Steepness is a statistical property of dominance hierarchies defined as the slope fitted to normalized David's scores. The steepness package computes this metric from observed sociomatrices and estimates statistical significance via randomization tests. Authors David Leiva and Han de Vries developed this method for analyzing dyadic dominance indices.
Use Cases
- Calculate hierarchy steepness based on normalized David's scores.
- Estimate statistical significance of dominance patterns via randomization tests.
- Analyze dyadic dominance indices corrected for chance.
- Compute dominance hierarchies from proportions of wins.
- Model social structure using sociomatrix data.
Strengths
- Provides a defined metric (steepness) for quantifying hierarchy structure.
- Includes a method for statistical significance estimation via randomization.
- Authors are identified with institutional affiliations.
- Platform tags indicate cross-disciplinary relevance.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.