***** To join INSNA, visit http://www.insna.org *****
If I understand you correctly, you mean with "overall model" that you combined the ego-networks into one network with missing values for pairs of individuals who do not belong to the same ego-network. If you have a sufficiently large number of alters per ego, you could alternatively run an ERGM per network and then perform a meta-analysis on the results (similar to Tom Snijders and Chris Baerveldt, Journal of Mathematical Sociology, 2003, and myself, Social Networks, 2003) - in which you can also include ego or network characteristics as explanatory variables. Personal networks tend to be very heterogeneous in structure, and the use of one overall network doesnīt do justice to that. But of course, such an approach is only useful for large networks (Also, you would probably wish to leave the egos out).
As Lindsay said, HLM is an alternative approach to study the likelihood of a tie (or tie contents), but itīs typically used for ego-alter pairs, not among alters - if thatīs what youīre after. One exception is the paper of Hugh Louch (Social Networks, 2000) with small networks. You would need to assume that within personal networks, dyads are independent of one another.
There is also a paper by Yuval Kalish and Garry Robins (Social Networks, 2006), which gives a nice alternative to studying personal network structures cross-sectionally.
I hope this is useful for you. Good luck!
Ramon y Cajal Researcher
Laboratory for Personal Networks and Communities (egolab-GRAFO)
Department of Social and Cultural Anthropology
Autonomous University of Barcelona
----- Original Message -----
From: Emese Domahidi <[log in to unmask]>
Date: Monday, February 20, 2012 5:52 pm
Subject: [SOCNET] Ego Networks & ERGM
> ***** To join INSNA, visit http://www.insna.org *****
> Dear all,
> I am computing ERGM Models for ego networks of approximately 90
> egos and
> their alteri. The ego networks are not interconnected. I
> constructed an
> overall model for all networks, controlling for different network
> structures (e.g. edges and gwdsp) and different individual and dyadic
> attributes. The model is converging and the model fit is ok.
> My question is whether constructing an overall model instead of
> looking at
> each ego network individually is an appropriate way to model ERGM for
> ego networks. I would really appreciate comments on this approach
> as well as
> literature recommendations on ego networks and ERG models.
> Cheers, Emese
> 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.
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.