Machine Learning foR Psychologists

_{Last updated 2025-06-25.}

This Github repo contains all lesson files for Machine Learning in R. The goal is to impart students with the basic tools to construct, evaluate and compare various machine learning models, using tidymodels. (Prior to 2025, this course was based on the caret package - this version can still be found here.)

These topics were taught in the graduate-level course Machine Learning for Psychologists (Psych Dep., Ben-Gurion University of the Negev; Psych Dep., Tel-Aviv University). (Materials developed with Yael Bar-Shachar.) This course assumes basic competence in R (importing, regression modeling, plotting, etc.), along the lines of Practical Applications in R for Psychologists.

Notes:

• This repo contains only materials relating to Practical Applications in R, and does not contain any theoretical or introductory materials.
• Please note that some code does not work on purpose, to force students to learn to debug.

Setup

You will need:

1. A fresh installation of R (preferably version 4.3.2 or above).
2. RStudio IDE (optional, but recommended).
3. The following packages, listed by lesson:

Lesson	Packages
01 Intro with Regression	tidymodels, ISLR, stats, patchwork, kknn
02 Classification	tidymodels, ISLR, stats, parameters, kknn, palmerpenguins, themis, patchwork, probably, tailor, modeldata
03 Resampling and Tuning	tidymodels, kknn, finetune, ISLR, glue, modeldata, patchwork, datasets, tune, performance
04 The problem of over-dimensionality	tidymodels, leaps, ISLR, stats, MASS, insight, glmnet, tidyr, scales, tibble, ggplot2, vip, kknn, plsmod, BiocManager
05 SVM	tidymodels, kernlab, ISLR, palmerpenguins
06 Trees	tidymodels, rpart, rpart.plot, ISLR, scales, vip, MASS, tidymodels, baguette, randomForest, xgboost, MASS, forcats, vip
07 explain predictions	tidymodels, kknn, randomForest, patchwork, DALEX, DALEXtra, marginaleffects, ISLR, palmerpenguins, vip
08 unsupervised learning	tidyverse, patchwork, recipes, Rtsne, factoextra, ggrepel, cluster, randomForest, modeldata, psych, parameters, performance, dplyr, tidyr, datasets, GPArotation, psychTools

You can install all the packages used by running:

# in alphabetical order:

pkgs <- c(
  "baguette", "BiocManager", "cluster", "DALEX", "DALEXtra",
  "datasets", "dplyr", "factoextra", "finetune", "forcats", "ggplot2",
  "ggrepel", "glmnet", "glue", "GPArotation", "insight", "ISLR",
  "kernlab", "kknn", "leaps", "marginaleffects", "MASS", "modeldata",
  "palmerpenguins", "parameters", "patchwork", "performance", "plsmod",
  "probably", "psych", "psychTools", "randomForest", "recipes",
  "rpart", "rpart.plot", "Rtsne", "scales", "stats", "tailor",
  "themis", "tibble", "tidymodels", "tidyr", "tidyverse", "tune",
  "vip", "xgboost"
)

install.packages(pkgs, dependencies = TRUE)

Package Versions

The package versions used here:

• baguette 1.1.0 (CRAN)
• BiocManager 1.30.25 (CRAN)
• cluster 2.1.6 (CRAN)
• DALEX 2.4.3 (CRAN)
• DALEXtra 2.3.0 (CRAN)
• datasets 4.4.1 (Dev)
• dplyr 1.1.4 (CRAN)
• factoextra 1.0.7 (CRAN)
• finetune 1.2.0 (CRAN)
• forcats 1.0.0 (CRAN)
• ggplot2 3.5.1 (CRAN)
• ggrepel 0.9.6 (CRAN)
• glmnet 4.1-8 (CRAN)
• glue 1.8.0 (CRAN)
• GPArotation 2024.3-1 (CRAN)
• insight 1.3.0 (CRAN)
• ISLR 1.4 (CRAN)
• kernlab 0.9-33 (CRAN)
• kknn 1.3.1 (CRAN)
• leaps 3.2 (CRAN)
• marginaleffects 0.26.0.3 (Github: vincentarelbundock/marginaleffects)
• MASS 7.3-60.2 (CRAN)
• modeldata 1.4.0 (CRAN)
• palmerpenguins 0.1.1 (CRAN)
• parameters 0.26.0.1 (Dev)
• patchwork 1.3.0 (CRAN)
• performance 0.14.0 (CRAN)
• plsmod 1.0.0 (CRAN)
• probably 1.0.3 (CRAN)
• psych 2.4.12 (CRAN)
• psychTools 2.4.3 (CRAN)
• randomForest 4.7-1.2 (CRAN)
• recipes 1.1.0 (CRAN)
• rpart 4.1.23 (CRAN)
• rpart.plot 3.1.2 (CRAN)
• Rtsne 0.17 (CRAN)
• scales 1.3.0 (CRAN)
• stats 4.4.1 (Dev)
• tailor 0.0.0.9002 (Github: tidymodels/tailor)
• themis 1.0.3 (CRAN)
• tibble 3.2.1 (CRAN)
• tidymodels 1.2.0 (CRAN)
• tidyr 1.3.1 (CRAN)
• tidyverse 2.0.0 (CRAN)
• tune 1.2.1 (CRAN)
• vip 0.4.1 (CRAN)
• xgboost 1.7.8.1 (CRAN)