Inferential Statistics

Here we move from describing a single sample to drawing conclusions about the population behind it. You will learn to test hypotheses with t-tests, ANOVA, and post-hoc comparisons; to quantify how large and how reliable an effect is through effect sizes, p-values, and statistical power; and to choose the right approach when the assumptions of normality or equal variance break down. The block also covers how to identify outliers and apply data transformations honestly, and Case Studies I brings these tools together on realistic chemical data. It closes with Bayesian statistics — now that you know the frequentist toolkit, we revisit probability as a degree of belief and contrast the two frameworks.