Skip to contents

Mosaic plots.

Source: R/geom-mosaic.r, R/stat-mosaic-text.R, R/stat-mosaic.r
geom_mosaic.Rd

A mosaic plot is a convenient graphical summary of the conditional distributions in a contingency table and is composed of spines in alternating directions.

Usage

geom_mosaic(
  mapping = NULL,
  data = NULL,
  stat = "mosaic",
  position = "identity",
  na.rm = FALSE,
  divider = mosaic(),
  offset = 0.01,
  show.legend = NA,
  inherit.aes = TRUE,
  expected = NULL,
  ...
)

stat_mosaic_text(
  mapping = NULL,
  data = NULL,
  geom = "Text",
  position = "identity",
  na.rm = FALSE,
  divider = mosaic(),
  show.legend = NA,
  inherit.aes = TRUE,
  offset = 0.01,
  expected = NULL,
  ...
)

stat_mosaic(
  mapping = NULL,
  data = NULL,
  geom = "mosaic",
  position = "identity",
  na.rm = FALSE,
  divider = mosaic(),
  show.legend = NA,
  inherit.aes = TRUE,
  offset = 0.01,
  expected = NULL,
  ...
)

Arguments

mapping

Set of aesthetic mappings created by aes(). If specified and inherit.aes = TRUE (the default), it is combined with the default mapping at the top level of the plot. You must supply mapping if there is no plot mapping.

data

The data to be displayed in this layer. There are three options:

If NULL, the default, the data is inherited from the plot data as specified in the call to ggplot().

A data.frame, or other object, will override the plot data. All objects will be fortified to produce a data frame. See fortify() for which variables will be created.

A function will be called with a single argument, the plot data. The return value must be a data.frame, and will be used as the layer data. A function can be created from a formula (e.g. ~ head(.x, 10)).

stat

The statistical transformation to use on the data for this layer. When using a geom_*() function to construct a layer, the stat argument can be used to override the default coupling between geoms and stats. The stat argument accepts the following:

  • A Stat ggproto subclass, for example StatCount.

  • A string naming the stat. To give the stat as a string, strip the function name of the stat_ prefix. For example, to use stat_count(), give the stat as "count".

  • For more information and other ways to specify the stat, see the layer stat documentation.

position

A position adjustment to use on the data for this layer. This can be used in various ways, including to prevent overplotting and improving the display. The position argument accepts the following:

  • The result of calling a position function, such as position_jitter(). This method allows for passing extra arguments to the position.

  • A string naming the position adjustment. To give the position as a string, strip the function name of the position_ prefix. For example, to use position_jitter(), give the position as "jitter".

  • For more information and other ways to specify the position, see the layer position documentation.

na.rm

If FALSE (the default), removes missing values with a warning. If TRUE silently removes missing values.

divider

Divider function. The default divider function is mosaic() which will use spines in alternating directions. The four options for partitioning:

  • vspine Vertical spine partition: width constant, height varies.

  • hspine Horizontal spine partition: height constant, width varies.

  • vbar Vertical bar partition: height constant, width varies.

  • hbar Horizontal bar partition: width constant, height varies.

When omitted, divider can be inherited from mosaic_settings().

offset

Set the fixed gap at the deepest split. Gaps increase by a factor of 1.5 toward the outermost split. When omitted, the value can be inherited from mosaic_settings().

show.legend

logical. Should this layer be included in the legends? NA, the default, includes if any aesthetics are mapped. FALSE never includes, and TRUE always includes. It can also be a named logical vector to finely select the aesthetics to display. To include legend keys for all levels, even when no data exists, use TRUE. If NA, all levels are shown in legend, but unobserved levels are omitted.

inherit.aes

If FALSE, overrides the default aesthetics, rather than combining with them. This is most useful for helper functions that define both data and aesthetics and shouldn't inherit behaviour from the default plot specification, e.g. annotation_borders().

expected

Optional specification for loglinear model residual shading. Can be a formula (e.g., ~ Var1 + Var2), a character shortcut ("independence", "saturated", "conditional"), or NULL (default, no model). When omitted, the value can be inherited from mosaic_settings. An explicitly supplied layer value takes priority; in particular, expected = NULL turns off a plot-level model for that layer. When specified, Pearson residuals are calculated and automatically mapped to fill (unless fill aesthetic is explicitly set). Use with scale_fill_residual for a diverging color scale. Positive residuals receive a solid dark blue outline and negative residuals a dashed dark red outline by default. Residuals within numerical tolerance of zero receive a solid black outline. Set colour = NA to remove the outlines from both the cells and the residual legend.

...

other arguments passed on to layer. These are often aesthetics, used to set an aesthetic to a fixed value, like color = 'red' or size = 3. They may also be parameters to the paired geom/stat.

geom

The geometric object to use to display the data for this layer. When using a stat_*() function to construct a layer, the geom argument can be used to override the default coupling between stats and geoms. The geom argument accepts the following:

  • A Geom ggproto subclass, for example GeomPoint.

  • A string naming the geom. To give the geom as a string, strip the function name of the geom_ prefix. For example, to use geom_point(), give the geom as "point".

  • For more information and other ways to specify the geom, see the layer geom documentation.

Value

A ggplot2 layer that can be added to a plot.

Details

Variables mapped only to fill or alpha retain their historical role as innermost mosaic partitions, but they are not shown on the automatic product axes. Position axes label only variables explicitly mapped through x or conds. If an aesthetic variable is also included in product(), it remains eligible for an axis label.

Product variables are ordered from innermost to outermost. With the default mosaic divider, reversing two variables swaps their horizontal and vertical roles; for example, product(predictions, actual) places actual on the primary x axis.

Computed variables

x

location of center of the rectangle

y

location of center of the rectangle

xmin

location of bottom left corner

xmax

location of bottom right corner

ymin

location of top left corner

ymax

location of top right corner

Author

Gavin Klorfine

Examples


data(titanic)

ggplot(data = titanic, aes(x = product(Class), fill = Survived)) +
  geom_mosaic()

# good practice: use the 'dependent' variable (or most important variable)
# as fill variable

# if there is only one variable inside `product()`,
# `product()` can be omitted
ggplot(data = titanic, aes(x = Class, fill = Survived)) +
  geom_mosaic()


ggplot(data = titanic,
       aes(x = product(Class, Age), fill = Survived)) +
  geom_mosaic()


ggplot(data = titanic,
       aes(x = product(Class), conds = product(Age), fill = Survived)) +
  geom_mosaic()


# if there is only one variable inside `product()`,
# `product()` can be omitted
ggplot(data = titanic, aes(x = Class, conds = Age, fill = Survived)) +
  geom_mosaic()


ggplot(data = titanic,
       aes(x = product(Survived, Class), fill = Age)) +
  geom_mosaic()


# Variables can be transformed directly inside mosaic aesthetics
ggplot(data = mtcars,
       aes(x = product(factor(gear)), fill = factor(cyl))) +
  geom_mosaic()


# A fill-only variable colours and partitions the tiles without appearing on
# a position axis. Reverse the product order to put `actual` on the x axis.
set.seed(19790801)
predictions <- sample(iris$Species)
confusion <- as.data.frame(table(actual = iris$Species, predictions))
confusion$is_correct <- ifelse(
  confusion$actual == confusion$predictions,
  "Correct prediction", "Incorrect prediction"
)
ggplot(confusion, aes(
    weight = Freq,
    x = product(predictions, actual),
    fill = is_correct
  )) +
  geom_mosaic()


# Wrapped in donttest: still runs under R CMD check --run-donttest and is
# exercised by the package's tests to make sure it works.
# \donttest{
data(happy)

ggplot(data = happy, aes(x = product(happy))) +
  geom_mosaic(divider = "hbar")


ggplot(data = happy, aes(x = product(happy))) +
  geom_mosaic() +
  coord_flip()


# weighting is important
ggplot(data = happy, aes(weight = wtssall, x = product(happy))) +
  geom_mosaic()


ggplot(data = happy,
       aes(weight = wtssall, x = product(health), fill = happy)) +
  geom_mosaic() +
  theme(axis.text.x=element_text(angle=35))


ggplot(data = happy,
       aes(weight = wtssall, x = product(health), fill = happy)) +
  geom_mosaic(na.rm = TRUE)


ggplot(data = happy,
       aes(weight = wtssall, x = product(health, sex, degree), fill = happy)) +
  geom_mosaic(na.rm = TRUE)


# here is where a bit more control over the spacing of the bars is helpful:
# set labels manually:
ggplot(data = happy,
       aes(weight = wtssall, x = product(age), fill = happy)) +
  geom_mosaic(na.rm = TRUE, offset = 0) +
  scale_x_productlist("Age", labels=c(17+1:72))


# thin out labels manually:
labels <- c(17+1:72)
labels[labels %% 5 != 0] <- ""
ggplot(data = happy,
       aes(weight = wtssall, x = product(age), fill = happy)) +
  geom_mosaic(na.rm = TRUE, offset = 0) +
  scale_x_productlist("Age", labels=labels)


ggplot(data = happy,
       aes(weight = wtssall, x = product(age), fill = happy,
           conds = product(sex))) +
  geom_mosaic(divider = mosaic("v"), na.rm = TRUE, offset = 0.001) +
  scale_x_productlist("Age", labels=labels)


ggplot(data = happy,
       aes(weight = wtssall, x = product(age), fill = happy)) +
  geom_mosaic(na.rm = TRUE, offset = 0) +
  facet_grid(sex~.) +
  scale_x_productlist("Age", labels=labels)


ggplot(data = happy,
       aes(weight = wtssall, x = product(happy, finrela, health))) +
  geom_mosaic(divider = mosaic("h"))


ggplot(data = happy,
       aes(weight = wtssall, x = product(happy, finrela, health))) +
  geom_mosaic(offset = .005)


# Spine example
ggplot(data = happy,
       aes(weight = wtssall, x = product(health), fill = health)) +
 geom_mosaic() +
 facet_grid(happy~.)


# Residual shading with independence model
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  geom_mosaic(expected = "independence") +
  scale_fill_residual()


# Custom model formula
ggplot(data = titanic, aes(x = product(Class, Sex, Survived))) +
  geom_mosaic(expected = ~ Class + Sex) +
  scale_fill_residual()

# } # end of donttest