Abusers indoors and coronavirus outside: an examination of public discourse about COVID-19 and family violence on Twitter using machine learning
Bibliographic record
Abstract
Abstract Purpose This brief report aims to provide the first large-scale analysis of public discourse regarding family violence and the COVID-19 pandemic on Twitter. Method: We analyzed 301,606 Tweets related to family violence and COVID-19 from April 12 to July 16, 2020, for this study. We used the machine learning approach, Latent Dirichlet Allocation, and identified salient themes, topics, and representative Twitter examples. Results We extracted nine themes on family violence and COVID-19 pandemic, including (1) the Impact of COVID-19 on family violence (e.g., rising rates, hotline calls increased, murder & homicide); (2) the types (e.g., child abuse, domestic violence, sexual violence) and (3) forms of family violence (e.g., physical aggression, coercive control); (4) risk factors of family violence (e.g., alcohol abuse, financial constraints, gun, quarantine); (5) victims of family violence (e.g., LGBTQ, women, and women of color, children); (6) social services of family violence (e.g., hotlines, social workers, confidential services, shelters, funding); (7) law enforcement response (e.g., 911 calls, police arrest, protective orders, abuse reports); (8) Social movement/awareness (e.g., support victims, raise awareness); and (9) domestic violence-related news (e.g., Tara Reade, Melissa Derosa). Conclusions The COVID-19 has an impact on family violence. This report overcomes the limitation of existing scholarship that lacks data for consequences of COVID-19 on family violence. We contribute to the understanding of family violence during the pandemic by providing surveillance in Tweets, which is essential to identify potentially effective policy programs in offering targeted support for victims and survivors and preparing for the next wave.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.017 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.003 | 0.006 |
| Open science | 0.000 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".