Install any skill in seconds. Free to start, no credit card required.
Get Started Free →面向 Nature Portfolio 与高影响力期刊的实验设计一致型统计分析、审查和报告技能。用于定义独立实验单位与重复层级、制定统计分析计划、检查数据质量、选择模型与估计量、计算效应量和不确定性、处理多重比较与敏感性分析,并重写 Methods、Results、表格和图注中的统计文本。触发场景包括 Nature 统计、数据统计、statistical analysis、p value、sample size、effect size、confidence interval、replicates、multiple comparisons、统计方法、图注统计和审稿人统计意见。
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 61% | 0% |
从研究设计和估计目标出发,再选择模型和检验。统计显著性不能替代效应大小、数据质量或科学意义。
plan:在分析前定义问题、实验单位、主要终点、模型、校正与敏感性分析。analyse:用户提供数据后执行可复现分析,并保留数据处理与诊断记录。audit:审查现有统计方法、结果、表格与图注。rewrite:在事实充分时生成可粘贴的统计方法或结果文本。review-response:解析审稿人统计问题,给出验证路径与保守回复要点。复杂临床试验、监管分析或患者级决策必须服从协议、统计分析计划和专业统计师审核。
n 如何定义?这些事实不清时,不给出最终检验选择;使用 AUTHOR_INPUT_NEEDED。
python scripts/reporting_audit.py <file> --context methods|results|legend。实验单位、伪重复和常见故障读取 references/design-integrity.md。
textStatistical scope - Mode / input / boundary: - Scientific question and estimand: - Independent unit and n: - Design hierarchy: Analysis specification - Outcome / predictors / contrasts: - Model or test: - Assumptions and diagnostics: - Multiplicity: - Sensitivity analyses: Results - Effect estimate and uncertainty: - Exact inferential result: - Practical interpretation: Ready-to-paste reporting [Methods / Results / legend] AUTHOR_INPUT_NEEDED - [事实性缺口] Reviewer-risk note - [剩余风险]
n。| 任务 | 读取 | |---|---| | 实验单位、嵌套、重复测量、伪重复、缺失与排除 | references/design-integrity.md | | estimand、模型选择、诊断、效应量、多重比较与敏感性 | references/analysis-plan.md | | Methods、Results、表格、图注和统计图形报告 | references/reporting-and-figures.md |
Other measured skills in the registry, with their headline benchmark lift.