02-05-2015 06:07 PM
I am using Add Health social network data. I have about 7000 subjects in my analytic set, and each has an average of 4 ties, (they had the option of up to 10), but not necessarily to participants in the analytic set. There are about 90,000 participants total. I want to weight the ties; 1 for unreciprocated outgoing ties, 2 for unreciprocated incoming ties, and 3 for reciprocated ties. So I need the nomination information of all the 90,000, in order to know who has nominated those in my analytic set.
The data is in the format:
Participant ID Male friend 1 ID (MF1AID) MF2AID MF3AID MF4AID MF5AID Female Friend 1 ID (FF1AID) FF2AID FF3AID FF4AID FF5AID
Weighting ties could be done in an adjacency matrix, but from what I am reading, that is a pretty big adjacency matrix. After the ties are weighted, I want to analyze the data for bridging. Ideally, I would also like to use the data from the alters to determine how many alters reported smoking. Can I use proc optnet for this? Do I need to convert the data to an adjacency matrix to do it?
I am new to using SAS for any type of social network analysis.
Any help would be greatly appreciated.