***** To join INSNA, visit http://www.insna.org *****
I have a question about which statistical models work best for combining
both monadic and dyadic independent variables. Hope you can help!
So far, I have used dyadic data to analyse whether there is a tendency that
people will coordinate their choice of mobile phone operator (because calls
are cheaper if two people are on the same network). The dyadic data I used
are for example friendship relations and whether two respondents are of the
same nationality. I estimated a logit model and then adjusted the standard
errors by a QAP permutation procedure.
In a next step, I would like to add attribute data of individuals into the
model and see whether, for example, more price sensitive individuals are
more likely to coordinate operator choice. (The attribute data could be
binary, categorical or continuous. The intention of my model is to estimate
parameters for all independent variables - they are not just confounding
Are there extensions of QAP permutation tests to incorporate both dyadic and
monadic data as the one mentioned above? Or should I rather use an
alternative statistical model such as p*? Can you point me to some
references? I would also appreciate any suggestions of software with which
such models are relatively easy to estimate.
Thank you very much in advance,
Nottingham University Business School
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.