2.6.1.2.7 GRTS non-stratified design pros & cons

Generalized Random-Tessellation Stratified (GRTS) design produces a probability sample with design-based variance estimators.  It provides a spatially balanced,random sample. allows for unequal probability sampling, and can provide an over-sample of sample sites to accomdate field implementation issues.  There may often be factors which divide up the population into sub-populations (groups / strata) and we may expect the measurement of interest to vary among the different sub-populations. This has to be accounted for when we select a sample from the population in order that we obtain a sample that is representative of the population. This is achieved by stratified sampling.  A non-stratified sample does not take separate samples from strata or sub-groups of a population.

Tools:

Software to implement such a design is available on the Aquatic Resource Monitoring web site or directly from R project package spsurvey.

Pros and Cons:

The following pros and cons of GRTS non-stratified designs should assist you in determining if it is appropriate for your monitoring needs.

Site selection

Pros:

• Provides representative sample
• Provides spatially balanced sample 
• May achieve desired sampling error with smaller sample size than a simple random sample design
• Allows replacement of sites if sites are dropped (for valid reasons)

Cons:

• Does not incorporate characteristics of the population or information known about target population
• Can require larger sample sizes than other spatial designs that incorporate information about the target population to achieve desired sampling error
• Sample selection procedures are not as readily available as for simple random sample
• Are relatively new, not yet widely used and it can be difficult to understand the GRTS site selection procedure
• Can increase field operation costs due to inaccessibility of sites
• Errors in the sampling frame may result in the sampling frame excluding some sites that are in the target population or including some sites that are not in the target population
• Sampling frames based on different GIS scales (e.g., 1:100000 and 1:24000) may result in different estimates for the target population

Statistical Inference

Pros:

• Procedures for estimating characteristics of target population are available and result in unbiased estimates and associated unbiased estimates of variance
• Variance estimates are unbiased and may result in smaller variance estimates (compared to simple random sample) when a local neighborhood variance estimator is used.
• Allow standard statistical methods to be used
• Precision and power depend on sample size and variability of the metric
  

Cons:

• Local neighborhood variance estimator is difficult to understand and compute without using  spsurvey software
• When simple random sample variance estimators (i.e., Horwitz-Thompson) are used instead of local neighborhood variance estimator and the sampled variable exhibits a spatial pattern, variance estimates are biased high

Next:  Spatial Design Results and Next Steps | Go Back