***** To join INSNA, visit http://www.insna.org *****
Thanks for the suggestion Tom, I shall try.
THE SAMPLE:I'm currently working with a data set in spss using
standard statistical procedures. The survey data is collated from a
sample of people from around the world who all have a related
interest and similar behaviour patterns around this interest. I'm
assuming that this group of people is a social network, with all the
accompanying characteristics (core, periphery, nodes, hubs, loose
networks of people linked in certain ways etc).
THE ANALYSIS; I am currently testing the correlations between an
index of information sources and people within this network who
exhibit particular behavioural characteristics. I have correlation
scores to indicate the strength of relationships between the two
variables. However these correlation scores are pegged to standard
data sets based on the approximate principles that r=.10 to +-.29 is
a weak correlation, r=0.30 to+- 0.49 is medium correlation and r=0.5
to +-1.0 is strong correlation.
THE QUESTION: My question relates specifically to this. Is there
any work done to show whether these correlation score cut offs are
that are specifically relevant to a collectivity measured as a social
network. Or to put it another way, whether correlation of two such
variables from a standardised social network data set has specific
cut off rates that better indicate the strength of the relationship
between two variables.
I would appreciate any insights or help you could give me on whether
a)this question can be asked of social network data sets and b) if
there is an answer to it.
>***** To join INSNA, visit http://www.insna.org *****
>I'm a research student (Australia) and I'm playing around on the
>periphery of social network methods and theory. I wondered if anyone
>could point me in the direction of work on levels of correlation
>between members of a social network. (Or even if such statistics can
>be used against this form of social modelling).
>Many thanks for your help,
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.