MOVICI Article 2 reproducibility package
Data audit

Participant table
- 96 participant rows.
- 34 debate clusters.
- 47 Competition and 49 Cooperation participants.
- Verified debate coding for 85 participants; 11 rows retain missing debate-derived values.

Dyad table
- 212 ordered evaluator-to-evaluatee rows.
- 91 evaluators across 29 debate clusters.
- No self-evaluation dyads.
- Missing perceived expertise: 16.
- Missing perceived benevolence: 0.
- Missing mean argument evaluation: 20.

Data minimization
- Released columns match the variables required by the analysis script.
- Exact dates and times, workstations, internal identifiers, demographics, raw transcripts, free text, and unrelated MOVICI variables are excluded.
- Article 1 and Article 2 use independent public pseudonyms to prevent cross-package linkage.

Partner-composite provenance
- Perceived expertise: 223 of 223 complete source totals reproduced.
- Perceived benevolence: 224 of 224.
- Perceived social hierarchy: 224 of 224.
- First argument total: 202 of 202.
- Second argument total: 161 of 161.

Bootstrap policy
- The retained exploratory indirect path uses 500 debate-cluster resamples.
- The path is exploratory and is not interpreted causally.
