Module 3 · Inferential Statistics in R

Thu Oct 29, 2026 · noon–2pm PT

Materials coming soon

This page will fill in before the session.

Designed for someone with intermediate knowledge of R and RStudio. The first two modules exist to prepare you for this one.

In this session we will:

run the analyses that appear most often in papers — t-tests, chi-squared, ANOVA
discuss the considerations that longitudinal datasets like ABCD force on you, including violated normality assumptions and missing data
introduce linear models for testing hypotheses

Important

Module 3 is a prerequisite for Module 4, which is a hands-on project session assuming you can already choose and run a test.

The data

We work on data/data.RDS, the same ABCD-shaped dataset as Module 2 — 500 participants across five annual visits. Two of its properties drive this session: about 27% of cells are missing, and the missingness is structured rather than random, so complete-case analysis is never free. The dataset reference page sets out where the gaps come from.

Preparation

Module 2, or equivalent comfort with tidyverse pipelines and ggplot2
The missing-data section of Module 1 — we pick that thread up properly here
Download the dataset and its dictionaries into your project’s data/ folder

Resources

longitudinal.dev — a collection of statistical analyses for longitudinal studies
Running a paired t-test on ABCD data