| ||Format||Pages||Price|| |
|PDF (84K)||6||$25||  ADD TO CART|
|Complete Source PDF (1.5M)||97||$55||  ADD TO CART|
Two aspects of a sample design that seem crucial in deciding whether specialized statistical expertise will be required to make the design work are: (1) frame versus nonframe sampling and (2) random versus patterned sampling. These terms are illustrated and some subcategories are suggested. Even a clearly defined frame may be less than adequate. Patterned sampling can yield small sampling variances but the size of this variance may be difficult to discover. The assistance of a statistician seems most needed when there is no frame, or a less than adequate one, and when a highly patterned sampling design is used.
random sampling, patterned sampling, sampling distributions, nonframe sampling, frame adequacy, statistical consulting
Professor, North Carolina State University, Raleigh, NC