[R] Overdispersion in count data

[Gavin Simpson]

On Thu, 2008-04-03 at 01:24 +0000, David Winsemius wrote:
> "Wade Wall" <wade.wall at gmail.com> wrote in
> news:e23082be0804021244q57e359e8ic724b6f619f90153 at mail.gmail.com: 
> 
> > Thanks for the recommendations, insights.  I tried using glm.nb, but
> > it didn't seem to like my data.  I received the message (subscript)
> > logical subscript too long.  I am using the same dataframe as my
> > previous glm.  Do you know if I need to put the data in a different
> > format? 
> 
> I was wondering about your data layout. You said you had the flower/no-
> flower data in two different columns. That is not the way I usually 
> offer data to glm(). I would have imagined that log(burn_time) would 
> have been an offset. It might help if you at least offered the audience 
> a sample of ten rows, the results of str() for the data.frame, and the 
> call to the glm function.
> 

David, Wade,

You can supply a two-column matrix to glm() with families
(quasi)binomial as the two columns represent successes and failures
respectively; see ?glm and Details section.

However this is not possible with a (quasi)poisson family or with
glm.nb, which is one reason why Wade might have been getting the error.
My fault here - I didn't grep exactly what Wade had written. Even if
this isn't causing the error, Wade won't be fitting the model he thought
he was with a two column response in glm.nb.

In a similar vein to David's suggestion, could one not use offset but on
the total number of flowers sampled in each location, e.g.
offset(log(totalPlants)) in the formula where totalPlants is the
variable containing the total number of plants encountered. You need to
include this in the formula for glm.nb as that function does not have an
argument 'offset'. Wade's y is then just a vector of counts of flowering
individuals. In this way one would account for differences in the number
of flowers encountered between sites.

HTH

G

-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%