This repository contains source code for a tidyverse-focused revision to Learning statistics with R. A preview of the book can be found at https://djnavarro.github.io/tidylsrbook (though the eventual home for the online version of the book will be https://learningstatisticswithr.com). The current plan for the book is roughly:
- Part 1: Working with data
- Part 2: Learning from data
- Part 3: Tools of the trade
- Appendices
Chapters in this section start from the presumption that a data set already exists, and the analysts goal is to organise it, tidy it, describe it and visualise it.
Chapters in this section start from the presumption that the analyst has a more ambitious goal than describing data, and introduces tools for making sound inferences from data.
- Statistical learning: orthodox and Bayesian (t-test and chi-square as examples; cover bootstrapping, cross-validation, bias-variance etc here)
- Linear modelling in R (linear regression, ANOVA)
- [haven't decided what else yet]
Chapters in this section discuss other methodological tools that are central to psychological research
- Reproducibility, documentation and version control [git, GitHub, OSF, reprex, RMarkdown]
- Measurement and the design of experiments [sneak some philosphy of science in here]
- Implementing behavioural experiments in R [hm. Shiny, interactive graphics devices, maybe get off my lazy arse and write wrappers for jsPsych]
- Data structures in R
- Programming in R
- Probability theory