In models of the fraction of fish recaptured in field experiments on g
ear efficiency the binomial error distribution is usually assumed. How
ever, variance in excess of that defined by the error distribution (ov
erdispersion) is typical in fish capture because of heterogeneity amon
g and within groups of individuals and incomplete model specification.
Quasi-likelihood offers a parsimonious solution to the typical proble
m of incomplete definition of an error distribution with discrete resp
onses. An example is given from the recapture of marked fish following
rotenone treatment in lake enclosures, in which a generalized linear-
logistic model includes an extra-binomial variance as a function of th
e mean. Estimated standard errors of fitted parameters were two to thr
ee times lower in a linear-logistic maximum likelihood model than in t
he quasi-likelihood model because extra-binomial variation (overdisper
sion) was ignored in the former model. In a cross-validation trial, 95
% confidence intervals included 85% of independent observations with t
he quasi-likelihood model compared with 69% with the maximum likelihoo
d model.