CoChat
Sign Up
Cited by 144Hate Speech and Cyberbullying DetectionSocial and Intergroup PsychologyMisinformation and Its Impacts

The DOI matches its Crossref record; 14 of the 15 references it cites were found in a registry or an index.

Empathy-based counterspeech can reduce racist hate speech in a social media field experiment

Open access
Hangartner, Dominik, Gennaro, Gloria, Alasiri, Sary, Bahrich, Nicholas, Bornhoft, Alexandra, Boucher, Joseph, Demirci, Buket Buse, Derksen, Laurenz, Hall, Aldo, Jochum, Matthias, Munoz, Maria Murias, Richter, Marc, Vogel, Franziska, Wittwer, Salomé, Wüthrich, Felix, Gilardi, Fabrizio, Donnay, Karsten·2021·Proceedings of the National Academy of Sciences
DOI 10.1073/pnas.2116310118PMC PMC8685915
Open this paper in CoChat — free accountOpen PDF

Abstract

Despite heightened awareness of the detrimental impact of hate speech on social media platforms on affected communities and public discourse, there is little consensus on approaches to mitigate it. While content moderation—either by governments or social media companies—can curb online hostility, such policies may suppress valuable as well as illicit speech and might disperse rather than reduce hate speech. As an alternative strategy, an increasing number of international and nongovernmental organizations (I/NGOs) are employing counterspeech to confront and reduce online hate speech. Despite their growing popularity, there is scant experimental evidence on the effectiveness and design of counterspeech strategies (in the public domain). Modeling our interventions on current I/NGO practice, we randomly assign English-speaking Twitter users who have sent messages containing xenophobic (or racist) hate speech to one of three counterspeech strategies—empathy, warning of consequences, and humor—or a control group. Our intention-to-treat analysis of 1,350 Twitter users shows that empathy-based counterspeech messages can increase the retrospective deletion of xenophobic hate speech by 0.2 SD and reduce the prospective creation of xenophobic hate speech over a 4-wk follow-up period by 0.1 SD. We find, however, no consistent effects for strategies using humor or warning of consequences. Together, these results advance our understanding of the central role of empathy in reducing exclusionary behavior and inform the design of future counterspeech interventions.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1073/pnas.2116310118

Cites

14 of 14 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1. Examples of a human-controlled “bot” account (Left) and counterspeech messages (Right) using warning of consequences (green box), humor (red box), and empathy-based (blue box) strategies. The hate tweet and Twitter handles are obscured to protect users’ anonymity.
Fig. 1. Examples of a human-controlled “bot” account (Left) and counterspeech messages (Right) using warning of consequences (green box), humor (red box), and empathy-based (blue box) strategies. The hate tweet and Twitter handles are obscured to protect users’ anonymity.CC BY-NC-ND · PubMed Central
Fig. 2. Point estimates along with 90% and 95% CI from OLS regressions. Effects of the three counterspeech strategies on the “No. of Xenophobic Tweets,” “No. of Total Tweets,” and the ratio of the two measures (“Xenophobic Tweet Share”), over the 4-wk follow-up period (Left); the propensity that the user deleted the original xenophobic tweet (“Xenophobic Tweet Deleted”), the share of retrospectively deleted tweets (“Tweet Deletion Rate”) and the average “Vader Negativity” score over the 4-wk follow-up (Right). All regression outcomes are standardized to have mean = 0 and SD = 1. Mean and SD of the original, nonstandardized variables are provided in parentheses.
Fig. 2. Point estimates along with 90% and 95% CI from OLS regressions. Effects of the three counterspeech strategies on the “No. of Xenophobic Tweets,” “No. of Total Tweets,” and the ratio of the two measures (“Xenophobic Tweet Share”), over the 4-wk follow-up period (Left); the propensity that the user deleted the original xenophobic tweet (“Xenophobic Tweet Deleted”), the share of retrospectively deleted tweets (“Tweet Deletion Rate”) and the average “Vader Negativity” score over the 4-wk follow-up (Right). All regression outcomes are standardized to have mean = 0 and SD = 1. Mean and SD of the original, nonstandardized variables are provided in parentheses.CC BY-NC-ND · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 15 references

Research works printed but not linked: 2

  1. A. Siegel, Online Hate Speech in Social Media and Democracy: The State of the Field, Prospects for Reform (Cambridge University Press, 2020).
  2. I. Gagliardone, D. Gal, T. Alves, G. Martinez, Countering Online Hate Speech (United Nations Educational, Scientific and Cultural Organization, 2015).

Cited by

Sorted by citations

See all 144

Related

Metadata

Authors
Hangartner, D., Gennaro, G., Alasiri, S., Bahrich, N., Bornhoft, A., Boucher, J., Demirci, B. B., Derksen, L., Hall, A., Jochum, M., Munoz, M. M., Richter, M., Vogel, F., Wittwer, S., Wüthrich, F., Gilardi, F., & Donnay, K.
Affiliations
  • ETH Zurich (CH)
  • Center for Comparative and International Studies, Eidgenössische Technische Hochschule Zurich, 8092 Zurich, Switzerland;
  • University of Zurich (CH)
Corresponding author
Hangartner, Dominik
ORCID
Venue
Proceedings of the National Academy of Sciences
Publisher
National Academy of Sciences
Published
2021-12-06
Volume
118
Issue
50
ISSN
  • 0027-8424 (ISSN-L)
  • 1091-6490
DOI
10.1073/pnas.2116310118
PMC
PMC8685915
OpenAlex
W4200232392
Cited by
144
References
15
Citations by year
2026: 20 · 2025: 44 · 2024: 42 · 2023: 23 · 2022: 13 · 2021: 2
FWCI
7.07
Type
journal article
Language
en
Open access
Licence
Locations
  • Proceedings of the National Academy of Sciences (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • Zurich Open Repository and Archive (University of Zurich) (acceptedVersion, cc-by-nc-nd)
  • Repository for Publications and Research Data (ETH Zurich) (submittedVersion, cc-by-nc-nd)
  • Zurich Open Repository and Archive (University of Zurich) (submittedVersion, cc-by-nc-nd)
  • ETH Zürich Research Collection
  • Zurich Open Repository and Archive (University of Zurich)
Funders
  • InnoSuisse (46165.1 IP-SBM)
  • Swiss Federal Office of Communications (NA)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
  • Social Sciences › Sociology and Political Science › Social and Intergroup Psychology
  • Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
Keywords
counterspeech, hate speech, racism, xenophobia, empathy, humor, social media, Twitter, hate speech mitigation, online harassment
MeSH terms
Empathy, Hate, Humans, Language, Social Media, Racism

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review