• Important points are emphasized / annotated

• Axes, symbols, and colors are described

• Visual content clarifies (does not distract)

• Is accurate, clear, and improves understanding

• An “effective graph” communicates clearly

• Patterns

• Relationships \(\leadsto\) compare & contrast values

• Anomolies

• Focus / reduction of information

• What is the role of the following in data visualization:
  • Art / aesthetics?
  • Entertaining / engagement?
• We want to see the truth, to support some set of claims with data–but how can we do that if people can’t “see” it because of disinterest?
  • Hans Rosling on Third vs. Western Word
• How you present data is important … but entirely secondary to presenting data clearly

• Hide or obfuscate data
• Lack context, labeling, or description
• Are inaccurate or misleading
• Focus more on art, iconography, or technology than delivering content, ideas, and data

• Avoiding pie charts
• Avoiding 3D plot elements
• Complications with multivariate visualization
• Bad baselining examples
• Complexity of judging differences

• It is bad at doing what it is designed to do: Difficult to judge relative size of the pie slices

• Inefficient / inflexible use of space

• Need many colors and high contrast to make wedges distinct

• We’re much worse at estimating area than length — we’re especially bad at perceiving small differences in area

• Pie charts make judging trends difficult

3D effects make graphs harder to read

• Are we to judge length? Area? Volume?

• Display looks 3D when the angular perspective is offset, which makes referencing values on the axes harder

• Display looks 3D when shading is employed, which clutters the graph and makes it harder to read

When you have multiple variables to compare, there are several possibilities:

• Stacked bar plots
  • Efficient use of space & clean
  • Variables at bottom easier to compare than variables at top
• Separate bar plots
  • Good scaling control & clean
  • Extremely inefficient with space cross-variable comparisons can be difficult
• Grouped bar plots
  • Cross-variable comparisons are natural
  • Creates chart clutter and can be difficult to read

• Plotting 2D values using a scatter plot is easy

• If we have a categorical variable, we can sometimes use shading or color to add a third dimension

• But if we have another numeric dimension, it’s challenging

• Why not use point size (area)?

• People don’t judge small differences in area very well
• Using radius distorts values since the area increases with the square of the radius
• Double the radius means quadruple the area

• Items compared should have the same baseline for comparison

• That baseline should not distort the true data values

• Scaling should be set properly for comparison (apples-to-apples)

• Scaling should not distort the true data values

• Data should always be properly adjusted

• A notional concept by Edward Tufte that argues for keeping visualizations as simple as possible
• Idea is to maximize: \(\frac{ink\;required\;to\;represent\;actual\;data}{total\;ink\;used\;in\;the\;graphic}\)
• What is “data ink”? Anything that, were it erased, the underlying data would be removed (values, proportions, etc.)
• What is “non-data ink”? Anything that can be erased without damaging the underlying data (e.g., iconographic images, non-data shading, borders, etc.)
• “Above all else, data.” Tufte 1983
• It is a good rule of thumb, but shouldn’t be followed blindly

The ggplot2 library:

• uses a consistent grammar of graphics
• provides a high-level plots specification
• allows user to think in terms of a layers data visualization pipeline
• provides extensive visualization functionality for many common graphics
• is widely used

• Users specify building plots for a plot, then layer them as desired
• Building blocks include:
  • data sets
  • aesthetic mapping (what fields map to what visual elements)
  • geometry objects (how to visually encode those elements)
  • transformations, coordinate systems, and scaling mechanisms
  • thematic fine-tuning (fonts, annotations, positions adjustments)
  • faceting (multi-panel plotting)
• Each layer is added onto the old layer, making it intuitive and natural

• The ggplot() is the foundational function in the ggplot2 library
• It initializes a plotting object, setting up:
  • The data set to use
  • Which variables map to what plot elements (x, y, fill, size, etc.)
  • The core plot object
• The object returned by ggplot() is not a visualization by itself
• It is extended and interpretted with subsequent function calls for visualziation

• Most plot tools have functions that take explicit data types as arguments (e.g., the built-in plot() function takes R vectors for x and y)
• ggplot consumes R data frames, and it understands the concept of variables (fields) in a data frame
• ggplot understands the difference between numeric and categorical data and treats them differently
• You can flexibly map variables to whatever visual elements you prefer
• Data frames should be in long (statistical) form

Elements you choose to visualize may be set explicitly, or may be mapped to a variable using aes()

• x – Position along the x-axis
• y – Position along the y-axis
• size – Point width, line thickness, etc.
• linetype – Type of line (dotted, dashed, solid, etc.)
• color – Usually the color of the outer border of something
• fill – The color of the inner fill of some shape

Additional elements you choose to visualize may be set explicitly, or may be mapped to a variable using aes()

• shape – The symbol being used for some point
• linetype – The style of line objects (e.g., solid, dashed)
• alpha – The transparency of an object
• label – Text labels associated with an object

• Load the ggplot2 library
• Call ggplot(), providing:
  • The data frame to use
  • The variable mapping, aes()

library(ggplot2)
rd = data.frame(
  Student = c("Bob", "Sue", "Cat", "Lin"),
  NumberGrade = c(96, 82, 97, 74),
  LetterGrade = factor(c("A","B","A","C")) )

p = ggplot(rd, aes(y=NumberGrade))

ggplot2 interprets plot elements using geom objects:

• geom_point – Points
• geom_bar – Bars
• geom_line – Lines
• geom_smooth – Smoothed curves
• geom_polygon – Polygons
• geom_boxplot – Boxplot

• geom_density – Density plots
• geom_histogram – Histograms

There are other geomtetry objects: http://docs.ggplot2.org/current/

Aside from geom objects, there are other kinds of layers:

• Scales – Scaling controls for the mapping between data and aesthetics (scale_x_discrete(), scale_size_continuous(), etc. In general: scale_AESTHETIC_QUALIFER
• Coordinate systems – For transforming data to other coordinate systems (coord_flip(), coord_polar(), etc.)
• Faceting – Splitting up data into trellis displays (facet_grid(), facet_wrap(), etc.)
• Themes – Changing non-data elements of plot (element_text(), etc.)

http://docs.ggplot2.org/current/

• Use the “+” operator to add visualization layers to the plot object
• Layers are placed in the order that they are added, in a pipeline
• It’s easy to keep each layer separate and explicit

library(ggplot2)
rd = data.frame(Student = c("Bob", "Sue", "Cat", "Lin"),
                NumberGrade = c(96, 82, 97, 74),
                LetterGrade = factor(c("A","B","A","C")) )

ggplot(rd, aes(x=Student,y=NumberGrade)) +   # Build the plot object
    geom_point(size=5) +                     # Encode visually using points
    xlab("Student Name") +                   # Label the X axis
    ylab("Numeric Grade") +                  # Label the Y axis
    ggtitle("Course Grade Results")          # Give the plot a title

• The same basic plot object can be interpretted differently:

library(ggplot2)
rd = data.frame(Student = c("Bob", "Sue", "Cat", "Lin"),
                NumberGrade = c(96, 82, 97, 74),
                LetterGrade = factor(c("A","B","A","C")) )

ggplot(rd, aes(x=Student,y=NumberGrade)) +
    geom_bar(stat="identity") +              # Only line that changed...
    xlab("Student Name") + 
    ylab("Numeric Grade") + 
    ggtitle("Course Grade Results")

library(ggplot2)
rd = data.frame(Student = c("Bob", "Sue", "Cat", "Lin"),
                NumberGrade = c(96, 82, 97, 74),
                LetterGrade = factor(c("A","B","A","C")) )

ggplot(rd, aes(x=Student,y=NumberGrade)) +
    geom_bar(stat="identity") + 
    coord_flip() +
    xlab("Student Name") + 
    ylab("Numeric Grade") + 
    ggtitle("Course Grade Results")

• Other variables can be mapped to other plot features, as well

library(ggplot2)
rd = data.frame(Student = c("Bob", "Sue", "Cat", "Lin"),
                NumberGrade = c(96, 82, 97, 74),
                LetterGrade = factor(c("A","B","A","C")) )

ggplot(rd, aes(x=Student,y=NumberGrade,fill=LetterGrade)) +
    geom_bar(stat="identity") + 
    xlab("Student Name") + 
    ylab("Numeric Grade") + 
    ggtitle("Course Grade Results")

• In ggplot2:
  • color refers to the color of borders
  • fill refers to the color of the inside fill
• Except with the default point shape, which is actually a font – and so uses color only
• Unless you change the shape to a drawn shape (e.g., shape 21)

library(ggplot2)

myData = data.frame(Furbletude=rnorm(30),
                    Blehmekness=rnorm(30))

ggplot(myData, aes(x=Furbletude, y=Blehmekness)) +
  geom_point(color="lightblue", fill="darkblue", size=4)

library(ggplot2)

myData = data.frame(Furbletude=rnorm(30),
                    Blehmekness=rnorm(30))

ggplot(myData, aes(x=Furbletude, y=Blehmekness)) +
  geom_point(color="darkblue", fill="lightblue", size=4, shape=21)

• R has hundreds of prebuilt colors (type colors() at the console to list)
• But you can also specify custom colors in several ways, including:
  • In RGB hex via a string – e.g., “#992B1A”
  • Using the rgb() function – e.g., rgb(0.26, 0.52, 0.87)
  • Using the hsv() function – e.g., hsv(0.17, 0.98, 0.66)
• You can construct palettes as lists of these
• Or you can use prebuilt palletes

library(ggplot2)

myData = data.frame(Count=sample(1:10, 30, replace=T),
    Awesomeness=sample(c("CoolThings", "SillyThings", "Meh"), 30, replace=T))

ggplot(myData, aes(x=Awesomeness, y=Count)) +
  geom_bar(stat="identity", color="white", fill=rgb(0.12, 0.76, 0.9))

library(ggplot2)

myData = data.frame(Count=sample(1:10, 30, replace=T),
        Awesomeness=sample(c("CoolThings", "SillyThings", "Meh"), 30, replace=T),
        TypeOfThing=sample(c("A", "B", "C"), 30, replace=T))

ggplot(myData, aes(x=Awesomeness, y=Count, fill=TypeOfThing)) +
  geom_bar(stat="identity", color="black")

• In ggplot2, the scale_…() functions are used to override data-driven properties like colors
• Get used to using RColorBrewer
• It’s nice because it has some good pre-built discrete and continuous palettes
• So install it, if you have not already

library(ggplot2)
library(RColorBrewer)

myData = data.frame(Count=sample(1:10, 30, replace=T),
        Awesomeness=sample(c("CoolThings", "SillyThings", "Meh"), 30, replace=T),
        TypeOfThing=sample(c("A", "B", "C"), 30, replace=T))

ggplot(myData, aes(x=Awesomeness, y=Count, fill=TypeOfThing)) +
  geom_bar(stat="identity") +
  scale_fill_brewer(palette="Set2") +  # Set2 is a color-blind friendly palette
  theme_bw()  # Make the background white and the grid lines black

library(ggplot2)

ggplot(mtcars, aes(x=mpg,y=hp)) + 
  geom_smooth(size=1.5, color="darkgray") + 
  geom_point(aes(size=gear,color=cyl)) +
  xlab("Miles per Gallon") + 
  ylab("Horse Power")

library(ggplot2)

ggplot(mtcars, aes(x=mpg,y=hp)) + 
  geom_point(size=4, shape=21, fill="lightblue", color="darkblue") +  
  xlab("Miles per Gallon") + 
  ylab("Horse Power") +
  theme(text=element_text(size=18, family="Times"))

library(ggplot2)

ggplot(diamonds, aes(carat)) + 
  geom_histogram(binwidth=0.5, fill="wheat", color="black") +  
  xlab("Carat") + 
  ylab("Count") +
  ggtitle("Diamond Carat Distribution")
  theme(text=element_text(size=18, family="Times"))

• Harvard’s R Graphics Tutorial
• R-Statistics’ ggplot2 Tutorial
• DataCamp’s Visualization with ggplot2 course
• R CookBook’s Colors (ggplot2)