Population-scale Social Network Analysis

Lecture by Frank Takes for the Interaction Data Lab of the Center for Research and Interdisciplinarity (CRI) in Paris.

Abstract

This talk considers responsibly anonymized population-scale social network data on all 17 million inhabitants of the Netherlands. The data stems from country-wide administrative register data, and has the potential to shed new light on contemporary social scientific problems such as segregation, inequality, loneliness and poverty. The talk discusses how the formal links (family, household, work, school and neighbor ties) in this social network require one to critically rethink network science concepts such as the unit of analysis, measurement errors effects and the boundary specification problem. Moreover, it allows us to in a unique way revisit the well-known concept of closure and the small-world phenomenon in a population-scale social network context. The talk furthermore presents initial findings on the relation between the network structure and spatial distribution of the population.