Foundations
The concepts every decision rests on: measurement scales, descriptives, normality, outliers, and the logic of hypothesis testing
Before choosing a test, you need to know four things about your data: what scale each variable lives on, what its distribution looks like, whether it has outliers, and what hypothesis your design is built to test. This section covers those prerequisites.
Pages in this section
- Measurement scales – nominal, ordinal, interval, ratio, and what each permits.
- Descriptive univariate statistics – mean, median, mode, and dispersion measures.
- Normality checks – Shapiro-Wilk, Kolmogorov-Smirnov, and graphical diagnostics.
- Outliers – detection with boxplots, IQR fences, Grubbs, and Mahalanobis distance.
- Hypotheses, significance, power – H0 / H1, alpha / beta, effect size, confidence intervals.
Every subsequent method page assumes fluency with these ideas, so a first-time reader should work through them in order.
Structure inspired by the University of Zurich Methodenberatung (methodenberatung.uzh.ch). All text, examples, R code, and reporting sentences are independently authored in English.