Development of the Survey Design
Survey sampling is intended to characterize the entire population of interest; therefore, all members of the target population must have a known chance of being included in the sample. The survey design specifies the process for selecting sites at which a response will be determined. Includes a probability model for inference based on the randomized selection process.
Simple Random Sampling Designs:
Simple random selection ensures that the sample is representative because all members of the population have an equal chance of being selected. Random selection can be thought of as a kind of lottery drawing to determine which stream reaches, for example, are included in the sample. The selection is non-preferential towards any particular reach or group of reaches. In general, simple random sampling is not recommended for aquatic resource monitoring program designs. Monitoring program objectives usually include requirements that simple random sampling can't address efficiently.
Several random sampling strategies have been developed to accommodate specific objectives and attain enhanced attributes for the selected sample. The Monitoring Design and Analysis Team has developed the following:
Generalized Random Tessellation Stratified Designs - GRTS
- Spatially balances sample across the resource (improved precision)
-
Enables design-based estimators including variances
- Precise control over inclusion probabilities
- Element & region variable probability assignment
- Joint inclusion probability can be determined
- Controls sample and subsample spatial balance
- Nested subsamples easily selected
- Unified theory for point, network, and areal resources such as lakes, streams, and coastal waters
GRTS Design Options
- Multiple density categories to allocate samples, supports unequal selection probability
- Nested subsamples for measuring additional indicators or duplicate samples
- Panels for monitoring over time
- Over-sample selection to address non-target and inaccessible sites
- Special study areas within study-wide design
- Explicit stratification
- Incorporate multiple stage sampling