Data Visualization Using R

Data visualization is instrumental to convey ideas, trends, and relationships and can be used to help make informed decisions since it transforms data into information that can be acted upon.  Within today's computing environment, creating the desired visualization can be accomplished by almost anyone once they know the rules and tools to do so.  The R programing language (which is free and has an enormous number of libraries that add specialized statistical analysis and visualization tools in almost every scientific and engineering field) is used in this hands-on course to build a wide range of scientific and technical graphics.  The student is expected to bring their own lap-top with R and R-studio already installed; instructions how to do this will be provided prior the course. 

In this three-day course, no R programing experience is required but some experience with a programing language or Excel is helpful.

At the end of this course the student will be able to select and construct the appropriate STEM-H data visualization based on the type of data and the needs of the graphic.  This will be done using the R programing language through a combination of discussion, demonstration, and practice.  Specific topics covered include:

• Data types and appropriate representation for each
• "Rules" for good graphics
• Simplified introduction to R
• Construction of different types of graphics using R

1. Overview
   1. Types of data – categorical, continuous
   2. Purposes of graphics – evaluation, comparison, trends
   3. Essentials of good graphics
2. Fundamentals of R programing
   1. R's data types
   2. R syntax simplified
   3. Data input – reading *.csv and *.xlxs files
   4. Data wrangling
   5. Data output – writing to *.pdf and *.txt files
   6. R packages (libraries)
3. General types of data graphics
   1. Distributions
   2. Ranking
   3. Correlation
   4. Part of a whole
   5. Evolution
   6. Flow
   7. Mapping
4. Controlling the "look" of R graphics
   1. Colors, sizes, shapes
   2. Multiple graphics in one window
   3. Adding text:  legends and descriptions
   4. The "par" function
5. Getting graphic I – constructing traditional R graphs
   1. Distributions:  2- and 3-D histograms, density, box and whisker plots
   2. Ranking: 2- and 3-D bar plots, spider plots
   3. Correlation: 2- and 3-D scatter (line and point) plots and bubble plots
6. Getting graphic II – constructing advanced R graphics
   1. Distributions: violin plots
   2. Correlations: heatmaps, correlograms
   3. Part of a whole: tree maps, pie charts, dendrograms
   4. Flow: Sankey, arc diagram, chord diagram, edge bundling

Stuart Munson-McGee, Emeritus Professor, is a retired chemical and food engineering professor from New Mexico State University where he taught both graduate and undergraduate courses for nearly 30 years.  His PhD from the University of Delaware is in chemical engineering. He has an extensive background in materials science and experimental statistics.