statistical-analysis

Test selection guidance matches your experimental or observational data to t-tests, ANOVA, regression, chi-square, correlations, non-parametric, and Bayesian workflows. The skill evaluates your dataset structure to recommend the appropriate statistical test.

Assumption checking validates normality, homogeneity of variance, linearity, and outliers through automated diagnostics. It generates residual diagnostics and heterogeneity checks to ensure your ANOVA or regression models meet required statistical assumptions.

APA-style reporting templates format your results with effect sizes and power analyses included. The skill produces code-ready outputs that follow APA reporting standards for t-tests, ANOVA, and regression results.

Bayesian workflows are supported alongside traditional hypothesis testing for comprehensive analysis. The skill guides both Bayesian and frequentist approaches, applying reproducible workflows across experimental and observational datasets.

Reproducible workflows are enforced through structured scripts and references using numpy, pandas, scipy, matplotlib, and seaborn. The skill generates analysis plans, performs diagnostics, and outputs APA-ready results consistently.

T-test assumption checks verify normality and homogeneity of variance to prevent invalid conclusions. The skill performs automated diagnostic tests for outliers and residual patterns before executing hypothesis testing and calculating effect sizes.