computeTMEgoDis: Compute Fujimoto, Snijders, and Valente's (2018) Ego...
In dream: Dynamic Relational Event Analysis and Modeling

computeTMEgoDis

R Documentation

Compute Fujimoto, Snijders, and Valente's (2018) Ego Homophily Distance for Two-Mode Networks

Description

This function computes the ego homophily distance in two-mode networks as proposed by Fujimoto, Snijders, and Valente (2018: 380). See Fujimoto, Snijders, and Valente (2018) for more details about this measure.

Usage

computeTMEgoDis(net, mem, standardize = FALSE)

Arguments

`net`	The two-mode adjacency matrix.
`mem`	The vector of membership values that the homophilous four cycles will be based on.
`standardize`	TRUE/FALSE. TRUE indicates that the sores will be standardized by the number of level 2 nodes the level 1 node is connected to. FALSE indicates that the scores will not be standardized. Set to FALSE by default.

Details

The formula for ego homophily distance in two-mode networks is:

Ego2Dist_{i} = \sum_{a}y_{ia}{1 - |v_i - p_ia |}

where:

\sum_a sums across all level 2 nodes in the network
y_{ia} is the 1 if node i is tied to node a and 0 else.
v_i is the value of the respondent. Within the function this is predefined to be 1 if there are multiple categories.
p_ia is the proportion of same-category actors that are tied to node a not including the ego itself.
|v_i - p_ia| is equal to 1 if all the level 1 nodes that are tied to the level 2 node share the same categorical membership and 0 if all level 1 nodes are a different category.

If the ego is a level 2 isolate or a level 2 pendant, that is, only one level 1 node (e.g., patient) is connected to that specific level 2 node (e.g., medical doctor), then they are given a value of 0. In particular, the contribution to the ego distance for a pendant is 0. The ego distance value can be standardized by the number of groups which would provide the average ego distance as a proportion between 0 and 1.

Value

The vector of two-mode ego homophily distance.

Author(s)

Kevin A. Carson kacarson@arizona.edu, Diego F. Leal dflc@arizona.edu

References

Fujimoto, Kayo, Tom A.B. Snijders, and Thomas W. Valente. 2018. "Multivariate dynamics of one-mode and two-mode networks: Explaining similarity in sports participation among friends." Network Science 6(3): 370-395.

Examples


# For this example, we use the Davis Southern Women's Dataset.
data("southern.women")
#creating a random binary membership vector
set.seed(9999)
membership <- sample(0:1, nrow(southern.women), replace = TRUE)
#the ego 2 mode distance non-standardized
computeTMEgoDis(southern.women, mem = membership)
#the ego 2 mode distance standardized
computeTMEgoDis(southern.women, mem = membership, standardize = TRUE)

dream documentation built on Aug. 8, 2025, 6:36 p.m.