You could try to fit the generalized event count model to see what
dispersion pattern the data tell you unless you have a theory to predict
more complex dispersion patterns.
Kosuke
On Sat, 12 Apr 2003 dhopkins(a)fas.harvard.edu wrote:
Dear All,
Quick question: in the data we are working with, people were asked how many
public meetings they had attended in the last year. The range of answers is
from 0 to 60, but the distribution across those values is spiked at the even
numbers (i.e. 612 people say 10 meetings and 634 say 12 meetings, while only 8
say 11 meetings, and 8 say 13 meetings). We were hoping to use the generalized
event count model, but were curious if that was a good choice given the data
and what seems to be its alternating pattern of overdispersion and
underdispersion.
Best,
Colin and Dan
--
PhD Candidate in Government
Proctor
Harvard University
(617) 493-8402
_______________________________________________
gov2001-l mailing list
gov2001-l(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov2001-l