D. Lambert et K. Roeder, OVERDISPERSION DIAGNOSTICS FOR GENERALIZED LINEAR-MODELS, Journal of the American Statistical Association, 90(432), 1995, pp. 1225-1236
Generalized linear models (GLM's) are simple, convenient models for co
unt data, but they assume that the variance is a specified function of
the mean. Although overdispersed GLM's allow more flexible mean-varia
nce relationships, they are often not as simple to interpret nor as ea
sy to fit as standard GLM's. This article introduces a convexity plot,
or C plot for short, that detects overdispersion and relative varianc
e curves and relative variance tests that help to understand the natur
e of the overdispersion. Convexity plots sometimes detect overdispersi
on better than score tests, and relative variance curves and tests som
etimes distinguish the source of the overdispersion better than score
tests.