Dear SOCNETters,

I am doing research on the social capital of groups of individuals who work
together.  My study focuses on Episcopal Church vestries, which are governing
boards for individual congregations.  My theoretical framework ties network
closure and cohesion with regard to (1) ideological agreement and (2) the
extent of friendly and long-lasting ties among group members to group-level
social capital, and then ties group-level social capital to group
performance.  I'm hoping that the subscribers to the listserv might be
willing to help me with a couple of questions:

1.  The vestries have 8-15 members each.  To do whole-network/vestry/group
measures I obviously need a high response rate from each
network/vestry/group.  The data is quite sensitive and somewhat difficult to
collect.  From a research standpoint, is there a minimum number of groups
needed to do meaningful cross-group comparisons?  How do you account for
non-respondents in data analysis?

2.  Another data-collection idea is to ask individual vestry members
questions like "Who agrees with each other with regard to theology?" and ask
respondents to assess whether member A agrees with members B, C, D, etc (and
B with C, D, etc.)--and so respondents would assess such agreement for every
possible dyad within the group.  (Member A would be the respondent, allowing
for ego-centric measures as well.)  Have other researchers tried this sort of
approach?  Is it a valid way of gathering data about the social structures of
a particular group?  And finally, has anyone studied whether one's position
within a particular group affects his or her responses to such questions (for
example, do more central or more isolated people differ in some sort of
predictable way in their perceptions about who is connected to who in the

I'm grateful for any advice you can provide.  Thanks in advance--

Harry Van Buren
Doctoral student, University of Pittsburgh (Katz Graduate School of Business)
Visiting Instructor, University of Northern Iowa
319.273.2020 telephone
319.273.2922 facsimile
[log in to unmask] (personal)
[log in to unmask] (academic)