***** To join INSNA, visit http://www.insna.org ***** Hi Socnet, 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 variables.) 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, Daniel Nottingham University Business School Doctoral Programme _____________________________________________________________________ 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.