Introduction

Basic Explanation of `ggmosaic`

designed to create visualizations of categorical data
can produce bar charts, stacked bar charts, mosaic plots, and double decker plots
plots are constructed hierarchically, so the ordering of the variables is very important.
integrated in ggplot2 as a geom
allows for ggplot2 tools, such as facetting and layering

Creation of `ggmosaic`

ggmosaic was created primarily using ggproto and the productplots package

ggproto allows you to extend ggplot2 from within your own packages

ggmosaic began as a geom extension of the rect geom
used the data handling provided in the productplots package
calculates xmin, xmax, ymin, and ymax for the rect geom to plot

ggplot2 limitations

ggplot2 is not capable of handling a variable number of variables

current solution: read in the variables x1 and x2 as x = product(x1, x2)
product function:
a wrapper function for a list
allows for it to pass check_aesthetics

These limitations also lead to issues with the labeling, but those can be manually fixed.

geom_mosaic: setting the aesthetics

Aesthetics that can be set:

weight : select a weighting variable
x : select variables to add to formula
- declared as x = product(x1, x2, …)
fill : select a variable to be filled
- if the variable is not also called in x, it will be added to the formula in the first position
conds : select a variable to condition on

These values are then sent through productplots functions to create the formula for the desired distribution

Formula: weight ~ fill + x | conds

From the aesthetics to the formula

Example of how the formula is built

weight = 1
x = product(Y, X)
fill = W
conds = Z

These aesthetics set up the formula for the distribution:

Formula: 1 ~ W + Y + X | Z

Because a mosaic plot is constructed hierarchically through alternating spines, the ordering of the variables is very important.

Weight ~ X

 ggplot(data = NHANES) +
   geom_mosaic(aes(weight = Weight, x = product(SleepHrsNight), fill=factor(SleepHrsNight)), na.rm=TRUE) +
   labs(x="Hours of sleep a night ", title='f(SleepHrsNight)') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

Weight ~ Y + X

 ggplot(data = NHANES) +
   geom_mosaic(aes(weight = Weight, x = product(SleepHrsNight, AgeDecade), fill=factor(SleepHrsNight)), na.rm=TRUE) +    theme(axis.text.x=element_text(angle=-25, hjust= .1)) + labs(x="Age in Decades ", title='f(SleepHrsNight | AgeDecade) f(AgeDecade)') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

Weight ~ X + Y / Z

 ggplot(data = NHANES) +
   geom_mosaic(aes( x = product(SleepHrsNight, AgeDecade), fill=factor(SleepHrsNight), conds=product(Gender)), na.rm=TRUE, divider=mosaic("v")) +    theme(axis.text.x=element_text(angle=-25, hjust= .1)) + labs(x="Age in Decades ", title='f(SleepHrsNight, AgeDecade | Gender)') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

Alternative to conditioning: facetting

ggplot(data = NHANES) +
   geom_mosaic(aes( x = product(SleepHrsNight, AgeDecade), fill=factor(SleepHrsNight)), na.rm=TRUE) +    theme(axis.text.x=element_text(angle=-25, hjust= .1)) +
   labs(x="Age in Decades ", title='f(SleepHrsNight, AgeDecade | Gender)')  + facet_grid(Gender~.) + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

Importance of ordering

order1 <- ggplot(data = NHANES) + geom_mosaic(aes(weight = Weight, x = product(SleepHrsNight, Gender), fill=factor(SleepHrsNight)), na.rm=TRUE, offset=0.015) + labs(x="Gender ", title='f(SleepHrsNight | Gender)  f(Gender)') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE)) + theme(plot.title = element_text(size = rel(1)))

order2<- ggplot(data = NHANES) + geom_mosaic(aes(weight = Weight, x = product(Gender, SleepHrsNight), fill=factor(SleepHrsNight)), na.rm=TRUE, offset=0.015) + labs(x="", y="Gender", title='f(Gender | SleepHrsNight)  f(SleepHrsNight)') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE)) + theme(plot.title = element_text(size = rel(1))) + coord_flip()

grid_arrange_shared_legend(order1, order2, ncol = 2, nrow = 1, position = "right")

Other features of geom_mosaic

Arguments unique to geom_mosaic:

divider: used to declare the type of partitions to be used
offset: sets the space between the first spine

Divider function: Types of partitioning

Four options available for each partion:

vspine: width constant, height varies.
hspine: height constant, width varies.
vbar: height constant, width varies.
hbar: width constant, height varies.

a2 <- ggplot(data = NHANES) +
   geom_mosaic(aes( x = product(SleepHrsNight), fill=factor(SleepHrsNight)), divider="hbar", na.rm=TRUE) + theme(#axis.text.x=element_text(angle=35, hjust= 1),
     legend.position="none")+labs(x=" ", title='divider= "hbar"') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))


a1 <- ggplot(data = NHANES) +
   geom_mosaic(aes( x = product(SleepHrsNight), fill=factor(SleepHrsNight)), divider="hspine", na.rm=TRUE) + theme(#axis.text.x=element_text(angle=35, hjust= 1),
     legend.position="none") + labs(x=" ", title='divider= "hspine"') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))


b2 <- ggplot(data = NHANES) +
   geom_mosaic(aes( x = product(SleepHrsNight), fill=factor(SleepHrsNight)), divider="vbar", na.rm=TRUE) + theme(legend.position="none") + labs(y=" ", x="", title='divider= "vbar"') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))
 
b1 <- ggplot(data = NHANES) +
   geom_mosaic(aes(  x = product(SleepHrsNight), fill=factor(SleepHrsNight)), divider="vspine", na.rm=TRUE) + theme(legend.position="none") + labs(y=" ", x="", title='divider= "vspine"') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

grid_arrange_shared_legend(a1, a2, b1, b2, ncol = 2, nrow = 2, position = "right")

Partitioning with one or more variables

mosaic()
- default
- will use spines in alternating directions
- begins with a horizontal spine
mosaic(“v”)
- begins with a vertical spine and then alternates
ddecker()
- selects n-1 horizontal spines and ends with a vertical spine
Define each type of partition
- c(“hspine”, “vspine”, “hbar”)

#set.separators(c(":", ";","|"))

m1 <-ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=mosaic("h")) +
   theme(axis.text.x=element_blank(), legend.position="none")+labs(x=" ", title='divider= mosaic()') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

m2 <-ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=mosaic("v")) +
   theme(axis.text.x=element_blank(), legend.position="none")+labs(x=" ", title='divider= mosaic("v")') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))


m3 <-ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=ddecker()) +
   theme(axis.text.x=element_blank(), legend.position="none")+labs(x=" ", title='divider= ddecker()') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

grid_arrange_shared_legend(m1, m2, m3, ncol = 3, nrow = 1, position = "right")

m4 <-  ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=c("vspine", "vspine", "hbar")) +
   theme(axis.text.x=element_blank(), legend.position="none")+labs(x=" ", title='divider= c("vspine", "vspine", "hbar")') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

m5 <-  ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=c("hbar", "vspine", "hbar")) +
   theme(axis.text.x=element_blank(), legend.position="none")+labs(x=" ", title='divider= c("hbar", "vspine", "hbar")') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

m6 <-  ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=c("hspine", "hspine", "hspine")) +
   theme(axis.text.x=element_blank(),
legend.position="none")+labs(x=" ", title='divider= c("hspine", "hspine", "hspine")') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

m7 <-  ggplot(data = NHANES) + geom_mosaic(aes(x=product(SleepHrsNight, Gender, AgeDecade), fill = factor(SleepHrsNight)), na.rm=T, divider=c("vspine", "vspine", "vspine")) +
   theme(axis.text.x=element_blank(), legend.position="none")+labs(x=" ", title='divider= c("vspine", "vspine", "vspine")') + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

grid_arrange_shared_legend(m4, m5, m6, m7, ncol = 2, nrow = 2, position="right")

geom_mosaic: offset

offset: Set the space between the first spine

default = 0.01
space between partitions decreases as layers build

Adjusting the offset

oo1 <- ggplot(data = NHANES) +
  geom_mosaic(aes(weight = Weight,  x = product(Age), fill=factor(SleepHrsNight)), na.rm=TRUE) +   theme(axis.text.x=element_text(angle=0, hjust= .5))+labs(x="Age", y=" ",  title=" offset = 0.01") + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE)) 

o1 <- ggplot(data = happy) +
  geom_mosaic(aes(weight = wtssall,  x = product(age), fill=marital)) +
  theme(axis.text.x=element_text(angle=0, hjust= .5))+labs(x="Age", y=" ",  title=" offset = 0.01")+ guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE)) 

oo2 <- ggplot(data = NHANES) +
  geom_mosaic(aes(weight = Weight,  x = product(Age), fill=factor(SleepHrsNight)), offset=0, na.rm=TRUE) +   theme(axis.text.x=element_text(angle=0, hjust= .5))+labs(x="Age", y=" ",  title=" offset = 0")+ guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE)) 

o2 <- ggplot(data = happy) +
  geom_mosaic(aes(weight = wtssall,  x = product(age), fill=marital), offset = 0) +
  theme(axis.text.x=element_text(angle=0, hjust= .5))+labs(x="Age", y=" ",  title=" offset = 0") + guides(fill=guide_legend(title = "SleepHrsNight", reverse = TRUE))

grid_arrange_shared_legend(oo1, oo2, nrow = 1, ncol =2, position="right")

Plotly

gg <- ggplot(data = NHANES) + geom_mosaic(aes(x = product(SleepHrsNight, Gender, 
    AgeDecade), fill = factor(SleepHrsNight)), na.rm = T, divider = ddecker(), 
    offset = 0.025)

# just for now commented out ggplotly(gg)

Mosaic plots with `ggplot2`

Haley Jeppson and Heike Hofmann

2017-02-08

Introduction

Basic Explanation of `ggmosaic`

Creation of `ggmosaic`

ggplot2 limitations

geom_mosaic: setting the aesthetics

From the aesthetics to the formula

Weight ~ X

Weight ~ Y + X

Weight ~ X + Y / Z

Alternative to conditioning: facetting

Importance of ordering

Other features of geom_mosaic

Divider function: Types of partitioning

Partitioning with one or more variables

geom_mosaic: offset

Adjusting the offset

Plotly

Mosaic plots with ggplot2

Haley Jeppson and Heike Hofmann

2017-02-08

Introduction

Basic Explanation of ggmosaic

Creation of ggmosaic

ggplot2 limitations

geom_mosaic: setting the aesthetics

From the aesthetics to the formula

Weight ~ X

Weight ~ Y + X

Weight ~ X + Y / Z

Alternative to conditioning: facetting

Importance of ordering

Other features of geom_mosaic

Divider function: Types of partitioning

Partitioning with one or more variables

geom_mosaic: offset

Adjusting the offset

Plotly

Mosaic plots with `ggplot2`

Basic Explanation of `ggmosaic`

Creation of `ggmosaic`