Evaluation of the National Agricultural Workers Survey (NAWS)

Federal government agencies use NAWS findings for many reasons, including estimating the number and characteristics of farm workers and their dependents, conducting occupational injury and health surveillance, and designing and evaluating child health and education programs. The current NAWS contractor selects the survey sample and conducts interviews in three cycles each year to reflect the seasonality of agricultural production and employment.  Between 1,500 and 4,000 workers are interviewed each year, depending on the information needs and resources of federal agency stakeholders. The data collection contractor also computes the analysis weights.

In 2009, ETA contracted with Mathematica to review the work done by the NAWS data collection contractor on the computation of the sampling weights and nonresponse adjustments in the survey. Mathematica provided a report of findings and recommendations in 2009.  Subsequently, ETA again contracted again with Mathematica to review the changes the data collection contractor made in response to those findings and recommendations. These subsequent projects included a review of the sample design, of the computation of the analysis weights, and of the field procedures used to select the sample of farmworkers. We continue working with ETA and the current NAWS contractor to improve the sampling and weighting procedures used in the NAWS.