Intro

A number of my colleagues need to analyze data and are learning a programming language for the first time. It does not need to be said, this is a very heavy lift and occurs in the presence of a wide diversity of well intentioned opinion. It also occurs in the presence of academic bullshit. This is not well intentioned. It is designed to sell books and software licenses. It is my opinion that the appropriate first scripting language to learn for the aspiring social scientist is R. Arguements could be made for perl, python or lisp. I think these great languages can be learned later. For describing relationships with data and graphing results, R has a great set of libraries.

R Front ends

There are a number of front-ends for R. The most popular of which is R-studio. I think it leaves much to be desired. It is also produced by a company that sells enhancements to R that most users will never need. I find the better and more cost effective solution to be Emacs-Speaks-Statistics (ESS). Much of the power of the language comes from being able to read from files and streams. Combining this with the uber editor Emacs only makes sense. For many data manipulation problems the student can solve it either with either tool.

The intellectual history of this tutorial is taken from Charles DiMaggio's excellent introduction to R and two brilliant tutorials from Stephen Elgen An introduction to ESS from 2011 and Caimbridge R bootcamp.

Here is our introduction:

Emacs

• Why Emacs
• History of Emacs
• Moving around
• Cut and paste
• Loading/saving files
• Windows
• Search and replace
• Macros
• Modes
• Completion
• Getting help
• libraries
• Hooks
• Philosophy

ESS

• Why ESS?
• History of ESS
• ESS Installation
• Editing and viewing R code
• Making comments
• Indenting your code.
• Viewing help files
• Making help files (.Rd)
• Inferior ESS Processes (*R*)
• Which version of R will it find?
• What can I do in *R*?
• What is emacsclient?
• Transcripts
• How many versions of R can I run?
• Sending code from an R buffer to *R*
• Customizing ESS
• When things go wrong with ESS

Org Mode

• intro
• exporting to LaTeX or html
• R examples

R language

• introduction
• foundations
• functions
• packages
• graphics
• data
• variables

Advanced stats

• power
• web/online
• bayes (multilevel, hierarchical)
• spatial
• meta-analysis

More to come…