Online Program

Evaluation of Using CVs as a Publication Standard
Wendy Barboza, USDA/NASS 
*Jason L Bell, USDA/NASS 

Keywords: NASS, publication standards, CV, coefficient of variation, graphics, county estimates

Each year, the National Agricultural Statistics Service (NASS) conducts a County Agricultural Production Survey (CAPS). The purpose of this survey is to estimate the acreage and production of specific commodities at the county level within each state. To avoid disclosure issues, NASS has two criteria for their publication standards. Either the county must have 30 usable, positive reports, or it must have a certain number of usable, positive reports and the reported data must account for at least 25\% of the county-level estimate. In the past, Coefficients of Variation (CVs) were not available because CAPS was not a probability-based sample. This has since changed, thus the CV can now be used as a publication standard.

In order to evaluate this, plots were generated to display the relationship between usable, positive reports and the CVs of the final survey estimate for each county. In general, the plots revealed that there is a negative relationship between the two variables. More importantly, the CV of some publishable counties was unexpectedly high, suggesting that a minimum CV might be an appropriate criterion for determining whether or not a county is published.