# Eval Anova

> Run Design-of-Experiments (DoE) evaluations with ANOVA over a matrix of agent configurations — comparing models, thinking-effort levels, prompts, or other factors across shared test cases, with repeated-measures / mixed-effects statistics that account for case difficulty plus a cost/quality Pareto view. Use whenever the user wants to compare models or configurations on an eval, decide which model or config is best, sweep or grid factors, run replications, or check whether a difference in eval scores is statistically significant (F, p, effect size) — even if they don't say "ANOVA" or "DoE". Also use when an eval.yaml has a matrix block or the user asks to fan an eval out across configurations.

## Facts
- Page: https://tashan.sh/capability/skill-opendatahub-io-eval-anova
- tashan id: skill:opendatahub-io/eval-anova
- Source: https://github.com/opendatahub-io/agent-eval-harness
- Type: skill
- Category: other
- tashan score: not scored (catalogued only — too little public evidence)
- Adoption: 9.0
- Upkeep: 97.0
- Freshness: 94.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- License: Apache-2.0
- Official: no

## Install

```sh
cp -r eval-anova ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-09-12 by tashan (https://tashan.sh) from public evidence. Scorer s5.