Answers are grounded in this paper and cite their sources.
Needham and Usinger’s analysis of data on the total numbers of animals from 100 samples, each 1 ft2, from a riffle is shown to be erroneous. Their conclusion that 73 samples were needed to arrive at significant figures on total numbers at a 95% level of confidence is optimistic; 448 samples are required to give a sample mean within 5% of the population mean at a 95% level of confidence. Data from sampling sites where the sampling apparatus is totally submerged are particularly variable.
No takes yet. Share an insight, caveat, or question.
F. M. Chutter (1972) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: