felix_bach at gmx.de
Thu Jul 26 15:51:50 CEST 2007
I agree with David. A dispersion parameter of 25 suggests that you have
mainly 0's in your data set and your model is not adequate. Perhabs you
should dichotomize your data in 0 and 1's and use a logistic mixed model but
be aware of small numbers of events.
That amount of overdispersion would make the use of a poisson model
very questionable, and will very likely result in estimated standard
errors that are too low, hence the change in statistical significance
when you switch to quasipoisson.
View this message in context: http://www.nabble.com/dispersion_parameter_GLMM%27s-tf3354683.html#a11810939
Sent from the R help mailing list archive at Nabble.com.
More information about the R-help