R Workshop

Module 1: Introduction to R

2018-03-14
Bobae Kang
(Bobae.Kang@illinois.gov)

This workshop will help you to get started and provide you with the basic skills and techniques in using R for research and data analysis.

Ultimately, this workshop seeks to help you to gain the knowledge and confidence necesary to learn what they need to know for you own research projects.

Source: pixabay.com

• Download datasets (to the Download folder)
• Examine each dataset in MS Excel
• Copy + paste data into a single spreadsheet
• Open the combined dataset in SPSS
• Run a regression analysis in SPSS
• Write a report in MS Word
• Submit the report

• Set up a directory
• Write an R program to download datasets
• Write an R program to combine and clean datasets
• Write an R program to run regression analysis and draw plots
• Write an R program to generate a report
• Submit the report

• Automation
• Modularity
• Reproducibility
• Version control

• Implementing the research work in programs that will run later to automatically execute the work
• Producing consistent results

In software design, modularity refers to a logical partitioning of the “software design” that allows complex software to be manageable for the purpose of implementation and maintenance.
- “Modularity”, Wikipedia

• Breaking down different stages or steps of research work into smaller but meaningful parts
• Separate programms for separate tasks
• Writing custom functions

Reproducibility refers to the ability of a researcher to duplicate the results of a prior study using the same materials and procedures as were used by the original investigator. […] Reproducibility is a minimum necessary condition for a finding to be believable and informative.
- U.S. NSF Subcommittee on Replicability in Science

• Greater productivity in a collaborative project

• The practice of managing changes in a document or a program in a systematic fashion
• Protecting the work from (unintentional) corruptions
• An example of version control system: Git

[1] "Hello World!"

Source: r-project.org

“R is a language and environment for statistical computing and graphics.”
- The R Foundation

• Built for data analysis and visualization
• One of the the most popular choices of programming language among academic researchers and data scientists

Source: flickr.com

• Open source (free!)
• Built for statistical analysis
• Reproducible and transparent
• Extensible through powerful third-party libraries
• Enabling researchers to tackle a variety of tasks using a single platform

Source: Wikimedia Commons

• License cost
• Speed and scalability
• Visualization
• Complex and advanced analysis
• Reproducibility

• License cost (again)
• Syntax
• Visualization
• Reporting

• License cost (DUH!)
• Reproducibility
• Data manipulation
• Complex and advanced analysis

• No intention to degrade other tools
  • Ease of use for intended tasks
• R as a great addition to any researcher's toolbox
  • Highly performant, versatile, and flexible

• Best Integrated Development Environment (IDE) for R
• Powerful and convenient features
• Interactive workflow
• Open source (free again!)
• … and many more!

Source: Wikimedia.org

• Visit https://cran.r-project.org/

• Or simply google “download R” to find the link to download page

• Installation requires the Administrator account

  • Talk to DoIT!

• Visit https://www.rstudio.com/products/rstudio/download/

• Or simply google “download Rstudio” to find the link to download page

• Again, installation requires the Administrator account

  • Talk to DoIT!

Source: Wikimedia Commons

R basics

• Part 1. Fundamentals of R programming.
  • R objects, expressions, functions, environments
• Part 2: Gearing up for data analysis
  • tidyverse framework
  • Recommended R style guide

Data analysis with R

• Part 1. Getting started with tidyverse
  • Manipulating data with dplyr
  • Tidying up data with tidyr
• Part 2. More on data analysis
  • character strings
  • date/datetimes
  • Importing/exporting data

Data visualization with R

• Part 1. The Grammar of Graphics
  • ggplot2 package
• Part 2. Maps and interactive plots
  • Pacakges for maps
  • Pacakges for interactive plots

Statistical modeling with R

• Part 1. Basics of statistical modeling
  • Descriptive statistics
  • Linear modelss and generalized linear models
• Part 2. Options for advanced modeling
  • Survival analysis
  • Time series analysis
  • Spatial regression analysis
  • Machine learning

“To Infinity and Beyond”

• Part 1. Sharing your work
  • R Markdown documents
  • Presentation slides
  • Shiny applications
  • Websites
• Part 2. Leveraging online resources
  • Various online resources

Source: tenor.com

• Bollen, K., Cacioppo, J., Kaplan, R., Krosnick, J. & Olds, J. (2015). “Social, Behaviroal, and Economic Sciences Perspectives on Robust and Reliable Science”.
• Gentzkow, M. & Shapiro, J. M. (2014). “Code and Data for the Social Sciences: A Practitioner’s Guide”.
• Robinson, D. (2017). “The Impressive Growth of R”. Stack Overflow Blog.
• Soltoff, B. (2018). “Introduction to computing for the social sciences”. Computing for the Social Sciences.
• Wikipedia articles on relevant topics.