Identification of Peer Effects through Social Networks

By Yann Bramoullé, H. Djebbari, B. Fortin. 2009

1,351 citations recorded by the source corpus.

We provide new results regarding the identification of peer effects. We consider an extended version of the linear-in-means model where each individual has his own specific reference group. Interactions are thus structured through a social network. We assume that correlated unobservables are either absent, or treated as fixed effects at the component level. In both cases, we provide easy-to-check necessary and sufficient conditions for identification. We show that endogenous and exogenous effects are generally identified under network interaction, although identification may fail for some particular structures. Monte Carlo simulations provide an analysis of the effects of some crucial characteristics of a network (i.e., density, intransitivity) on the estimates of social effects. Our approach generalizes a number of previous results due to Manski (1993), Moffitt (2001), and Lee (2006).

Subfields: Education, Safety Research, Sociology and Political Science

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Education

In Safety Research

In Sociology and Political Science

Citations in the corpus

Criminal Networks: Who is the Key Player? (2012, methodology): “In particular, we highlight the di erence between the case with row-normalizedG (Bramoullé et al., 2009) and our case with non-row-normalized G. Endogenous network formation/correlated e ects Although this setting allows us to solve the reection problem, the estimation results might still be awed…”
NEIGHBORHOOD EFFECTS: ACCOMPLISHMENTS AND LOOKING BEYOND THEM* (2010, methodology): “New advanced on econometric techniques, like by Bramoullé et al. (2009) and others, offer new ways to take advantage of network structure in identifying decisions in network settings. They can be augmented by means of non-parametric techniques, like the ones proposed by Brock and Durlauf (2007), who”
Random Coefficients on Endogenous Variables in Simultaneous Equations Models (2014, methodology): “, Bramoullé et al. 2009 and Lin 2010). Card and Giuliano (2013) use this friendship data to construct pairs of best friends.”
Does Co‐authorship Lead to Higher Academic Productivity? (2015, methodology): “Bramoullé et al. (2009) propose the use of peers’ peers (and peers’ peers’ peers) characteristics as instrumental variables for the peers’ behaviour. Calvó-Armengol, Patacchini and Zenou (2009) and Lin (2010) use friendship networks of adolescents derived from the Add-Health data to identify peer an”
Distinguishing influence-based contagion from homophily-driven diffusion in dynamic networks (2009, positive): “(31)] has developed following seminal work by Manski (32) and Frank and Strauss (33), or models of the coevolution of networks and behaviors by Snijders (34), methods based on exogenous shocks to peers [e.”
Social Networks and the Diffusion of User-Generated Content: Evidence from YouTube (2012, methodology): “Thus, our method of identification is similar to recent approaches that measure social influence by identifying the nature of interactions structured through a network (e.g., Bramoulle et al. 2009).”
Social Networks, Reputation, and Commitment: Evidence From a Savings Monitors Experiment (2019, background): “42This multitude of channels is also the reason why we do not try to estimate and instrument a more structured model of spillovers in the spirit of Bramoulle et al. (2009).”
Matching and network effects (2010, methodology): “1− P ij = e−cb ij−1 = 1 1 + eXβ e ijbd ij−1 = 1 + e ′β cb ij−1 = log(1 + e β) ≈ eXβ (5)”
Do Peers Affect Student Achievement? Evidence from Canada Using Group Size Variation (2010, methodology): “Following Lee (2007), we estimate the model in 1More precisely, Manski studies a social interactions model which, in terms of identification, has the same properties as a model where individuals interact in groups and each individual is included in his peer group (see Bramoullé et al. 2009).”