Building an original typology to analyse the geography of migrants’ personal networks

International audienceMigration studies have shown that individuals, families and diasporas can develop their relationships and networks over large distances and transnational space, while places and spatial proximity continue to strongly structure social networks. The role of space and place in migrants’ personal networks has, however, often been analysed in terms of the place of origin and destination, which fails to capture the diversity of spatial patterns of personal networks, including alters’ clusters in some places but also their geographical scattering. In this paper, we define migration as moving beyond one’s place of living measured as large employment areas. We present a novel typology for analysing migrants’ personal networks based on their geographical scope, the distribution of network members (alters) across employment areas, regions and countries, and the extent to which alters live in the same area as the respondent (ego). By combining the disciplines of sociology and human geography, and qualitative and quantitative network analysis, we seek to develop an approach that accounts for the complexity of social networks and can be replicated across diverse databases. Our data come from two different egocentric network datasets: (1) the Panel of Caen (France), which is a qualitative follow-up study based on context-focused name generators capturing large networks (87 Egos along 5 survey waves with a mean size of 37 alters); (2) the nationally representative 2013 MOSAiCH survey including the ego-centric family networks of 666 adults living in Switzerland. Personal networks were based on a unique name generator asking respondents to self-define their meaningful family members. We apply our typology to these two datasets to identify and map the main geographical patterns of migrants’ personal networks. We then aggregate alters into places to generate networks of places from alters’ relationships and examine the main characteristics of these patterns using standard network indices. In particular, we characterise the part of the links between places that depends on the relations of the Ego with his/her alters, and the part that rather depends on the relations between alters. This allows us to measure the structural importance of Ego in the geography of his/her personal network. Finally, we analyse which migrants’ characteristics explain differences in these patterns of relationships between places. We finish our presentation by discussing the significance and limitations of our approach to integrate geographical information into the analysis of personal networks