Heterophily in social groups formation: a social network analysis

Research on heterophily as a relational phenomenon is mostly limited to specific professional contexts which do not include knowledge about status or value heterophily as a possible group-generating principle. This article tries to close this gap by proposing and applying a new method of data analysis to investigate this possible role of heterophily. We apply social network analysis to personal network data. As a preliminary, we introduce a conceptual distinction between two types of heterophily, what we call intra-category heterophily and inter-category heterophily, and we validate the adjusted residuals of contingency tables to measure these two types of heterophily. Then, we reconstruct the relational space of these heterophilous relations among categories or attributes defined by socio-demographic and status characteristics. Finally, we group by faction algorithm the categories or attributes that maintain denser heterophilous relationships with each other than with the rest. The methodology proves to be valid and useful for achieving the desired analytical objective, revealing that status-heterophilous relationships can be considered as guiding principles, or mechanisms, for generating groups of social categories.