Research

Working papers

[1] Exposure to better-performing peers: consequences and mechanisms
The existing literature on peer effects in education suggests that better peers can positively or negatively impact other students’ outcomes. However, the potential mechanisms leading to negative peer effects are not well understood. This study empirically investigates the effects of peer performance on a comprehensive set of outcomes, including short-term cognitive and non- cognitive abilities, as well as long-term educational achievement, using a sample of American high school students. I show that exposure to high-performing peers negatively impacts cognitive abilities, self-discipline, and the likelihood of college graduation. The common underlying driver of these results is performance competition, which fosters the development of a negative self-perception about one’s abilities relative to peers. To support this key finding, I propose a novel network game that incorporates the notion of self-perception. The theoretical predictions of this model provide a framework for interpreting a broader range of empirical results, including potentially negative ability peer effects.  
(WP Draft)

[2] Inside out: Including self-perception in peer effect modeling
Theoretical work on peer effects builds on the premise of positive peer effects. Yet, this assumption does not always support empirical findings that contradict it or document heterogeneous peer effects. I address this gap by proposing a novel model of social interactions that formalizes the mechanism of social comparison. The proposed model is a modified version of the local-average model and integrates the notion of self-perception. The core of this model is the outcome production function, which is a concavely increasing function of one’s own productivity and effort and the product of self-perceived ability and peer effort. Five main results are established in this study. First, a unique (but not alway interior) Nash equilibrium exists and is a weighted average of productivities, where the weights are a nonlinear function of the network structure, the degree of social comparison, and the perceived abilities. Second, peer productivity does not always positively affect individual equilibrium effort, and the direction of this effect strictly depends on one’s own and peer self-perception. Third, the overall distribution of self-perception in the network affects everyone’s effort, and both the network density and the degree of social comparison can exacerbate this effect, regardless of its direction. Furthermore, the presence of individuals with low self-concepts is not unconditionally harmful (only above a certain threshold), but the larger their proportion in the network, the lower the average effort in the network. This study is the first formalization of social comparison through a microeconomic model. Its conceptual differences from the existing models of social interactions has also significant implications for policy interventions. (WP Draft)

[3] Misclassification in Linear-in-Means Models: Theory and Application to Peer Effects Estimation (with Simone Balestra, and Beatrix Eugster) –Under Review
This paper investigates, both theoretically and empirically, the consequences of misclassification in an linear-in-means (LIM) model. We build the theoretical analysis on a simple form of an LIM model—including only an individual characteristic and its groupwise average—and demonstrate that under random group formation and nondifferential measurement error, the peer effect is biased by an “own” and a “smearing effect.” As the number of groups tends to infinity, the smearing effect approaches zero with almost probability one, while the own effect turns into a simple attenuation bias that is proportional to the misclassification rates. Applying the theoretical results to the estimation of the peer effect of students with learning disabilities on other students’ performance, we show that the results are in line with the theoretical predictions as long as the considered misclassified variables exclusively capture learning disabilities. (WP Draft, CEPR WP) 

Work in progress

[1] Discriminatory Popular Initiatives and Migrant Assimilation (with Massimo Troncone, and Roberto Valli). Data: Swiss admin data from the Swiss Federal Statistical Office. Status: Drafting. 

[2] Insight: IN-School Individualized coGnitive beHevioral Therapy. A field experiment in El Salvador (with Martina Jakob, and Miriam Prater). Status: Grant applications in progress. 

Pending

[1] Neighbourhood effects in adolescence on cultural assimilation (with Massimo Troncone, and Roberto Valli). Status: Idea formalised and data collected. Data: Swiss admin data from the Swiss Federal Statistical Office.