Social Space Diffusion

• In Event: 606 - Section on Social Psychology Roundtable Session (co-sponsored with Section on Sociology of Emotions)
  In Refereed Roundtable: Table 06. Social Support, Social Networks and Social Integration

Abstract

Social networks represent two different facets of social life. First, they represent the stable paths that contagions take to spread through a population, and second, they indicate a random draw from an underlying social space, which represents the relative positions of the people in the network to one another. The dual nature of networks as stable connections and as random draws from social space creates a challenge – if the observed network ties are a single random draw, is it realistic to expect that diffusion, or the spread of something through a connected group of people, only follows the observed network ties? This study integrates these perspectives by introducing a social space diffusion model. In it, network ties indicate positions in social space, and diffusion occurs proportionally to distance in social space, rather than strictly following the presence or absence of observed social network ties. Practically, the diffusion simulation occurs in two parts: the positions are estimated using a latent space model, and then the predicted probabilities of a tie from that model – representing the distances in social space – are used as weights in a weighted averaging framework. Using data from a sample of adolescent friendship networks in schools, I find that the social space diffusion model is robust to a variety of latent space model specifications, gives greater weight in the final consensus to people with high closeness centrality, and reaches the final consensus much more quickly than diffusion that only follows the observed network.

Author

• Jacob Charles Fisher, Duke University