2.1.6.2 Choosing a Spatial Survey Design

One of the first steps in developing a survey spatial design is to decide whether or nor you will be implementing a single or multi-stage or multi-phase design. 

• Multi-stage sampling is a kind of complex sample design in which two or more levels of units are imbedded one in the other.  For example: geographic areas (primary units), streams (secondary units), and stream reaches (tertiary units).  At each stage, a sample of the corresponding units is selected.  At first, a sample of primary units is selected, then, in each of those selected units, a sample of secondary units is selected.  All ultimate units (stream reaches, for instance) selected at the last step of this procedure are then surveyed.  Multi-stage sampling is sometimes used when no general sample frame exists.  In this case, a first step is to select, at random, a sample of areas (e.g watersheds) from a list of all watersheds in the domain.  Then, for each selected primary unit, a comprehensive enumeration of all units of lower rank is made, thus obtaining a local sample frame among which a sample of secondary units will be selected.
• Multi-phase sampling is similar to multi-stage sample except that the primary and secondary units are the same type.  Double sampling is an example of a two-phase survey design. 
• A cluster sample design is essentially a two-stage sample where the secondary probability is 100 percent.
• A nested sample design is essentially a two-phase sample where one set of variables is measured followed by selecting a sub-sample (second phase) of the first sample where an additional second set of variables is measured.

The reasons for adopting multi-stage or multi-phase designs may be reducing costs or reducing the sample error.  They can, however, be more complicated to implement and may increase sampling error for estimates based on second set of variables if strong relationship with variables in first set does not exist.

Once you have decided on a single or multi-stage/phase approach, you next must determine the specific survey design that is most appropriate for your monitoring needs.  You can choose a different survey design for each multi-stage or phase.  We have developed a dichotomous key that will assist you in selecting the most appropriate design. 

...click here to access this dichotomous key

Next: Go Back