 |
 |
***** To join INSNA, visit http://www.insna.org ***** Hi all, Thanks to everyone who had suggestions. These I've forwarded to colleagues, who are meeting next week to discuss them. They had requested I refrain from disclosing details of the project & I feel obligated to honor that. I'll speak with them next week. The query: > Some colleagues are interested in network analyses geo-coded egonets. They > have the geographic location of ego and alters, but they do not have any > information on ties between alters. I don't think network analysis has > anything to offer. > > Any suggestions? > Responses: Do they have any attribute/membership/event info on the egos/alters? This can be used to perform a two-mode analysis of possible ties. Other than that, you are right, SNA has nothing to offer. Valdis Krebs http://orgnet.com -------------------------------------------------- sounds like you're looking for a way to determine the weight of ties. How about the simply the euclidian distance? Steffen - Dr. Steffen Blaschke Assistant Professor at the Chair of Organization and Management Faculty of Business, Economics, and Social Sciences University of Hamburg ------------------------------------------------ > Do they have any attribute/membership/event info on the egos/alters? This can be used to perform a two-mode analysis of possible ties. In this case, one can use the attribute information to compute their position the in vector space (e.g., cosine values) and project distances in that space as a network on the geographical distances. See also at http://www.leydesdorff.net/gmaps . With best wishes, Loet -------------------------------------------------------------------------- passed to me from the SOCNET - feel free to add this reply to the other replies you get from SOCNET, if you think it helps or enhances the discussion If sociometric networks can be constructed from the information, then network analysis may indeed have something to offer. In other words, if there is enough information present on both egos and alters, to say who is who, then it could be possible to develop identifiers for both that are on the same par (same ID = same person). I know there's already one matching variable... geography. If there are enough other variables to safely conclude that person X is the same as person Y (regardless of whether X or Y is an ego or an alter), then you have network analysis liftoff. One other consideration... using geographical coordinates/places AS NODES in a network could conceivably result in a network analysis, even if the above is not true. (but probably not as interesting or as fruitful as knowing WHO the people in the network are, in my experience.) ----------------------------------------------------------- Lawrence E. Raffalovich Associate Professor Department of Sociology e-mail: [log in to unmask] University at Albany State University of New York Voice: (518) 442-4456 1400 Washington Ave. Fax: (518) 442-4936 Albany, NY 12222 _____________________________________________________________________ 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.