***** To join INSNA, visit http://www.insna.org *****

People have used loglinear models to analyze such data, especially
egocentric data. As Garry points out, attribute-based models can also be
analyzed with the new ERGMs. Both are versions of generalized linear
models, and the relation between them is discussed here:

Koehly, L., S. Goodreau, and M. Morris. (2004). "Exponential family models
for sampled and census network data." Sociological Methodology 34:
241-270.

The relation is a bit more subtle than usual. This paper also reviews
much of the earlier literature.


On Sun, 4 Feb 2007, Balazs Vedres wrote:

> ***** To join INSNA, visit http://www.insna.org *****
>
> Dear Socnetters,
>
> This is a question about model choice with categorical independent
> variables.
>
> I have categorical attributes, and network data.
>
> When I look at the attributes one-by-one, and produce contingency tables
> with the dyadic data, each attribute has a highly significant chi-squared.
>
> When I look at them with QAP multiple regressions approach, the R-squared is
> practically zero. The network is sparse with about a thousand nodes.
>
> Of course, a QAP regression with a binary dependent variables is not really
> appropriate. I think when coefficients are referred to as probabilities in
> the QAP linear regression context, it is not appropriate. Like they are
> referred to here:
>
> http://faculty.ucr.edu/~hanneman/nettext/C18_Statistics.html#tworeg
>
> But even when I calculate geodesics (and have an interval scale dependent
> variable with an approximately normal distribution) I have no R-squared.
>
> So:
>
> Is there a way to model multiple, node level categorical independents, and a
> dyadic dependent variable?
>
> Log-linear models?
>
> A logistic regression model?
>
> Thanks
> Balazs
>
> _____________________________________________________________________
> 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.
>

****************************************************************
  Professor of Sociology and Statistics
  Director, UWCFAR Sociobehavioral and Prevention Research Core
  Box 353340
  University of Washington
  Seattle, WA 98195-3340

  Office: (206) 685-3402
  Dept Office: (206) 543-5882
  Fax: (206) 543-2516

[log in to unmask]
http://faculty.washington.edu/morrism/

_____________________________________________________________________
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.