statistics
This skill guides you through test selection for continuous, categorical, and time-to-event data across different study designs. It includes assumption verification methods, multiple comparison corrections, and effect size benchmarks to ensure rigorous analysis. Follow integrated reporting standards to communicate results with full transparency.
Statistics helps you select the right statistical test and verify assumptions for your research data.
AI-generated summary based on this skill's SKILL.md
Install
beita6969/ScienceClaw/statistics · repository language: TypeScript
git clone https://github.com/beita6969/ScienceClaw
cp -r ScienceClaw/skills/statistics ~/.claude/skills/statisticsnpx skillfed install beita6969/ScienceClaw/statisticsFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How do I choose the right statistical test for my research data?
Statistics guides you through test selection by first identifying your data type (continuous, categorical, or time-to-event) and study design. For continuous data, check normality using Shapiro-Wilk or Kolmogorov tests; if normal, use t-tests or ANOVA; if non-normal, use Mann-Whitney U or Kruskal-Wallis. For categorical data, use chi-square or Fisher's exact test. For survival data, apply Cox proportional hazards. The skill helps you verify assumptions before committing to any test.
What are ANOVA assumptions and post hoc tests I should know?
Statistics covers ANOVA assumptions: normality of residuals, homogeneity of variance (test with Levene's test), and independence of observations. When ANOVA is significant, use post hoc tests like Tukey HSD or Bonferroni correction to compare groups while controlling family-wise error rate. For repeated measures ANOVA with paired data, the skill also addresses sphericity assumptions and appropriate corrections.
How should I report effect sizes and confidence intervals correctly?
Statistics emphasizes reporting effect sizes (Cohen's d, eta-squared, odds ratios) alongside p-values and 95% confidence intervals. This provides readers with practical significance, not just statistical significance. The skill includes interpretation benchmarks and shows how to calculate and present these metrics according to scientific standards, ensuring your results communicate both magnitude and precision of findings.
What multiple comparison corrections should I apply to my analysis?
Statistics guides you through multiple comparison corrections including Bonferroni (conservative, controls family-wise error rate) and False Discovery Rate (FDR) methods (less conservative, controls proportion of false positives). The choice depends on your study goals and number of comparisons. The skill helps you decide which correction fits your research context and implement it correctly.
How do I verify statistical assumptions before running my analysis?
Statistics provides methods to check key assumptions: use Shapiro-Wilk or Kolmogorov-Smirnov tests for normality, Levene's test for homogeneity of variance, and visual tools like Q-Q plots. The skill walks you through interpreting these tests and deciding whether to proceed with parametric tests or switch to non-parametric alternatives like Wilcoxon signed-rank or Mann-Whitney U tests.
What reporting standards should I follow when publishing results?
Statistics integrates STROBE, CONSORT, and PRISMA reporting guidelines to ensure transparency. These standards require clear reporting of methods, effect sizes, confidence intervals, and p-values. The skill helps you structure your results section to meet journal requirements and communicate findings with full methodological transparency, building reader confidence in your statistical rigor.
SKILL.md
rendered from the published skill — quoted content, verbatim
Statistical Analysis & Quality Control
Overview
Comprehensive statistical methodology for scientific research. Covers test selection, assumption verification, power analysis, effect size reporting, and reporting standards.
Test Selection Guide
| Data Type | Groups | Paired? | Normal? | Recommended Test |
|---|---|---|---|---|
| Continuous | 2 | No | Yes | Independent t-test |
| Continuous | 2 | No | No | Mann-Whitney U |
| Continuous | 2 | Yes | Yes | Paired t-test |
| Continuous | 2 | Yes | No | Wilcoxon signed-rank |
| Continuous | 3+ | No | Yes | One-way ANOVA + post hoc |
| Continuous | 3+ | No | No | Kruskal-Wallis + Dunn |
| Continuous | 3+ | Yes | Yes | Repeated measures ANOVA |
| Categorical | 2x2 | — | — | Chi-square / Fisher's exact |
| Time-to-event | 2+ | — | — | Log-rank + KM curves |
| Time-to-event | Adjusted | — | — | Cox proportional hazards |
| Continuous | Prediction | — | — | Linear/logistic regression |
Assumption Checks
- Normality: Shapiro-Wilk (n < 50), Kolmogorov-Smirnov (n > 50), Q-Q plot visual
- Homoscedasticity:
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 1 file
skills/statistics/SKILL.md