R Learning Resources
Conducting a meta-analysis in R can be a serious undertaking. Using R for meta-analysis also requires a strong statistical foundation in regression modeling approaches. Thankfully, there are numerous resources online that can help you wrap your head around R and its functions. On this page you will find resources and tutorials that we have found effective in guiding our learning of R and meta-analysis. Additionally, our Video Guides provides basics on installing R and RStudio, calculating effect sizes, and creating meta-analytic models. Before you get started, consider this piece of advice from MALIC as you navigate your R learning journey:
Set goals! Solid goals and objectives provide a roadmap to guide your learning experience. These goals and objectives should be specific and attainable. For example, "learn how to do a meta-analysis in R" can be the ultimate goal. Specific objectives can include "learning how to calculate effect sizes using the metafor package" or "create a meta-regression model in R using clubSandwich." Setting goals and objectives can recenter you as you navigate R's seemingly endless list of packages and functions. Don't worry though! Set goals. Take breaks. Breathe deeply. You can do this!
Note: Some of the resources provided on this page may require familiarity with Github to access the learning material.
Created as an introduction to R and RStudio, the 25 Days of Christmas R Calendar will walk you through 25 short daily exercises to get familiar with the basics of R. Notably, this tutorial provides a starting point for learning data cleaning and manipulation using the "dplyr" package. Most lessons can be completed in less than 10 minutes and can be a perfect starter for someone wanting to dip their toes in R.
This tutorial is one-part statistics textbook and one-part comic book adventure! Andy Field, a professor of psychopathology at the University of Sussex, developed this tutorial using his background in psychology. Researchers and practitioners in psychological disciplines may find some familiarity in Dr. Field's instruction style and analysis techniques. Tutorial contents follow the typical analysis workflow of importing and cleaning data, examining associations and correlations between constructs, and building regression models for hypothesis testing. You will be required to download and install the "learnr" package in R before starting the tutorial.
Keith McNulty is a veteran data scientist specializing in people analytics data. He has extensive knowledge of programming in Python and R. His recent textbook, Regression Modeling in People Analytics, was recently published and is available online for free. McNulty's tutorial is notable in his focus on using basic R functions. This can be valuable for a learner who wants to get a handle on basic R functions before moving on to individual packages. Again, this tutorial focuses primarily on statistical hypothesis testing using regression models, but also provides instruction in advanced techniques such as hierarchical data modeling, survival analysis, and power analysis for sample size optimization.
As the title suggests, this tutorial is geared towards researchers and practitioners in the field of education. Written Ryan Estrellado, Emily A. Freer, Jesse Mostipak, Joshua M. Rosenberg, and Isabella C. Velasquez, the Data Science for Education textbook aims to accomplish the daunting goal of mapping out a clear path for data science in the field of education. While less focused on specific functions and packages, this tutorial is designed to demonstrate how data science approaches in R can be used to address specific topics in education. These include visualization of gradebook data, longitudinal analysis of achievement outcomes in special education, hierarchical modeling, as well as an introduction to basic machine learning for predictive analytics in education.
The R for Data Science provides a comprehensive introduction to the "tidyverse" package for processing data in R. Cleaning, subsetting, and manipulating datasets is an important first step in analyzing data. Tidyverse. This package -- created by Hadley Wickham, a chief data scientist with RStudio, has been recognized as a standard tool for data cleaning and manipulation in R.
Wolfgang Veichtbauer, the author of the metafor package, maintains an updates a website dedicated to understanding how to use metafor for meta-analysis in R. You can follow the link above to his Github page, which serves as a one-stop shop for metafor updates, documentation, and news related to workshops, webinars, and other meta-analysis related events.