[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: poisson vs. ologit for pinprick data
The ordered logit approach seems more appropriate here than does Poisson.
The latter is intended to model events, of which the independence between
them is moot. -ologit- also appears to be appropriate since your response
variable is indeed ordered, but not interval (and, reading between the
lines of your post, is measured only once).
The variance being greater than the mean value appears to be neither here
nor there: all that is telling you is that there is a lot of unexplained
variation in your sample (how large is it, out of interest?), which your
independent variables may or may not account for.
Hope that helps! :-)
Tim Wade wrote:
> Dear Statalisters:
> I've been off this list for a while and am greatly
> enjoying my return to this excellent resource and
> I'm seeking advise for the appropriate model for
> pinprick data scaled from 1 to 5 where 1 represents
> the most sensitive and 5 the least to the pinprick
> stimulus. I need to model this as a function of
> several dependent variables. The pinprick data are
> distributed something like Poisson, but the variance
> is considerably less than the mean (mean=1.28,
> var=0.52), and it does not seem to be a real "count"
> type of outcome. Could a poisson model be appropriate
> here? Or an ordered logit? Any other suggestions?
> Thanks very much.
CLIVE NICHOLAS |t: 0(44)191 222 5969
Politics Building |e: firstname.lastname@example.org
School of Geography, |f: 0(44)870 126 2421
Politics & Sociology |
University of |
NE1 7RU |
United Kingdom |http://www.ncl.ac.uk/geps
(c) The University of Newcastle upon Tyne is a registered trademark of
Universities UK, a division of World Wrestling Entertainment, a
subdivision of Fox Inc., part of Muddy Fox, a joint-stock partnership with
the band Mud. Muddy Fox is a subdivision of IMG/McCormack, a division of
Coca-Cola, part-owner of God, Jesus, Moses and Son (GJMS) Miracles plc.
GJMS Miracles is a registered trademark of the University of Newcastle
upon Tyne. All rights reserved 2003.
* For searches and help try: