Embedded Files
Working papers:
Identification of a Rank-dependent Peer Effect Model (With Myungkou Shin)
Abstract: We develop a model that captures peer effect heterogeneity by modeling the endogenous spillover to be linear in ordered peer outcomes. Unlike the canonical linear-in-means model, our approach accounts for the distribution of peer outcomes as well as the size of peer groups. Under a minimal condition, our model admits a unique equilibrium and is therefore tractable and identified. Simulations show our estimator has good finite sample performance. Finally, we apply our model to educational data from Norway, finding that higher-performing friends disproportionately drive GPA spillovers. Our framework provides new insights into the structure of peer effects beyond aggregate measures.
Identifying Gender Bias in Grading (With Ragnar Alne)
Abstract: This paper shows how to identify gender bias for equally skilled students. Our identification result relies on the existence of a correctly timed blind exam, and only require an assumption on the blindness of said exam. The results are simple to implement and can be applied to sub-groups based on skills and socioeconomic characteristics. Applying our results to detailed registry data, we show that gender bias is highly heterogeneous and is strongest for students with low-skill and low-socioeconomic background. These biases have significant effects on students later life outcomes, such as attending higher-education and labor market performance.
Fairness in grading: Randomizing Ethnicity and Gender in Teacher Assessments (With Ragnar Alne and Rune Borgan Reiling)
Related policy report using the results and media coverage [1] and [2] (in Norwegian)
Abstract: We investigate the impact of ethnicity and gender on teacher assessments through a randomized controlled experiment involving 203 teachers grading a single class's assignments. Within teachers we randomly assigned names, signaling gender and ethnicity, with a control group grading anonymized assignments. Our results reveal significant variation in grades assigned by teachers for identical assignments, with some evidence of both ethnic and gender biases. Additionally, we examine how teachers' implicit biases interact with grading biases. Through counterfactual policy simulations, we demonstrate that our estimates imply an economically relevant trade-off between precision and accuracy for grading policies in schools.
How Socioeconomic and Parental Background Shape Peer Networks and Educational Spillovers (With Ragnar Alne and Andreas Myhre)
Abstract: This paper examines how socioeconomic background and student characteristics influence friendship formation and educational outcomes. We take advantage of the combination of survey data combined with a rich set of registry data to observe both student friendships and detailed information on parental and socioeconomic backgrounds. We find significant effects of parental background-specifically age, ethnic background, and social security status-on student friendship formation. Parental income also plays a role, though we find no significant effects of parental wealth. The strongest determinants of friendship formation are shared gender and class membership, along with evidence of assortative matching based on academic skills. To estimate peer effects, we instrument for friends' academic performance using pre-existing skill measures. Our results indicate substantial spillovers: a one-standard-deviation increase in friends' GPA leads to approximately a 0.62 standard deviation increase in a student's own GPA. We leverage these findings to assess how classroom structure shapes academic outcomes through its influence on student friendships. We demonstrate that the realized social network significantly impacts individual achievement, suggesting that policies targeting peer interactions could be an effective tool for improving student outcomes.
Published papers:
Applying Contrastive Learning to an Attention Neural Model in a Multilingual Context
25th International Conference on Web Engineering (ICWE 2025) (With Phillip Gottschalk and Flavius Frasincar)
"Residential Investment and Recession Predictability"
International Journal of Forecasting (2019) (With André K. Anundsen and Knut Are Aastveit)
Selected Work in Progress:
Estimating peer effects and Network formation models with missing network links (Currently being restructured)
Abstract: Estimates of peer effects may suffer from bias if the network data has missing links. Moreover, if links are not missing at random, estimates of parameters in network formation models may also be biased. I contribute to the literature on identification of peer effects and network-formation models with partially sampled network data in three ways. My first contribution is to develop a consistent peer-effects estimator that uses link-formation probabilities from the true network-formation model. My second contribution is to develop two estimators of a network formation model that are robust to the missingness of links correlating with unobservable link-specific shocks. These are an inverse-probability-weighted likelihood estimator that uses the probabilities of observing links as weights, and a semi-parametric estimator. The first estimator requires the researcher to estimate the probability of a link being observed, while the second does not at the cost of a slower convergence rate. My third contribution is to show sharp partial identification of endogenous peer effects when there is no information on how the network is formed. The bounds from this exercise will be more informative if the researcher has information about the unobserved network. I apply my peer-effect estimator to a new dataset from two Norwegian schools that I merge with administrative data. In this data, I observe the complete network as well as a partial sample of links constructed by restricting students to only list some of their friends. Using the complete and partial sample of network links from the data, I find that my method reduces the bias in the peer-effect estimator by 65%. Finally, I demonstrate that naive estimators can lead to misleading results about household behavior in microfinance take-up by applying my peer-effect estimator to the dataset of Banerjee et al. (2013).
Economics of Scientific Research: Funding, Structure, and Outcomes (Project with Bruce Weinberg and Donna Ginther, received a million dollar grant from the Sloan Foundation).
An Empirical evaluation of trade-offs in hiring: Education vs. Experience (With Ragnar Alne and Christine Blandhol)
Inference for the solutions to Linear Programs (With Jonas Lieber)
Page updated
Google Sites
Report abuse