I started learning statistics because
my supervisor asked me to I find it really useful and mind-opening. In the early days, I mainly consulted Andy Field’s textbook Discovering Statistics Using R – if you are okay with numerous examples that involve sperms and childhood anecdotes, I’d say give it a try. Alternatively, there’s Probabilistic Models in the Study of Language (in progress) and Quantitative Methods in Linguistics, both of which are tailor-made for us linguists.
Speaking from my very limited experience, however, the best way to learn statistics with R is to just sit down and spend hours on some actual data (and maybe call your mom and cry every other hour). I prefer studying alone (or occasionally with my imaginary friends), so whenever a problem occurs, my strategy is to google it, youtube it, or stackexchange it before begging a stats God for their help. Bottom line: there are numerous ways to learn stats and R, but you won’t actually learn anything unless you make it relevant for you.
Rambling aside, here are some materials that I have been using regularly with the hope that one day I will turn into an R magician myself. Thanks to the people mentioned here, I have eventually made peace with statistics and even started to have fun with it! (Wait, how? Find out at the end of this page!)
(Inspired by the #barbarplots campaign, I made this cake back in June 2016)
Tips & tutorials from R ninjas
- Steve Politzer-Ahles has written many really useful tutorials on his website
- Page Piccinini has some great online materials for learning R on her website
- Roger Peng has written some stuff on doing Exploratory Data Analysis with R
- Winston Chang’s Cookbook for R provides solutions to common tasks and (very basic) problems in analysing data with R
- Andy Field has posted some of his lectures on Youtube
Awesome stats blogs
- Andrew Gelman: statistical modeling, causal inference, and social science
- CogTales: a blog run by young female cognitive scientists. Covering topics about life in the academia as well
- Data Colada: fairly short posts involving quantitative analyses, replications, and interesting discussion
- Error Statistics Philosophy: stats blog run by Deborah Mayo
- Simply Statistics: stats blog by Rafa Irizarry, Roger Peng, and Jeff Leek
- Shravan Vasishth’s Slog: statistical computing, simulation and stochastic modeling
- Rolf Zwaan: Psychological Experimentation, Cognition, Language, and Academia
- sometimes i’m wrong: stats and social science blog run by Simine Vazire
- The 20% Statistician: statistics, methods, and open science
- The Hardest Science: a psychology blog about the mind, science, and society, written by Sanjay Srivastava
Websites for general resources
- R Bloggers: a blog full of R news and tutorials, contributed by hundreds of R bloggers
- StatsBlogs: similar to R Bloggers, contributed by stats bloggers
Modeling & Analyses
- Mixed-effects modeling in R, workshop given by Steve Politzer-Ahles at University College London in May 2016
- Logistics regression in R
- ANOVA: one-way / two-way
- d-prime (SDT): NYU handout / UCLA tutorial / WISE interactive tutorial
- Continuously Cumulating Meta-Analysis [Braver et al 2015]
- ‘retimes’ package for ex-Gaussian analysis [see also Zhou &
Cheat sheets for plotting in R
And finally…… Fun with R (no kidding)!
- Suffering from the imposter syndrome? Have R say sweet things to you by using the praise package.
- Party of #rcatladies: check out the cats package for cat-related functions, including how to grab a random cat from the internet. For other animals, see the cowsay package.
- Games in R: there’s the sudoku package, and the fun package which includes the classical Mine sweeper, gomoku, and sliding puzzles.
- Too lazy to even open R? Sometimes I play this little game called Guess the Correlation. Let’s just call it work-life balance.