Investigations of network inequality have long relied on the logic of preferential attachment, holding that newcomers to a social network prefer to form ties with central actors — who presumably are more valuable as network partners — rather than peripheral actors. We develop an alternative account to explain the emergence of network inequality that is based on local structural processes that do not require network actors to have knowledge about the social position of others.
Instead, we propose that central actors benefit from being exposed to more opportunities for triadic closure, which confounds the quality- or popularity-based signal that their greater connectedness might also send. After theorizing about our mechanism, we develop a simulation model to describe how triadic closure might cause network inequality. We then present results from an observational study and a field experiment across multiple professional conferences to disentangle an attendeeâ€™s network centrality from her opportunities for triadic closure as correlates of her tendency to form new ties.
In the field experiment, we compare a randomly assigned set of conference attendees for whom opportunities for triadic closure are highly visible to another set of attendees for whom such opportunities are obscured. The findings from this field experiment demonstrate that by minimizing exposure to possibilities for triadic closure, the tendency of already central actors to become even more central all but disappears. We discuss the implications of our findings for those who often find themselves at the periphery of a social network (e.g., racial/ethnic minorities and women).
De Vaan, Mathijs, and Dan Wang. "Micro-Structural Foundations of Network Inequality: Evidence from Observation Data and Field Experiments." Columbia Business School, 2019.
Each author name for a Columbia Business School faculty member is linked to a faculty research page, which lists additional publications by that faculty member.
Each topic is linked to an index of publications on that topic.