Files
Abstract
Following the 2002 Census of Agriculture, a Classification Error Study was conducted to
estimate the number of operations misclassified (either as farms or non-farms) in the census.
This was done by matching operations who reported in the Area Frame June 2002 Agricultural
Survey to their census report and comparing their answers. The information on the June
Agricultural Survey was assumed to be correct, since it was collected in person by trained
enumerators, while the census data were obtained in most cases through a self-administered mail
form. Misclassification estimates for 2002 were generated based on cases where the census
report was classified differently than the matching June Agricultural Survey report. The
estimated misclassification rate was small but it was clear that, in some cases, the assumption of
the June Agricultural Survey response being correct was not justified.
Since the 2002 misclassification estimates were not used to adjust published census estimates, a
different approach was taken for the classification error study in 2007. For 2007, the focus was
on understanding why operations reported differently in the June Agricultural Survey than they
did on the census, rather than on estimating misclassification rates. Census records were
matched to operations’ reports from the 2007 June Agricultural Survey, but neither report was
assumed to be “the truth.” As in 2002, this study targeted operations classified as farms in one
case and non-farms in the other. In addition, it focused on operations who reported total acres
operated that differed by more than 25 percent between June and the census. Instead of
assuming one source was correct, these operations were reinterviewed, shown their June
Agricultural Survey and census questionnaires, and asked to resolve and explain the
discrepancies. In addition, operators were asked general questions related to suspected problems
in reporting their acreage.
The reinterviews uncovered several different sources of errors in reporting. These occurred in
both the June Agricultural Survey and the census, with the majority of errors in the June
Agricultural Survey. Errors were related to respondents, enumerators and National Agricultural
Statistics Service procedures and show that a multi-part solution is needed to address them.