What problem does it solve? Choosing the right statistical test, verifying its assumptions, and reporting results correctly is error-prone for researchers. This Skill provides a guided workflow for hypothesis testing, regression, correlation, and Bayesian analysis with automated assumption checks and APA-style reporting templates. ## Core Features & Use Cases - Test Selection & Assumption Checking: Decision trees for choosing t-tests, ANOVA, chi-square, or non-parametric alternatives, plus an automated assumption_checks.py module for normality, homogeneity of variance, linearity, and outlier detection. - Effect Sizes & Power Analysis: Calculate Cohen's d, eta-squared, Cramér's V, and Bayes Factors with confidence intervals, and run a priori or sensitivity power analyses via statsmodels. - APA Reporting: Ready-made templates for reporting t-tests, ANOVA, regression, chi-square, and Bayesian results with all required statistics. - Use Case: A graduate student comparing exam scores across three teaching conditions uses the Skill to verify ANOVA assumptions, run the test with Tukey post-hoc comparisons, compute partial eta-squared, and produce a publication-ready results paragraph. ## Quick Start Use the statistical-analysis skill to compare my two groups in scores.csv, check the assumptions, and report the results in APA format.
