Contents - Index

NETWORK>COHESION>RECIPROCITY

PURPOSE Calculate the amount of ties that are reciprocated in a network.

DESCRIPTION A tie is reciprocated if whenever a tie is connected from actor A to actor B then there is a tie from actor B to actor A. We can either count the number of dyads connected by a tie (which may or may not be reciprocated) and calculate the proportion of dyads that have reciprocated ties (the dyad based method) or we can count the number of arcs (directed edges) and calculate the proportion of arcs that are reciprocated (the arc based method). The routine also calculates the amount of reciprocated ties between partitions of the actors, in this case a combination of arc based and dyad based measures is possible and this is known as the hybrid method.

PARAMETERS
Input dataset
Name of file containing dataset to be analyzed. Data type: Directed graph.

Method:
Arc based. The number of reciprocated arcs divided by the total number of arcs
Hybrid. The arc based method is used on the reciprocities between the groups and the dyad based method is used within the groups. Note if the data is not partitioned this is the same as the dyad method.

Row partitioning / Blocking (if any):
The name of an Ucinet dataset .To partition the rows of the data matrix into blocks, specify a blocking vector by giving the dataset name, a dimension and an integer value. For example, to use the second column of a dataset called ATTRIB, enter "ATTRIB COL 2". The program will then read the second column of ATTRIB and use that information to sort the rows of the matrix. All rows with identical values on the criterion vector (i.e. the second col of attrib) will be placed in the same block of the matrix.

Column partitioning / Blocking (if any):
The name of an Ucinet dataset.To partition the columns of the data matrix into blocks, specify a blocking vector by giving the dataset name, a dimension and an integer value. For example, to use the second column of a dataset called ATTRIB, enter "ATTRIB COL 2". The program will then read the second column of ATTRIB and use that information to sort the columns of the matrix. All columns with identical values on the criterion vector (i.e. the second col of attrib) will be placed in the same block of the matrix.

Utilize diagonal values:  (Default = NO)
For square matrices NO means that diagonal entries are ignored.

Output node dataset: (Default=NodeReciprocity)
Name of file containing reciprocity values for each actor in the network

Output group dataset
Name of file containing reciprocity values within and between groups.

LOG FILE If no partition is given the output gives the reciprocity of the whole network for the selected method.
If a partition is given the output gives the groups followed by a blocked reciprocity matrix within and between groups using the measure selected. Note this matrix must be symmetric for the dyad based method.

Reciprocity proportions are then given for each actor by considering the ties to and from each as taken as ego. (Note these are the same at actor level regardless of the method selected and are based on dyads) These are given in columns as follows.

Symmetric: The total number of reciprocated ties involving ego divided by the number of ties to and from ego.
Non-symmetric. One minus the symmetric score.
Out/NonSym. The number of non-symmetric out-going ties divided by the total number of non-symmetric ties.
In/NonSym. The number of non-symmetric in-coming ties divided by the total number of non-symmetric ties.
NonSym/Out. The number of non-symmetric out-going ties divided by the total number of out-going ties.
NonSym/In. The number of non-symmetric in-coming ties divided by the total number of in-coming ties.

TIMING N/A