![[logo] US EPA](../gif/logo_epaseal.gif){kind=logo}
Establishing Subpopulations and Domains
Often there is interest in obtaining information on subsets of the target population
- Definition requirements for each subset the same as for Target Population
Aquatic Resources - Examples
- Individual ecoregions, biogeographical regions within study area
- All lentic resources in region with area < 100 ha
- All lotic resources with Strahler order < 4
- All lotic resources with < 20% riparian canopy cover
- All 5-th field HUCs with > 10 NWI polygons
- All 6-th field HUCs with > 25% Federal land ownership
Impact on Study Design
- Objectives identify critical subpopulation with expected sample sizes: Information required for each subpopulation/domain (increased sample size)
- Survey design addresses domain sample size requirements:
- Explicitly using stratification or unequal weighting
- Implicitly when other requirements provide sufficient sample sizes
- Other subpopulations can not be defined prior to sample selection
Stratification and Unequal Probability Selection
Stratification: reasons
- Improve precision by constructing strata that are more homogeneous
- Operational/administrative efficiency
- Regions or States may need to be operationally independent
- Different subpopulations require different survey designs
- Design for extensive wetlands (Everglades) may be different from prairie pothole wetlands
Unequal Weighting: Reasons
- Allocate sample effort to subpopulations
- Improve precision of results (desired sample sizes each subpopulation)
- Based on auxiliary information