X. EXPORT SURVEY AND STATISTICAL ANALYSIS METHODS
A. Overview
The key points for the survey design and sampling and statistical analysis for this 2015/2016 Export Cargo Survey are as follows:
- Samples were proportionately stratified according to Export Outlet (EO) – Texas and NOLA.
- To achieve a maximum ± 10% relative margin of error (Relative ME) for the U.S. Export Aggregate level and to ensure proportional sampling from each EO, the targeted number of total samples was 167 samples, with 132 to be collected from Texas and 35 to be collected from NOLA.
- Samples were provided by the U.S. Department of Agriculture’s (USDA) Federal Grain Inspection Service (FGIS) field offces at ports in the respective EOs.
- Export inspections of shipments generated 136 samples from the Texas EO and 46 samples from the NOLA EO. Since the number of samples collected in each EO was in excess of the targeted number of samples, the U.S. Export Aggregate averages for the quality factors were weighted according to the targeted proportion by EO.
- To evaluate the statistical validity of the number of samples surveyed, the Relative ME was calculated for each of the quality attributes at the U.S. Export Aggregate level. The Relative MEs for the quality factor results were all less than ± 10%.