statistical-analysis

To generate APA-style statistical reports, this Skill applies hypothesis testing, computes effect sizes with confidence intervals, and formats publication-ready results. It structures the end-to-end workflow from test selection to final documentation for academic studies.

The best way to run assumption checks is using built-in diagnostic scripts that test normality and homogeneity of variances. The Skill provides actionable recommendations and visualizations to validate model assumptions before proceeding with statistical tests.

To select the right statistical test, the Skill guides you through t-tests, ANOVA, chi-square, and nonparametric options based on your experimental design and data structure. It evaluates your dataset characteristics to recommend the appropriate analytical approach.

Yes, you can use statsmodels and pingouin for regression diagnostics and power analysis. The Skill leverages these dependencies alongside scipy.stats and pymc to perform comprehensive model checks and compute statistical power across varied sample sizes.

Bayesian statistical analysis works with pymc and arviz by integrating them into the workflow for model fitting and result visualization. The Skill supports these libraries to complement traditional hypothesis testing and expand analytical capabilities for complex research data.

Limitations of automated statistical test selection include the necessity of clean input data and proper study design definition. While the Skill validates assumptions and recommends tests, researchers must still correctly specify their experimental design and interpret effect sizes contextually.