Socioeconomic Correlations and Stratification in Social Communication Networks

Yannick Léo 1, 2 Eric Fleury 2, 1 Carlos Sarraute 3 Márton Karsai 2, 1
1 DANTE - Dynamic Networks : Temporal and Structural Capture Approach
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme, IXXI - Institut Rhône-Alpin des systèmes complexes
Abstract : The uneven distribution of wealth and individual economic capacities are among the main forces, which shape modern societies. They arguably bias the emerging social structure, lay behind spatial and social segregation, and induce differences in terms of consumer behaviour, mobility, or even communication patterns. However, the observation of these correlations is difficult due to the com- plexity of their dependencies, and due to the lack of multidimensional data, which simultaneously records the economic status and social network of individuals. In this work we aimed to close this gap through the analysis of a coupled dataset of mobile phone communication records and bank credit history of a large set of individuals living in Mexico. After mapping the spatially em- bedded social structure and estimating individual economic capacities we addressed four different aspects about their correlations: (a) We show that individual economic indicators such as average monthly purchases, and surprisingly individual debts are unevenly distributed in the population in agreement with the Pareto’s principle (b) After grouping egos into nine socioeconomic classes we detect signatures of status homophily and find evidences of stratification in the socioeconomic network (see Fig.1a). We also observe that the social structure is upward biased towards wealthier classes (see Fig.1b) and show assortative correlations giving rise to strongly connected ”rich-clubs” in the network. (c) Further we demonstrate that people of the same class live closer to each other, and their commuting distances are positively correlated with their economic status; (d) We de- tect purchasing patterns typical to socioeconomic classes and show them to be correlated with the social strata and finally we quantify the importance of the social network in terms of shaping consumption behaviour.
Type de document :
Communication dans un congrès
NetSci'16, May 2016, Seoul, South Korea
Liste complète des métadonnées
Contributeur : Márton Karsai <>
Soumis le : vendredi 25 novembre 2016 - 17:12:42
Dernière modification le : vendredi 20 avril 2018 - 15:44:27


  • HAL Id : hal-01403331, version 1


Yannick Léo, Eric Fleury, Carlos Sarraute, Márton Karsai. Socioeconomic Correlations and Stratification in Social Communication Networks . NetSci'16, May 2016, Seoul, South Korea. 〈hal-01403331〉



Consultations de la notice