Identify Peer Effects Research Articles

Heterogeneous peer effects of college roommates on academic performance

Understanding how student peers influence learning outcomes is crucial for effective education management in complex social systems. The complexities of peer selection and evolving peer relationships, however, pose challenges for identifying peer effects using static observational data. Here we use both null-model and regression approaches to examine peer effects using longitudinal data from 5,272 undergraduates, where roommate assignments are plausibly random upon enrollment and roommate relationships persist until graduation. Specifically, we construct a roommate null model by randomly shuffling students among dorm rooms and introduce an assimilation metric to quantify similarities in roommate academic performance. We find significantly larger assimilation in actual data than in the roommate null model, suggesting roommate peer effects, whereby roommates have more similar performance than expected by chance alone. Moreover, assimilation exhibits an overall increasing trend over time, suggesting that peer effects become stronger the longer roommates live together. Our regression analysis further reveals the moderating role of peer heterogeneity. In particular, when roommates perform similarly, the positive relationship between a student’s future performance and their roommates’ average prior performance is more pronounced, and their ordinal rank in the dorm room has an independent effect. Our findings contribute to understanding the role of college roommates in influencing student academic performance.

Identifying Peer Effects in Networks with Unobserved Effort and Isolated Students

Identifying Peer Effects in Networks with Unobserved Effort and Isolated Students

Efficient peer effects estimators with group effects

We study linear peer effect models where peers interact in groups and individual’s outcomes are linear in the group mean outcome and characteristics. We allow for unobserved random group effects as well as observed fixed group effects. The specification is in part motivated by the moment conditions imposed in Graham (2008). We show that these moment conditions can be cast in terms of a linear random group effects model and that they lead to a class of GMM estimators with parameters generally identified as long as there is sufficient variation in group size or group types. We also show that our class of GMM estimators contains a Quasi Maximum Likelihood estimator (QMLE) for the random group effects model, as well as the Wald estimator of Graham (2008) and the within estimator of Lee (2007) as special cases. Our identification results extend insights in Graham (2008) that show how assumptions about random group effects, variation in group size and certain forms of heteroscedasticity can be used to overcome the reflection problem in identifying peer effects. Our QMLE and GMM estimators accommodate additional covariates and are valid in situations with a large but finite number of different group sizes or types. Because our estimators are general moment based procedures, using instruments other than binary group indicators in estimation is straight forward. Our QMLE estimator accommodates group level covariates in the spirit of Mundlak and Chamberlain and offers an alternative to fixed effects specifications. This model feature significantly extends the applicability of Graham’s identification strategy to situations where group assignment may not be random but correlation of group level effects with peer effects can be controlled for with observable group level characteristics. Monte-Carlo simulations show that the bias of the QMLE estimator decreases with the number of groups and the variation in group size, and increases with group size. We also prove the consistency and asymptotic normality of the estimator under reasonable assumptions.

Peer effects on corporate R&D investment policies: A spatial panel model approach

Identifying peer effects has important policy implications because endogenous peer effects can amplify policy effects. Using a spatial panel model with network-structure peer groups and data on listed manufacturing firms in China, we investigate whether a firm’s R&D expenditure affects its peers’ R&D policies. Our results show: (1) peer firms’ R&D expenditures play an essential role in determining corporate R&D investment policies; (2) competition and information incentives are the two main underlying channels; and (3) R&D subsidies can indirectly increase a firm’s R&D via the endogenous effect. Using a novel identification strategy, this study provides additional empirical evidence of peer effects on corporate policies. The findings contribute to our understanding of the impacts of R&D subsidies and indicate that the overall policy effect can be leveraged by considering firms’ networks.

Identifying Peer Effects on Student Academic Effort

Identifying Peer Effects on Student Academic Effort

Peer correlations in income: Evidence from a Guanxi network in rural China

We investigate peer correlations in income using a novel panel of data from a village in China. We construct a complete and heterogeneous social connections (guanxi) network that includes all households, and use spatial-network econometric models to estimate peer correlations. The results provide unique empirical evidence for the existence of a peer correlation in income, even after netting out influences of family characteristics, contextual effects, and various unobservable factors. We identify peer effects in land and labor allocations as the drivers of peer correlations in income.

Thy Neighbor’s Vote: Peer Effects in Proxy Voting

Institutional investors’ proxy voting decisions are influenced by their neighbors. I identify peer effects in proxy voting using close-call votes on shareholder-sponsored governance proposals at publicly traded financial institutions (focal institutions). I first show that the passage of a governance proposal at a focal institution makes the institution more likely to vote against management at its portfolio firms. Using a triple-difference approach, I find that the neighboring institutions of the focal institution that passes a governance proposal become more likely to vote against management in stocks that are heavily held by the focal institution. These results suggest that peer influence is an important determinant of proxy voting behavior. This paper was accepted by David Sraer, finance. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2022.4517 .

Are You Listening?’: Sentiment Diffusion in GitHub

Do expressed sentiments affect remote-workers? My paper addresses this question in the important context of open innovation, using extensive online communication data on GitHub. I use the VADER sentiment analyzer to determine latent sentiment in open source software developer communication, and study the correlation of individual sentiments within developer teams. I employ an instrumental variable identification strategy to identify peer effects in sentiment diffusion. I find significant, positive correlation within in-group sentiments, which suggests potential channels through which sentiments can propagate. I also find sentiments expressed by an individual does not affect the entire team, and this suggests expressed sentiments tend to attenuate rather than magnify via propagation. These findings provide useful lessons to managers of remote working teams regarding ways to manage employee morale and team cohesion.

Let’s Stick Together: Peer Effects in Secondary School Choice and Variations by Student Socio-Economic Background

AbstractDespite the vast body of research focusing on peer effects in education, the role of the immediate peer environment in school choice has been understudied to date. We study the extent to which students from the same primary school cluster in the same secondary school, and how this effect varies by a student’s socio-economic background. We use register data from the Netherlands, covering six cohorts of students (2013–2019), that enable us to account for selection into primary schools and other endogeneity issues when identifying peer effects. The results indicate that students are more likely to choose a secondary school when students from their primary school cohort also choose this school, even after accounting for school popularity trends. We find evidence that students from socially disadvantaged backgrounds are more likely to cluster in the same secondary school as their primary school peers, yet these differences are small.

Peer Effects in Product Adoption

We use de-identified data from Facebook to study the nature of peer effects in the market for cell phones. To identify peer effects, we exploit variation in friends' new phone acquisitions resulting from random phone losses. A new phone purchase by a friend has a large and persistent effect on an individual's own demand for phones of the same brand. While peer effects increase the overall demand for phones, a friend's purchase of a particular phone brand can reduce an individual's own demand for phones from competing brands, in particular if they are running on a different operating system. (JEL C45, D12, L63, M31, Z13)

The importance of high performing team members in complex team work: Results from quasi‐experiments in professional team sports

AbstractThis paper empirically assesses peer effects of high performing team members in reciprocally interdependent team production. Using data from the National Basketball Association (NBA), we identify peer effects by exploiting unforeseen in‐game injuries of high performing players. Results indicate that without a high performing player, other team members maintain efficiency and the division of tasks among each other. However, they slightly increase risk taking and decrease their individual output, resulting in an inferior overall outcome. These effects depend on whether a high performing player has a team‐oriented or a self‐oriented role in the team. Additionally, we observe that relatively skilled players try to step in for the absent high performing player.

Distinguishing Homophily from Peer Influence Through Network Representation Learning

Peer influence and homophily are two entangled forces underlying social influences. However, distinguishing homophily from peer influence is difficult, particularly when there is latent homophily caused by unobservable features. This paper proposes a novel data-driven framework that combines the advantages of latent homophily identification and causal inference. Specifically, the approach first utilizes scalable network representation learning algorithms to obtain node embeddings, which are extracted from social network structures. Then, the embeddings are used to control latent homophily in a quasi-experimental design for causal inference. The simulation experiments show that the proposed approach can estimate peer influence more accurately than existing parameterized approaches and data-driven methods. We applied the proposed framework in an empirical study of players’ online gaming behaviors. First, our approach can achieve improved model fitness for estimating peer influence in online games. Second, we discover a heterogeneous effect of peer influence: players with higher tenure and playing levels receive stronger peer influence. Finally, our results suggest that the homophily effect has a stronger influence on players’ behavior than peer influence. Summary of Contribution: The study proposes a novel computational method to separate peer influence from homophily in an online network. Using network embeddings learned from data to control latent homophily, the approach effectively addresses the challenge of correctly identifying peer effects in the absence of randomized experimental conditions. While simplifying the computational process, the method achieves good computational performance, thus effectively helping researchers and practitioners extract useful network information in various online service contexts.

Information and credible sanctions in curbing online cheating among undergraduates: A field experiment

The rapid increase in online instruction in higher education has heightened concerns about cheating. We use a randomized control design to test whether informing students that we can detect plagiarism reduces cheating. We further test whether informing students they have been caught cheating reduces subsequent cheating. We find informing students about our capability to detect plagiarism has little effect on cheating. Notifying students that they have been caught cheating and are on a watch list reduces subsequent cheating attempts by at least 65 percent depending on the class and sample. We test for peer effects but conclude we cannot credibly identify peer effects distinct from own-cheating propensities.

Identifying and decomposing peer effects on decision-making using a randomized controlled trial

Utilizing a randomized controlled trial (RCT) in traditional clusters of apparel and textile firms in Vietnam, this paper investigates peer effects on firm managers’ decisions to participate in seminars on export promotion. We invited 131 randomly selected firm representatives to three one-day seminars on export promotion. We use the number of randomly invited peers to identify peer effects. We further decompose the invited peers into peers invited to the same seminar, those invited to the earlier seminars, and those invited to the later seminars. We find that the former has a positive effect on firms' participation, whereas the latter two have no significant effect. These results imply that peer effects on participation primarily arise from the benefits of face-to-face interactions. The presence of positive peer effects suggests that multiple equilibria in terms of the share of participants within each village of firms may emerge, which is also consistent with our observations.

Bias and High-Dimensional Adjustment in Observational Studies of Peer Effects

Peer effects, in which an individual’s behavior is affected by peers’ behavior, are posited by multiple theories in the social sciences. Randomized field experiments that identify peer effects, however, are often expensive or infeasible, so many studies of peer effects use observational data, which is expected to suffer from confounding. Here we show, in the context of information and media diffusion, that high-dimensional adjustment of a nonexperimental control group (660 million observations) using propensity score models produces estimates of peer effects statistically indistinguishable from those using a large randomized experiment (215 million observations). Compared with the experiment, naive observational estimators overstate peer effects by over 300% and commonly available variables (e.g., demographics) offer little bias reduction. Adjusting for a measure of prior behaviors closely related to the focal behavior reduces this bias by 91%, while models adjusting for over 3700 past behaviors provide additional bias reduction, reducing bias by over 97%, which is statistically indistinguishable from unbiasedness. This demonstrates how detailed records of behavior can improve studies of social influence, information diffusion, and imitation; these results are encouraging for the credibility of some studies but also cautionary for studies of peer effects in rare or new behaviors. More generally, these results show how large, high-dimensional datasets and statistical learning can be used to improve causal inference. Supplementary materials for this article are available online.

Sibling Spillovers

AbstractIt is notoriously difficult to identify peer effects within the family. Using administrative data on children from both Florida and Denmark, the paper examines the effects of having a disabled younger sibling. To address the identification challenge, the paper compares the differential effects for first- and second-born children in three-plus-child families, taking advantage of the fact that birth order influences the amount of time that a child spends in early childhood with their younger siblings, disabled or not. The paper finds evidence that, relative to the first born, the second child in a family is differentially affected when the third child is disabled.

GMM identification and estimation of peer effects in a system of simultaneous equations

This paper considers the identification and estimation of network models with agents interacting in multiple activities. We establish the model identification using both linear and quadratic moment conditions. The quadratic moment conditions exploit the correlation of individual decisions within and across different activities, and provide an additional channel to identify peer effects. Combining linear and quadratic moment conditions, we propose a general GMM framework for the estimation of simultaneous equations network models. The GMM estimator improves the asymptotic efficiency of the existing IV-based linear estimators in the literature. Simulation experiments show that the GMM estimator performs well in finite samples.

Thy Neighbor's Vote: Peer Effects in Proxy Voting

Institutional investors' proxy voting decisions are influenced by their neighbors. I identify peer effects in proxy voting using close-call votes on shareholder-sponsored governance proposals at publicly traded financial institutions (focal institutions). I first show that the passage of a close vote at a focal institution increases the likelihood that the institution votes against management. Using a triple-difference approach, I find that the passage increases the likelihood that the neighboring institutions vote against management in stocks that are heavily held by the focal institution. These results suggest that peer influence is an important determinant of proxy voting behavior.

Peer Effects in Legislative Voting

We exploit seating rules in the European Parliament to identify peer effects in legislative voting. Sitting adjacently leads to a 7 percent reduction in the overall likelihood that two members of the European Parliament (MEPs) from the same party differ in their vote. Peer effects are markedly stronger among pairs of women, MEP pairs from the same country, and in close votes. Using variation in seating across the parliament's two venues (Brussels and Strasbourg), we show that peer effects are persistent: MEPs who have sat together in the past disagree less even when they are not seated adjacently. (JEL D72, Z13)

Real Option Exercise: Empirical Evidence

Abstract We study when and why firms exercise real options. Using detailed project-level investment data, we find that the likelihood that a firm exercises a real option is strongly related to peer exercise behavior. Peer exercise decisions are as important in explaining exercise behavior as variables commonly associated with standard real option theories, such as volatility. We identify peer effects using localized exogenous variation in peer project exercise decisions and find evidence consistent with information externalities being important for exercise behavior. (JEL G30, G31, G32)