NETWORK > EGO NETWORKS >CSS BROKERAGE
PURPOSE Counts brokerage scores for cognitive slices of the same network and the user defined true network. These slices may either be a number of views of the whole network or the view of the whole network through all egos.
DESCRIPTION The CSS input is a set of k adjacency matrices, each of the form A(i,j) stacked into a three-dimensional matrix, A(i,j,k). This form is useful for cognitive social structures, where k refers to the perceiver of a relation from i to j. This routine counts the number of structural holes, that is the number of non-connected alters for each of the k adajcency matrix as well as the user defined true network. The alters and the ties in the networks can be based on a combination of reciprocated and unreciprocated ties from either the true network or the perceived networks.
Dataset containing individual perceived networks (CSS):
Name of file containing any set of matrices representing the same network. Data type: Valued graph. Multirelational.
Name of file contating the same actors that represents the true network
Output Measures (Default = 'CSS Brokerage Counts')
Output file that will contain brokerage counts graph .
What is Ego to Alter tie?
Buttons to select how to define the alters in terms of directions of ties from ego to alter.
What is Alter to Alter tie?
For the aletrs defined above whether ties need to be reciprocated or not to count as ties.
Define Alter to be based on?
Buttons to chose which network or combination of perceived and true network to use to define the ties from ego to alter.
Frequencies gives total counts. Proportions give fraction of eact type of hole.
LOG FILE A table giving the number of holes size and number of pairs. Each row is an ego and hence row p will use as the perceived network the pth matrix in A. A hole is simply a pair of alters who are not connected. The columns are as follows:
TP = No. of holes found in both true and perceived networks
Tp = No. of holes in true network but not perceived network
tP = No. of holes in perceived network but not true network
tp = No. of pairs of alters that are not holes in either true or perceived network
Size = Number of alters in the combined ego network
Pairs = Number of pairs of alters in the combined ego network
REFERENCES Krackhardt D. (1987). 'Cognitive social structures'. Social Networks 9, 104-134. (Note this paper explains the CSS structure and does not discuss brokerage)