Claudia Marangon
Claudia Marangon

Postdoctoral Fellow

Harvard University

Welcome to my website! I am a Postdoctoral Fellow at Harvard University, funded by the Swiss National Science Foundation. I am also a Research Affiliate at the Group for Law, Economics and Data Science and the Public Policy Group at ETH Zurich. I obtained my Ph.D. in Economics at ETH Zurich in May 2025 and was Visiting Fellow at the Harvard Kennedy School of Government in Spring 2023.

My research lies at the intersection of Media Economics and Economics of Crime, with a focus on the role of gender and racial identity. In particular, I leverage techniques in machine learning and natural language processing to shed light on the effect of identity among professionals, like journalists or judges.

I also co-organize the Online Seminar in Economics + Data Science. You can join our mailing list to be up to date with our events and contact me directly if you are interested in presenting!

CV
Interests
  • Political Economy
  • Media Economics
  • Economics of Crime
  • Natural Language Processing
Education
  • Ph.D. Candidate in Economics, 2019-2025

    ETH Zurich

  • Visiting Fellow, Spring 2023

    Harvard Kennedy School of Government

  • M.Sc. in Economic and Social Sciences, 2016-2019

    Bocconi University

  • B.A. in Economics and Social Sciences, 2013-2016

    Bocconi University

Research

Working Papers

Work in Progress

Identity in Journalism: Evidence from News Reporting of Violence Against Women.
Abstract
This paper investigates whether reporter identity shapes news content and reader behavior. Using data on the universe of murders with female victims in Italy from 2006 to 2022, matched with associated newspaper articles, I measure gender differences in reporting. I construct novel, validated narrative indices capturing two opposing reporting styles: victim-centered narratives, which focus on the victim and provide information on support resources, and perpetrator-centered narratives, which emphasize the perpetrator’s motives, like jealousy or a fit of rage. To adjust for variation in the stories and topics covered, I exploit the conditionally random assignment of journalists to crime events based on newsroom schedules. The results show substantial gender differences in reporting, with female journalists more likely to use victim-centered narratives and less likely to use perpetrator-centered narratives. Combining reporter assignment with provincial readership data in a staggered difference-in-differences design, I show that higher exposure to articles by female reporters after a murder increases calls to the helpline number for violence against women. I provide experimental evidence to support the victim-centered narratives as a key mechanism: victim-centered framing increases the perceived prevalence of help-seeking behavior, shifting norms around seeking support. By contrast, perpetrator-centered narratives lead to higher engagement among male readers while generating backlash among female readers, suggesting that identity-driven and audience-maximizing motives may reinforce one another.
Visual Justice: The Effect of Media Mugshots on Attitudes and Judicial Outcomes.
with Elliott Ash, Sergio Galletta, and Benjamin Kohler
Gender Differences in Judging.
with Clémentine Abed Meraim, Elliott Ash, and Talia Gillis

Conference Publications

WCLD: Curated Large Dataset of Criminal Cases from Wisconsin Circuit Courts. Proceedings of the 37th Conference on Neural Information Processing Systems (NEURIPS). 2023
with Elliott Ash, Naman Goel, Nianyun Li, and Peiyao Sun
Teaching
Data Science for Public Policy: From Econometrics to AI — ETH Zurich (Spring 2025)
Lecturer for graduate-level course connecting econometric methods with modern AI applications in public policy.
Text Data in Economics — University of Basel (Fall 2023, Fall 2024)
Lecturer for a PhD-level course on natural language processing methods applied to Economics
Building a Robot Judge: Data Science for Decision Making — ETH Zurich (Fall 2020–2023)
Teaching assistant for graduate-level course on causal inference and machine learning methods for decision-making
NLP for Law and Political Economy — ETH Zurich (Spring 2020–2022)
Teaching assistant for graduate-level course on natural language processing methods applied to Law and Political Economy