CoChat
Sign Up
Cited by 5Stress and Burnout ResearchWorkplace Health and Well-beingEmotional Labor in Professions

The DOI matches its Crossref record, with a small difference in title or year; 10 of the 10 references it cites were found in a registry or an index.

Perceptions and Experiences of Severe Content in Content Moderation Teams: A Qualitative Study

Open access
Strongylou, Dimitra Eleftheria, Savio, Marlyn Thomas, Steiger, Miriah, Bharucha, Timir, Manuel, Wilfredo R. Torralba, Huang, Xieyining, Guevara, Rachel Lutz·2024·Lecture Notes in Networks and Systems
DOI 10.1007/978-981-97-4581-4_1
Open this paper in CoChat — free accountOpen PDF

Abstract

Existing Trust & Safety policies predominantly focus on protecting content moderators’ (CoMos) safety against severe content violations. Nevertheless, in reality, CoMos might encounter content across a broad severity spectrum while perceptions of content severity among CoMos are yet to be fully understood in relation to their well-being and job accuracy. The current study employs a sequential study design to qualitatively examine the views of 34 CoMos in the USA and the Philippines and of 166 professionals (i.e., mental health and learning experience staff, team leads) supporting CoMos on content severity and its perceived impact on CoMos well-being and job accuracy. Of note, consistent with ethical guidelines and wellness best practices, all CoMos received tiered wellness support services from TaskUs Inc. Collected data were thematically analyzed and two overarching themes emerged. Within the first theme, namely, ‘perceived content categories’, both CoMos and professionals, after considering the potential distress caused by various contents, broadly distinguished between ‘moderate’ and ‘graphic’ content, while they further attributed their own definitions to each of the two categories. Under theme two—‘content impact on perceived stress- most study participants discussed how reviewing moderate content might have a more detrimental impact upon CoMos stress and job accuracy compared to reviewing graphic content in the lack of wellness support and care. This study is the first of its kind to give voice to both CoMos and support staff to discuss in depth their views on content moderation while also making participant-led recommendations for work policy and future research.

Overview

Integrity signals

Registry

Crossref · 10.1007/978-981-97-4581-4_1

Cites

10 of 10 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

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?

  • Perceived stress in this role arises from general work demands (e.g., workload, job demands, personal factors) coupled with exposure to disturbing content (45).

    “Perceived stress in this role arises from general work demands (e.g., workload, job demands, personal factors) coupled with the exposure to disturbing content (45).”

    From Longitudinal comparison of psychological outcomes of professional content moderators engaged in resilience training program (2025) · cites this paper as [45]

  • Content reviewers may not realize they are experiencing mental health issues, as such problems can manifest gradually.

    “Many content reviewers may not realize they are experiencing mental health issues, as such problems can manifest gradually [178] and reviewers may become accustomed to feelings of anxiety, stress, and unhappiness at work and view these emotions as common [32,168,178].”

    From Measuring the Mental Health of Content Reviewers, a Systematic Review (2025) · cites this paper as [178]

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 3

  1. Morrow G, Swire-Thompson B, Polny JM, Kopec M, Wihbey JP (2022) The emerging science of content labeling: contextualizing social media content moderation. J Am Soc Inf Sci 73(10):1365–1386
  2. Karunakaran S, Ramakrishan R (2019) Testing stylistic interventions to reduce emotional impact of content moderation workers. Proceed AAAI Conferen Hum Comput Crowdsourc 7(1):50–58
  3. Scheuerman MK, Jiang JA, Fiesler C, Brubaker JR (2021) A framework of severity for harmful content online. Proceed ACM Hum Comput Interact 5(CSCW2):1–33

Cited by

Sorted by citations

Related

Metadata

Authors
Strongylou, D. E., Savio, M. T., Steiger, M., Bharucha, T., Manuel, W. R. T., Huang, X., & Guevara, R. L.
Corresponding author
Strongylou, Dimitra Eleftheria
ORCID
Venue
Lecture Notes in Networks and Systems
Publisher
Springer Nature Singapore
Published
2024
Pages
1-12
ISSN
  • 2367-3370 (ISSN-L)
  • 2367-3389
DOI
10.1007/978-981-97-4581-4_1
OpenAlex
W4401723289
Cited by
4
References
9
Citations by year
2026: 2 · 2025: 2
FWCI
2.69
Type
conference paper
Language
en
Open access
Licence
Locations
Lecture notes in networks and systems (publishedVersion, cc-by)
Topics
  • Social Sciences › Psychology › Social Psychology › Stress and Burnout Research
  • Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Workplace Health and Well-being
  • Social Sciences › Sociology and Political Science › Emotional Labor in Professions
Keywords
content moderation, graphic content, stress, well-being

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