|
Hi, I am not a statistician and I would like to get some recommendation regarding whether to create a random or systematic sample plots. I will need to stratify it according to LULC but I don't know whether I should go for random or systematic within each strata. Are both as statistically robust? I would like to have peer-reviewed articles to back up any assertion one way or another. Thanks, Florence |
|
Thanks! I will try and digest. I will come back if I am still unsure. Cheers, Florence |
|
Dear Florence, Plese refer to this paper on "Dot sampling method for Area Estimation" (point 3.2) that includes an explanation of which method is more robust. Basically for practical reasons (easier to understand andto produce) we always reccomend the Systematic sampling. I reccomend you read the whole document as it containsa good explanation of the dot sampling method. When it comes to area calculations the procedure is the same ( please see section 2.1) regardless of random or systematic sampling as the probabilities are the same. Hope this helps! |
