estimating scale in the case of over-dispersion

[Maurice Vergeer]

Hi,
I have a number of variables that all can serve as dependent variables.
These variables are heavily skewed to the right, or even zero
inflated. In those cases Poisson regression, Negative Binomial
regression (or even ZING, ZIGP) would be applicable.

However, because the variables show over-dispersion the default scale
=1 is incorrect. It's way too optimistic: all independent variables
are significant. SPSS allows to estimate the scale: deviance or
pearson. The question then is which one to choose on what grounds.
Deviance is less optimistic and pearson is the least optimistic (i.e.
the least significant relations). I've read some books and articles on
this but don't find any clear cut advice, if that's possible.

So, maybe someone on this list may provide me with advice on this
problem or point me to a source that does.

sincerely
Maurice


--
___________________________________________________________________
Maurice Vergeer
Department of communication
Radboud University  (www.ru.nl)
PO Box 9104
NL-6500 HE Nijmegen
The Netherlands

Visiting Professor Yeungnam University, Gyeongsan, South Korea

contact:
E: [hidden email]
T: [phone] (direct)/ 3612372 (secretary) / maurice.vergeer (skype)
personal webpage: www.mauricevergeer.nl
blog:  http://blog.mauricevergeer.nl/
Journalism: www.journalisteninhetdigitaletijdperk.nl
CENMEP New Media and European Parliament Elections 2009
http://mauricevergeer.ruhosting.nl/cenmep

Recent publications:
- Vergeer, M. & Pelzer, B. (2009). Consequences of media and Internet
use for offline and online network capital and well-being. A causal
model approach. Journal of Computer-Mediated Communication, 15,
189-210.
-Vergeer, M., Coenders, M. & Scheepers, P. (2009). Time spent on
television in European countries. In R.P. Konig, P.W.M. Nelissen, &
F.J.M. Huysmans (Eds.), Meaningful media: Communication Research on
the Social Construction of Reality (54-73). Nijmegen, The Netherlands:
Tandem Felix.
- Hermans, L., Vergeer, M., &  d’Haenens, L. (2009). Internet in the
daily life of journalists. Explaining the use of the Internet through
work-related characteristics and professional opinions. Journal of
Computer-Mediated Communication, 15, 138-157.
___________________________________________________________________

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Ryan]

You can compute likelihood ratio tests to compare overall fit of the nested models and Vuong tests for the non-nested models. You could also compare the AIC and/or BIC between the non-nested models.

Maurice Vergeer wrote

Hi,
I have a number of variables that all can serve as dependent variables.
These variables are heavily skewed to the right, or even zero
inflated. In those cases Poisson regression, Negative Binomial
regression (or even ZING, ZIGP) would be applicable.

However, because the variables show over-dispersion the default scale
=1 is incorrect. It's way too optimistic: all independent variables
are significant. SPSS allows to estimate the scale: deviance or
pearson. The question then is which one to choose on what grounds.
Deviance is less optimistic and pearson is the least optimistic (i.e.
the least significant relations). I've read some books and articles on
this but don't find any clear cut advice, if that's possible.

So, maybe someone on this list may provide me with advice on this
problem or point me to a source that does.

sincerely
Maurice


--
___________________________________________________________________
Maurice Vergeer
Department of communication
Radboud University  (www.ru.nl)
PO Box 9104
NL-6500 HE Nijmegen
The Netherlands

Visiting Professor Yeungnam University, Gyeongsan, South Korea

contact:
E: [e-mail]
T: [phone] (direct)/ 3612372 (secretary) / maurice.vergeer (skype)
personal webpage: www.mauricevergeer.nl
blog:  http://blog.mauricevergeer.nl/
Journalism: www.journalisteninhetdigitaletijdperk.nl
CENMEP New Media and European Parliament Elections 2009
http://mauricevergeer.ruhosting.nl/cenmep

Recent publications:
- Vergeer, M. & Pelzer, B. (2009). Consequences of media and Internet
use for offline and online network capital and well-being. A causal
model approach. Journal of Computer-Mediated Communication, 15,
189-210.
-Vergeer, M., Coenders, M. & Scheepers, P. (2009). Time spent on
television in European countries. In R.P. Konig, P.W.M. Nelissen, &
F.J.M. Huysmans (Eds.), Meaningful media: Communication Research on
the Social Construction of Reality (54-73). Nijmegen, The Netherlands:
Tandem Felix.
- Hermans, L., Vergeer, M., &  d’Haenens, L. (2009). Internet in the
daily life of journalists. Explaining the use of the Internet through
work-related characteristics and professional opinions. Journal of
Computer-Mediated Communication, 15, 138-157.
___________________________________________________________________

=====================
To manage your subscription to SPSSX-L, send a message to
[e-mail] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD