I am currently using four measures of homophily using UCINet’s analyses to examine gender homophily, as well as internal/external tie homophily (is the alter within the same group or outside the group). In UCINet, select the analysis, your dataset and a categorical attribute to get the results. Here are my notes on the four types and the interpretation of the data.
If I am misunderstanding any of these measures, I would appreciate someone responding and clarifying
Pct Homophily = Percentage of ego's ties that have same (attribute); Percent: Number of ties between ego and an alter in the same attribute category divided by ego's total number of ties.
E-I Index = Index of whether ego's ties are homophilus, based on proportion of ties in-group (homophily) and out-group (heterophily)
Neg result = more internal (homophily)
Pos result = more external (heterophily)
Closer to zero = more balanced; Closer to 1 = 1 less balanced
Yules Q = measure of alter similarity = A measure of similarity which ranges from -1 for perfect heterophily to +1 for perfect homophily. A value of 0 means no pattern of homophily.
Cohen Kappa: measure of complete agreement (homophily)= The measure of categorical agreement on the specified attribute for the ego with each alter. A value of 1 means complete agreement. A value of 0 means no more agreement than expected by chance.
Julie M. Hite
Brigham Young University
Dept. of Educational Leadership & Foundations
Provo, UT 84602
(voice mail creates audio file to my email)