Quantifying qualitative data for understanding controversial issues
Bibliographic record
Abstract
Understanding public opinion on complex controversial issues such as 'Legalization of Marijuana' and 'Gun Rights' is of considerable importance for a number of objectives such as identifying the most divisive facets of the issue, developing a consensus, and making informed policy decisions. However, an individual's position on a controversial issue is often not just a binary support-or-oppose stance on the issue, but rather a conglomerate of nuanced opinions and beliefs on various aspects of the issue. These opinions and beliefs are often expressed qualitatively in free text in issue-focused surveys or on social media. However, quantifying vast amounts of qualitative information remains a significant challenge. The goal of this work is to provide a new approach for quantifying qualitative data for the understanding of controversial issues. First, we show how we can engage people directly through crowdsourcing to create a comprehensive dataset of assertions (claims, opinions, arguments, etc.) relevant to an issue. Next, the assertions are judged for agreement and strength of support or opposition, again by crowdsourcing. The collected Dataset of Nuanced Assertions on Controversial Issues (NAoCI dataset) consists of over 2,000 assertions on sixteen different controversial issues. It has over 100,000 judgments of whether people agree or disagree with the assertions, and of about 70,000 judgments indicating how strongly people support or oppose the assertions. This dataset allows for several useful analyses that help summarize public opinion. Across the sixteen issues, we find that when people judge a large set of assertions they often do not disagree with the individual assertions that the opposite side makes, but that they differently judge the relative importance of these assertions. We show how assertions that cause dissent or consensus can be identified by ranking the whole set of assertions based on the collected judgments. We also show how free-text assertions in social media can be analyzed in conjunction with the crowdsourced information to quantify and summarize public opinion on controversial issues.
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 | 0.051 | 0.222 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.015 | 0.016 |
| Science and technology studies | 0.003 | 0.004 |
| Scholarly communication | 0.007 | 0.009 |
| Open science | 0.003 | 0.007 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.010 | 0.002 |
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".