Learning to Code
Many colleagues have asked me about resources for learning R, Python, and causal inference statistics. I’m not expert enough in Python to recommend good learning resources, but these are my favorite resources for learning the basics of R and for learning to use R for particular tasks (like causal inference). Anyone can learn to code, and these resources help bring out the coder in all of us.
These are all free, online resources. They give you code that you can easily copy & paste into your R console, which allows you to play around with R and learn by doing. To keep the list manageable, I’ve limited it to one resource for learning the basics of R and five resources for learning to use R for specific tasks.
If you are brand new to coding, I recommend starting with the first resource (Hands-On Programming with R) and then selecting other options based on your learning interests.
Learning R
This book will teach you the basics of programming in R. It is written it for non-programmers and provides a friendly introduction to the R language. Throughout the book, you’ll use your newfound skills to solve practical data science problems. Start here.
Specific tasks in R
Data science: R for Data Science
This online book will walk you through everything you need to get started using R for data science. It is comprehensive and will provide you with a solid foundation for learning to do other tasks in R. This is a good place to start if you have some prior experience with R.
Causal Inference: Causal Inference: The Mixtape
This excellent book “introduces students and practitioners to the methods necessary to arrive at meaningful answers to the questions of causation, using a range of modeling techniques and coding instructions for both the R and the Stata programming languages.” Go here to learn about matching, difference-in-differences, regression discontinuity, synthetic controls, and more.
Machine learning: Basics of Statistical Learning
This online version of a stats course at the University of Illinois provides “a broad introduction to machine learning from the perspective of a statistician who uses R and emphasizes practice over theory”. Good for people with a basic understanding of R and Regression.
Distributed computing: Mastering Spark with R
This excellent online textbook will teach you how to analyze “big data” using R. It assumes no prior knowledge of R or of distributed computing.
Data visualizations: R Graphics Cookbook
This is “a practical guide that provides more than 150 recipes to help you generate high-quality graphs quickly, without having to comb through all the details of R’s graphing systems.”
General coding assistance
Learning a coding language is much like learning any other language. The guides on this page will help, but sometimes you need a translator. AI coding tools can function like a translator, both (1) translating your words into a coding language, and (2) translating code into words understandable to you. This does not mean AI will code for you; using a translator requires you to know what you want to say, to say it precisely, and (sometimes) to identify when the translator has made a mistake! But I have found these AI translators very helpful, especially for learning new tasks and for troubleshooting when code goes wrong.
There are many good AI tools for coding assistance, but I recommend Cursor. It provides a text editor where you can write your code, and a built in AI chat window that can see the code in the text editor, explain what the code is doing, and recommend improvements. It also allows you to interact with several AI language models.
Good luck!