statistical-analysis

Choosing the right hypothesis test requires a structured decision workflow to select appropriate methods like t-tests, ANOVA, or chi-square for group comparisons, and regression or correlation for relationship modeling. Bayesian alternatives are also supported for more complex study designs.

Checking statistical assumptions involves systematic diagnostic checks for normality, homogeneity of variance, outliers, and linearity. The skill provides explicit guidance and recommends remedial options like Welch's tests or non-parametric alternatives when assumptions are violated.

Calculating effect sizes and performing a priori power analysis are supported through built-in estimation guidance and sensitivity analysis. The workflow emphasizes effect sizes with confidence intervals to ensure rigorous study planning and multiple-comparison considerations.

Generating APA-style reporting for statistical analysis is a core feature that structures test statistics, diagnostics, and effect sizes into publication-quality writeups. It produces professional APA-aligned paragraphs ready for academic reporting.

Using Bayesian analysis instead of traditional hypothesis testing is appropriate when seeking alternative methods for group comparisons or relationship modeling. The skill guides you to Bayesian variants alongside standard regression, correlation, and hypothesis testing approaches.

This statistical analysis skill works for both group comparisons and relationship modeling across common academic study designs. It applies systematic decision workflows, assumption diagnostics, and effect-size calculations to satisfy diverse research requirements.