Premium
Bias in Estimating the Diversity of Large, Uncensused Communities
Author(s) -
Routledge R. D.
Publication year - 1980
Publication title -
ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.144
H-Index - 294
eISSN - 1939-9170
pISSN - 0012-9658
DOI - 10.2307/1935186
Subject(s) - quadrat , estimator , diversity index , statistics , sample (material) , index (typography) , sampling (signal processing) , mathematics , diversity (politics) , ecology , sample size determination , econometrics , computer science , biology , species richness , sociology , telecommunications , physics , anthropology , shrub , detector , world wide web , thermodynamics
If a community is too large to be censused, quantities such as the diversity must be estimated from a sample. Methods using data from a sample random sample of individuals to estimate either the Shannon—Wiener or the Simpson Index of diversity are of limited use. Seldom is it practical to obtain such a sample. It is usually easier to use sampling unit such as the plants in a quadrat or the animals trapped in a net. Methods which use the data from such a scheme to estimate the two commonly used indices are discussed here. (Computational procedures are outlined in a separate display.) The small sample bias, at least in the particular case examined in detail, is much smaller for an estimator of the Simpson Index than for any proposed estimator of the Shannon—Wiener Index. This property appears to provide the former index with a distinct advantage for measuring diversity and its ecological components.