CSSS508

Welcome to CSSS 508: Introduction to R for Social Scientists

University of Washington, Spring 2016

Instructor: Rebecca Ferrell, PhD student, UW Statistics

Canvas page for course (enrolled students only)

• R for Data Science textbook draft by Garrett Grolemund and Hadley Wickham. Some chapters not complete yet but this is the best overall text I can recommend at this time.
• Jenny Bryan's materials for STAT545 at UBC. My class is an abbreviated rip-off of her class, so more explanations can be found on her page.
• Instructions for using RStudio in the CSSCR lab (hopefully week 1 only)

1. Slides for week 1: Course logistics, R/RStudio, making R Markdown documents
• Example HW #1 Rmd solution, Example HW #1 knitted HTML
• R code in week 1 slides
• Get R
• Get RStudio
• RMarkdown documentation
• HTML document options (global formatting, etc.)
• PDF document options (requires LaTeX installation to output PDFs)
• Word document options (but please do not use Word output for this class!)
• Useful RStudio cheatsheets on R Markdown, RStudio shortcuts, etc.
• Pander documentation for making tables, etc.
• Shapes and line types and color names (PDF) in base R
2. Slides for week 2: plotting with ggplot2
• R code in week 2 slides
• HW #2 student example plots (html), (Rmd)
• ggplot2 documentation
• Cookbook for R graph reference
• R graph catalog at UBC
• ggthemes package
• cowplot package for publication ready graphs, multiple plots in single image, etc.
• gganimate for easy animations (saving GIFs requires ImageMagick or GraphicsMagick)
• Hadley Wickham on the grammar of graphics
• Tufte in R (if that's your sort of thing)
• Caption demonstration with gridExtra or R Markdown
3. Slides for week 3: manipulating and summarizing data with dplyr
• R code in week 3 slides
• HW #3 student example analyses (html), (Rmd)
• A cautionary tale about Excel
• dplyr stuff:
• dplyr cheatsheets with diagrams to help you remember functions
• Introduction to dplyr
• Window functions in dplyr
• Joining data in dplyr
• dplyr 0.4.3.9 update coming soon that may affect your code! Note behavior of distinct() will change to only keep variables mentioned. Useful new function case_when() will be added to make nested conditionals simpler.
• More advanced joins: sqldf for easy SQL in R
• nycflights13 documentation
4. Slides for week 4: R data structures
• R code in week 4 slides
• Setting up swirl for practice
• Homework:
• HW #4 solution key (html), (Rmd)
• Homework 4: R Markdown template (you will download this, fill in and submit on Canvas)
• Wikipedia reference on linear regression for more background on HW
5. Slides for week 5: data import, export, and cleaning with readr, tidyr, and lubridate
• Homework due in 2 weeks:
• Homework 5: R Markdown template (you will download this, fill in and submit on Canvas)
• 2012 general election voting data for King County (60 MB download)
• HW #5 solution key (html), (Rmd)
• Data in-class:
• R code in week 5 slides
• Billboard Hot 100 data from 2000
• One day of Seattle Police Department incident data
• Data import and export:
• readr documentation
• Column types in readr
• Using dput when asking for help
• readxl and openxlsx packages for Excel
• General data access and cleaning:
• New York Times article on "data janitor" work
• Quartz guide to bad data: a must read!
• Lots of resources on survey data sources and analysis in R
• rOpenSci (many packages for accessing particular data sources in R)
• Qualtrics API package and Rmonkey for Survey Monkey
• Tidying:
• tidyr vignette
• Tidy genomics (a walkthough of tidy data preparation and analysis)
• Dates and times:
• lubridate vignette
• Factors:
• Lots on factors from Jenny Bryan
6. Slides for week 6: looping with for loops
• R code in week 6 slides
• Homework: read this demonstration on using loops to download and import data, and how to fix errors along the way. I suggest you try to follow along at home with the code in the associated .Rmd file (though you will need to change some filepaths to reference directories on your own computer).
7. Slides for week 7: vectorization, functions, and function-centric ways to loop
• R code in week 7 slides
• Homework due in 2 weeks:
• Homework 6: R Markdown template (you will download this, fill in and submit on Canvas), HTML version
• Pronto! bike share data from fall 2014 through fall 2015
• HW #6 solution key (html), (Rmd)
• The R Inferno by Patrick Burns [PDF]: "Circles" 2, 3, and 4 are relevant after this week's material, and Circle 8 covers a lot of miscellaneous R weird things that may trip you up.
• Reference material on writing functions with lots of examples
• Code style guide for writing functions, etc.
• R, the master troll of statistical languages (to read if you feel a bit frustrated!)
8. Slides for week 8: functions to work with strings and character data
• R code in week 8 slides
• Seattle restaurant inspection data since 2012 (CSV, about 16 MB) from King County
• stringr vignette
• Site for regular expression testing with a good cheatsheet and hover explanations
• Blog post explaining paste for combining strings
9. Slides for week 9: mapping and labels in ggplot2
• R code in week 9 slides
• Lab in-class/homework due next week:
• HW #7 solution key (html), (Rmd)
• Homework 7: R Markdown template (you will download this, fill in and submit on Canvas), HTML version
• Seattle restaurant inspection data since 2012 (CSV, about 16 MB) from King County
• ggmap package examples
• More in depth ggmap examples
• ggrepel package vignette
10. Slides for week 10: web scraping and text mining
• R code in week 10 slides
• rvest for harvesting web data:
• SelectorGadget for getting tags
• Good demo on State of the Union speeches by Jerid Francom
• Another demo pulling Superbowl scores by David Radcliffe
• tm package for text mining:
• tm vignette
• Slides by Yanchang Zhao on tm and Twitter data
• tidytext for tidy text analysis
• quanteda package for another set of tools
• Social media data extraction tools:
• twitteR package for accessing Twitter in R
• Setting up API keys and secrets
• twitteR functions
• streamR for the streaming Twitter API
• Rfacebook
• Shiny for interactive R apps