We used the generalized random tessellation stratified sampling (GRTS) scheme in our study. Basically we adopted a sample size of 90 plots and an oversample size of 90 plots in case of inaccessible sites. Unfortunately, due to the high declination of access in a urban environment, we exhausted all 180 plots but could only sample 70 plots in total. My question is: should I take more samples? If so, what is the correct way to do so under the GRTS scheme? Or, should I just go ahead to analyze data using missing data techniques such as imputation? Any suggestions are highly appreciated.
Hi,
Your question: inaccessible sample plots using generalized random tessellation stratified (GRTS) sampling approach
http://www.nrs.fs.fed.us/pubs/jrnl/2009/nrs_2009_lister_002.pdf
http://www2.epa.gov/aboutepa/about-national-health-and-environmental-effects-research-laboratory-nheerl
http://alaska.usgs.gov/science/biology/ecomonitoring/pdfs/ANC_Forum_ThompsonB_Effective%20sampling%20designs.pdf
http://onlinelibrary.wiley.com/doi/10.1002/env.606/abstract
http://www.pnamp.org/sites/default/files/nbh.var_.environmetrics.pdf
http://www.clarkcountynv.gov/Depts/dcp/Documents/Library/dcp%20reports/2013/Monitoring%20Protocol,%20Test%20Use%20of%20Occup%20Sample%20to%20Detect%20Status%20Trends%20of%20Mojave%20DT%20%20in%20the%20BCCE%20v3%20-%20Feb2013.pdf
https://www.for.gov.bc.ca/hfp/frep/values/watersheds/FSW-Conceptual-Report-Nov-5-2008.pdf
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size