MétaCan
Menu
Back to cohort
Record W4410940256 · doi:10.2196/70881

Comparing Media and Law Enforcement Reports on Anti-Asian Hate Incidents During the COVID-19 Pandemic: Data Visualization Approach

2025· article· en· W4410940256 on OpenAlexvenueno aff
Young Ji Yoon, Su Hyun Shin, Dongwook Kim, Hee Yun Lee

Bibliographic record

VenueJMIR Formative Research · 2025
Typearticle
Languageen
FieldComputer Science
TopicHate Speech and Cyberbullying Detection
Canadian institutionsnot available
Fundersnot available
KeywordsPreprintCoronavirus disease 2019 (COVID-19)Law enforcement2019-20 coronavirus outbreakSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2)Political scienceVisualizationCriminologyLawSociologyComputer scienceVirologyWorld Wide WebData miningMedicineInfectious disease (medical specialty)

Abstract

fetched live from OpenAlex

BACKGROUND: During the COVID-19 pandemic, anti-Asian hate incidents (AAHIs) increased conspicuously. Literature reports discrepancies in how crimes are reported differently in media and law enforcement data, emphasizing potential biases and inconsistencies in AAHI reporting. Understanding the discrepancies in AAHI reporting between the two sources is crucial for improving documentation procedures and addressing systemic issues in reporting mechanisms. OBJECTIVE: This study aimed to (1) present the monthly trends in AAHI counts reported by media and law enforcement data from 2020 to 2021, (2) investigate variations in AAHI counts across states and counties for each year, (3) examine discrepancies in AAHI reporting between the two sources at state and county levels during the same period, and (4) delineate differences in the types and geographic distribution of incidents as represented by the two sources. METHODS: This study used two data sources for AAHIs, media data (n=1288) from The Asian American Foundation and law enforcement data (n=1086) from the Federal Bureau of Investigation, for the 2020-2021 period. Descriptive analyses were conducted to evaluate monthly trends, state and county-level variations, and differences in incident types and locations. Ratios of reported incidents between the two sources were calculated to assess discrepancies. Temporal trends were contextualized within key sociopolitical events to offer insights into reporting dynamics. RESULTS: First, both media and law enforcement data presented a sharp increase in reported AAHIs following the first confirmed COVID-19 case in the United States, peaking around March 2020, coinciding with controversial political rhetoric. A second peak occurred from March to April 2021, immediately following the pandemic's peak, and was followed by a decline as the situation improved. Second, in the two data sources, the state-level analysis indicated that California, Texas, New York, and Washington consistently reported the highest AAHI counts. In 2021, there were notable increases in reported incidents in states such as Wisconsin and Illinois. County-level data revealed persistent high counts in California, particularly in Los Angeles County. Ratios of AAHI counts between the two data sources presented significant discrepancies, with higher ratios in California and New York. Finally, the analysis of incident types revealed that media data reported a higher proportion of harassment (477/1288, 37%), while the law enforcement data reported more property-related incidents (239/1086, 22%). Regarding location types, media data frequently reported incidents in public areas (515/1288, 40%) and businesses (361/1288, 28%), whereas law enforcement data reported more incidents occurring in residential settings (201/1086, 18.5%). CONCLUSIONS: This study highlighted significant trends and disparities in AAHI reporting between media and law enforcement data, underscoring the need for a nuanced understanding of how these incidents were reported. Practice and policy implications suggested fostering community engagement to support Asian communities while enhancing the accuracy and consistency of hate crime reporting.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.714
Threshold uncertainty score0.922

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0030.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0010.000
Scholarly communication0.0010.001
Open science0.0010.002
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.133
GPT teacher head0.423
Teacher spread0.289 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations1
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJMIR Formative ResearchSame topicHate Speech and Cyberbullying DetectionFrench-language works237,207