TPRC46: Research Conference on Communications, Information and Internet Policy

Abstract
Homophily is a social concept with wide-ranging consequences. It has been shown (for the Swiss Parliament) that politicians match referendum outcomes based on homophily to their resident communities; it has been demonstrated that (United States-Mexico) immigration follows degrees of homophily; it has been invoked in forms to explain foreign policy decisions, political polarization, and echo chamber effects.

Emerging technologies such as social media are useful in studying these effects; their structure allows for analysis of personal networks. Prior research has showed that users of virtual spaces tend to exhibit homophily, and that aspects of homophily are clearly visible between Facebook users in the United States, and that this effect extends to state-level divisions.

In this paper, we use data supplied by Facebook – a network of over 2 billion people to date to examine the following questions:

1. Does social media connectivity between nations mirror homophily?
2. What shared factors between nation pairs, such as migrant stock, trade, geographical contiguity, shared languages and value systems, contribute most to the explanation of such digital connectivity?
3. From the above, can we infer how people of different nations form social connections, and can the study of social media networks provide a useful metric of people-to-people connectivity and homophily for the world?

To answer these questions, we consider traditional metrics of homophily as defined by McPherson, Smith-Lovin, & Cook (2001). This is expressed in our collected dataset as distances between countries, geographical contiguity, shared languages and value systems. We use regression modelling and clustering methods to identify correlations and patterns. Exploratory analysis has shown that Facebook connectivity expressed this way shows a high correlation to international migration and trade statistics.

As part of a secondary research question, we examine how the clustered data compares to models such as that by Huntington, S. (1997), which was influential in US politics, as well as the cultural dimensions theory put forward in Hofstede, G. (1980). The novelty of this research lies in the possibility of unearthing a data source that may help the world understand digital people-to-people connectivity at grand scales, using data-driven constructs, and the impacts to policy – especially in sociology, communication and economics - of such a metric.


References:
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual review of sociology, 27(1), 415-444

Huntington, S. (1997). The clash of civilizations and the remaking of world order. New York :Touchstone

Hofstede, G. (1980). Culture’s Consequences: International Differences in Work-Related Values. Cross-Cultural Research and Methodology. Thousand Oaks: SAGE