the standard error relative to the mean) for total household income and its off-farm income component was larger in 2012 and 2013 relative to 2010 and 2011. For the farm income component of household income, the coefficient was lower. See more statistics on the ARMS from recent years

See more documentation for the Agricultural Resource Management Survey.

Last updated: Tuesday, February 09, 2016

For more information contact: Daniel Prager