Twitter Sentiment About the US Federal Tobacco 21 Law: Mixed Methods Analysis
Bibliographic record
Abstract
BACKGROUND: On December 20, 2019, the US "Tobacco 21" law raised the minimum legal sales age of tobacco products to 21 years. Initial research suggests that misinformation about Tobacco 21 circulated via news sources on Twitter and that sentiment about the law was associated with particular types of tobacco products and included discussions about other age-related behaviors. However, underlying themes about this sentiment as well as temporal trends leading up to enactment of the law have not been explored. OBJECTIVE: This study sought to examine (1) sentiment (pro-, anti-, and neutral policy) about Tobacco 21 on Twitter and (2) volume patterns (number of tweets) of Twitter discussions leading up to the enactment of the federal law. METHODS: We collected tweets related to Tobacco 21 posted between September 4, 2019, and December 31, 2019. A 2% subsample of tweets (4628/231,447) was annotated by 2 experienced, trained coders for policy-related information and sentiment. To do this, a codebook was developed using an inductive procedure that outlined the operational definitions and examples for the human coders to annotate sentiment (pro-, anti-, and neutral policy). Following the annotation of the data, the researchers used a thematic analysis to determine emergent themes per sentiment category. The data were then annotated again to capture frequencies of emergent themes. Concurrently, we examined trends in the volume of Tobacco 21-related tweets (weekly rhythms and total number of tweets over the time data were collected) and analyzed the qualitative discussions occurring at those peak times. RESULTS: The most prevalent category of tweets related to Tobacco 21 was neutral policy (514/1113, 46.2%), followed by antipolicy (432/1113, 38.8%); 167 of 1113 (15%) were propolicy or supportive of the law. Key themes identified among neutral tweets were news reports and discussion of political figures, parties, or government involvement in general. Most discussions were generated from news sources and surfaced in the final days before enactment. Tweets opposing Tobacco 21 mentioned that the law was unfair to young audiences who were addicted to nicotine and were skeptical of the law's efficacy and importance. Methods used to evade the law were found to be represented in both neutral and antipolicy tweets. Propolicy tweets focused on the protection of youth and described the law as a sensible regulatory approach rather than a complete ban of all products or flavored products. Four spikes in daily volume were noted, 2 of which corresponded with political speeches and 2 with the preparation and passage of the legislation. CONCLUSIONS: Understanding themes of public sentiment-as well as when Twitter activity is most active-will help public health professionals to optimize health promotion activities to increase community readiness and respond to enforcement needs including education for retailers and the general public.
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 | 0.023 | 0.060 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
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".