***** To join INSNA, visit http://www.insna.org *****
Our next release of Statnet will have some tools for dealing with
egonets in both cross-sectional and dynamic networks. We'll be talking
about them in the "Stergm" workshop at the Sunbelt, which is when the next
release will be available.
Specifically, the methods will allow you to estimate an ERGM for a
complete network that has the statistics observable from the egonetworks
(e.g., degree distributions and selective mixing). Once you have the
estimated model, you can simulate (complete) networks with these
properties. Each simulation is a draw from the distribution of networks
specified by the model.
The dynamic form requires additional information on partnership dynamics
(at minimum, the average duration), and estimates a model that allows you
to simulate complete dynamic networks over time.
This is work that Mark Handcock originally developed, and it is being
implemented and extended by Pavel Krivitsky.
On Mon, 20 Feb 2012, Miranda Jessica Lubbers wrote:
> ***** To join INSNA, visit http://www.insna.org *****
> Dear Emese,
> 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!
> Miranda Lubbers
> 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.
Professor of Sociology and Statistics
Director, UWCFAR Sociobehavioral and Prevention Research Core
University of Washington
Seattle, WA 98195-4322
Office: (206) 685-3402
Dept Office: (206) 543-5882, 543-7237
Fax: (206) 685-7419
[log in to unmask]
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.