hypothesis-test

Hypothesis testing requires matching your test to your data structure and research question. This Skill uses decision-tree logic to guide you through test selection for one, two, or multiple variables, accounting for your sample size, variable types, and study design to recommend the appropriate t-test, ANOVA, or regression approach.

Statistical tests depend on assumptions like normality, homogeneity of variance, and independence of observations. This Skill provides structured guidance to check these assumptions, helps you interpret assumption-violation tests, and recommends alternative tests or transformations when assumptions fail.

P-values and effect sizes tell different stories about your data. This Skill explains how to report test statistics, degrees of freedom, p-values, and confidence intervals for effect sizes correctly, ensuring your conclusions match your results and avoiding common misinterpretations of statistical significance.

A paired t-test compares measurements from the same subjects at different times when observations are dependent. This Skill's decision tree identifies when paired designs apply versus independent-sample scenarios, and verifies the independence and normality assumptions required for valid results.

Assumption violations require different responses depending on severity and test type. This Skill guides you through resolving violations via data transformation, non-parametric alternatives, or robust test variants, and shows how to report when assumptions cannot be met.

Standardized reporting includes test statistic, degrees of freedom, p-value, effect size with confidence interval, and descriptive statistics. This Skill provides reporting guidance to present results with the precision and clarity required by your field's conventions.