reporting-effect-sizes

What problem does it solve?

This Skill prevents incomplete or misleading statistical reporting by requiring an effect size, a confidence interval, and a clear direction statement instead of a bare p-value.

Core Features & Use Cases

• Test-aware reporting: Chooses the right effect size for paired t-tests, Wilcoxon tests, chi-squared tests, McNemar tests, ANOVA, and regression.
• Confidence intervals: Adds a 95% confidence interval using the appropriate closed-form method or bootstrap approach.
• Publication-ready wording: Produces a report sentence that names the test, sample size, statistic, p-value, effect size, and which group was higher or lower.
• Guardrails against misuse: Refuses to label a result significant from p-values alone and avoids mismatched metrics such as unpaired d for paired data.
• Use case: A student or analyst can turn a raw hypothesis test output into a defensible results sentence for a paper, homework submission, dashboard, or presentation.

Quick Start

Ask the skill to rewrite your statistical test result into a publication-ready statement with the correct effect size, a 95% confidence interval, and a direction sentence.