Download (PDF 320 KB): https://research.fs.usda.gov/download/treesearch/15832.pdf
Abstract
Forest survey analysts usually are interested in tables of values rather than single point estimates. A common error is to include only plots on which nonzero values of the attribute were observed when computing the variance of a mean. Similarly, analysts often exclude nonforest plots from the analysis. The development of the correct estimates of forest area, attribute totals, and their means over the area of interest is described. Program TabGen was written to perform these calculations correctly assuming simple random sampling, stratified random sampling, or double sampling for stratification.
Parent Publication
Citation
Scott, Charles T. 2000. Estimating two-way tables based on forest surveys. In: Hansen, Mark; Burk, Tom, eds. Integrated tools for natural resources inventories in the 21st century. Gen. Tech. Rep. NC-212. St. Paul, MN: U.S. Dept. of Agriculture, Forest Service, North Central Forest Experiment Station. 234-238.
