***** To join INSNA, visit http://www.insna.org *****
Hello ISNA communitiy:
I'm about to use some basic measures on 14 networks in regard to "cooperation between public and private actors on road safety in the Netherlands" since part of the data is perfect netwerkdata. (I work on this research together with the Dutch Institute for Road Safety).
The networkdata contains 3 matrices with the following ties:
a. "frequency" (valued)
b. "positive contribution to the co-operation" (valued)
c. "knowledge about other's points of view in regard to road safety" (valued)
I tried out some of the basic measurements in Ucinet (6.29) on the frequeny network for 1 case in order to get everything right for the other 13 to follow:
- In and Outdegree (dichotomised data)
- Degree (symmetric data)
- Overall network centralization
- Subgroeps N-clicques
Because comparisation between the netwerks is important I always use the normalised output.
Overall I will manage to find my way with these measures, but 3 small questions remain, although they must be hiding somewhere in my recent pile of SNA manuals and literature:
1) What is most meaningfull: compute the overall network density (Ucinet->Network->Network Properties->Density) based on symmetric or assymetric data?
2) Maybe another easy one: how do I in the end compare network density between 14 networks of DIFFERENT sizes? The networks are relatively small (mean = 25 actors), but some might consist out of 10, others 30.
3) After I compute the In en Out degree centrality, I also compute the degree centrality based on symetric data.
Now: I find that the mean degree centrality (symmetric data) in this case equals exactly the networks density (based on symmetric data). Are these two the same kind or is it coincidental?
Thank you for any help offered!
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.