In the first part of our textbook "Whole Game" we took a quick survey the basic
tools of data science. Now we will take a deeper look at "Visualization" in
Part II.
This week in chapter 9 we look at the formal layered grammar of graphics.
How much information can we bring to a graphic?
Remember, aesthetics are the visual properties of a plot: position,
color, shape, ... Our data sets have many different types of values: double
(floating-point decimal), integer, strings, dates, factors, ... There are many
other visual properties and value types than we are listing here.
A good visualization is not about including the maximum amount of data possible,
but rather using different visual properties to communicate or investigate a
relationship among variables. We want to tell a story.
It is not reasonable to think that one visualization can tell a complete story
and it is also not reasonable to think that there is a relationship or story to
tell between any two variables.
The layered grammar of graphics (from whence ggplot gets its name)
provides us with functions called geoms that we layer (include
in the same visualization) to illustrate or investigate relationships between
variables.
It would be impossible to show you all the geom functions available to you and all
of the combinations of functions, attributes, and values that we could use, but
the text is going to give you a broad sample of what is available and show you
where you could look for more later.
Assessment deadlines will be 11:59pm each Saturday.
All assessments are submitted to the Homework Folder inside your assigned
Google Drive folder.
There are no make-ups for missed assessments. Contact me before a deadline
if you have an issue meeting the deadline and we will find a mutually
agreeable solution.
Homework
Homework 5 (due Saturday, February 8)
The instructions for your homework are contained in the R script
file
homework_05.R