Statistical Hypothesis Testing Tool
Paste the following prompt into your AI chat to install this skill:
Follow https://skillhub.cn/install/skillhub.md to install @engicool/hypothesis-testing.
About this skill
Problem
When working with a score column, treatment/control values, or frequency tables, the hard part is often deciding whether an observed difference is statistically significant and whether a parametric or nonparametric test is appropriate.
How It Works
This skill focuses on hypothesis testing for sample data, covering:
- Normality: Shapiro-Wilk / K-S
- Mean comparisons: one-sample, independent-sample, and paired t-tests
- Categorical analysis: chi-square goodness-of-fit and independence
- Multi-group comparisons: one-way ANOVA
- Assumption checks: Levene's test
It can accept CSV/TXT/Excel files, comma-separated lists, or numeric arrays, and relies on numpy, pandas, and scipy. The method is selected with --test, and the output reports the p-value, significance decision, and interpretation. --alpha, --expected, and --output can adjust the analysis. In practice, inspect sample size and distribution first, choose a parametric or nonparametric test, then evaluate effect size and confidence intervals for practical significance.
Boundaries
It answers whether a difference is significant, not whether it is causal. Independent-sample t-tests and ANOVA are more sensitive to normality and variance homogeneity. For severe skewness or unequal variances, prefer Mann-Whitney U or Wilcoxon. When running many tests, consider corrections such as Bonferroni.
Use Cases
- Compare treatment and control data using an independent-sample t-test, then check variance with Levene’s test.
- Inspect class score CSV files with Shapiro-Wilk before deciding whether parametric tests are acceptable.
- Validate die-roll counts or channel shares with a chi-square goodness-of-fit test against expected frequencies.
- Compare three teaching methods using one-way ANOVA to detect significant differences among group means.
Best For
- Data analysts who need a quick significance check and p-value for two-metric differences.
- Researchers who must verify normality and variance homogeneity before comparing experiments.
- Quality analysts who need to validate observed frequencies against expected distributions.
- Business analysts comparing groups from CSV data using t-tests, chi-square tests, or ANOVA.
Related Skills
Fetches Baidu Hot Search Top 10 titles using web_fetch first, validates same-day data, and falls back to browser automation when stale.
Generates an evening A-share policy and trading opportunity daily report by collecting same-day index, policy, and capital data, then applying a fixed template to highlight beneficiary sectors, drivers, and price directions.
Maps natural-language TikTok requests to KeyAPI REST scenarios, validates endpoints against docs, and executes data queries and analysis.
Extract city-specified AI jobs from BOSS Zhipin, save CSV/table data, mark new postings, and summarize salary trends, application advice, and HTML reports.