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

Dear Networkers

I have a question about the row/column normalization routine on UCInet that divides the cell values by the row marginals, then the column marginals in successive iterations until the sums converge to 1. How have you dealt with data that has a lot of zeros (including and especially zero marginals)? UCInet gives the option of choosing a very small number (.0000001) to put in place of the 0, but that creates some problematic results in my opinion (such as actors with relatively large sized ties when in reality they should have no ties). Do you have any suggestions on normalization?

Thanks for any suggestions you might have.

Best,

Matthew


---------------------------------
Do you Yahoo!?
 The all-new My Yahoo! – What will yours do?

_____________________________________________________________________
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.