REAL
Keystone sector methodology applied to Portugal
3.2. TT10 study
To achieve our goal we carried out a survey similar to the one described in
Laumann (1985). We identified 140 entities located in town including private, public
and non-profit institutions, and defined a direct interview with their main leaders. We
tested the survey in order to identify possible questions that leaders would raise during
the interview, to guarantee the privacy of each entity response. However, the surveyed
leaders were informed that a University researcher wrote the questions, and that they
were similar, for comparative purposes, to those within an international project.
We collected 83 usable responses (60%) and 3 were negative responses
considering privacy reasons. Nevertheless some of the 83 entities revealed some
reluctance to answer part of the survey, namely the questions related with money flows.
Based on the answers we knew about the existence/absence of relational links in
money, support and information each entity gives and receives to/from the other 82 on
the list to solve community affairs11.
The relational link between agents is used to study our test hypothesis. The
answers in the survey allowed us to complete six binary matrices (1/0). By convention,
each row in the matrix contained the sender of the flow, while each column housed the
receipt of that same flow. Consequently we define an 83x83 matrix with diagonal zero,
considering non-reflexive relationships (aij∙ gives no information to aij∙ himself).
Based on each pair of questions (give/receive) we defined three main relations
we denominate as INFORMATION, MONEY and ORGANIZATION12. As we studied
more than one link between agents at a time this is called a multiplex relation
analysis13.
The relational link between two actors is called a tie, so that a tie is a property
of the pair because it cannot pertain simply to one of the actors. As the relations are
directional (links goes from a node to another), the matrices do not have to be
symmetric.
In this analysis we look for dyads, which consist of a pair of actors and the
possible ties existing between them (Wasserman and Faust, 1994). As an example, we
10 TT is the symbolic name we choose for the studied town.
11 As defined in LAUMAN, E.O. (1985).
12 Because these are the main resources for collective action.
13 As defined in Wasserman and Faust (1994).
10