657,269
edits
m (Text replacement - " *sample " to " sample ") |
m (Text replacement - " *observations " to " observations ") |
||
Line 295: | Line 295: | ||
<HR> | <HR> | ||
<B> [[Extra-binomial variation|Extra-binomial variation]]</B>: Greater variability in repeat estimates of a population proportion than would be expected if the population had a 'binomial distribution. For example, suppose that <math>n </math> | <B> [[Extra-binomial variation|Extra-binomial variation]]</B>: Greater variability in repeat estimates of a population proportion than would be expected if the population had a 'binomial distribution. For example, suppose that <math>n </math> [[observation]]s are taken on *independent *Bernoulli variables that take the value l with *probability p, and the value 0 with probability <math>1 - p </math>. The [[mean]] of the total of the observations will be <math>np</math> and the ‘variance will be <math>np(1 - p)</math>. However, if the probability varies from variable to variable, with overall mean <math>p</math> as before, then the variance of the total will now be > <math>np(1 - p) </math>. In the context of plant and animal populations, extra-binomial variation may be termed over dispersion. See also INDEX OF DISPERSION. | ||
<HR> | <HR> |