Package {ggmosaic2}

Title:	Mosaic Plots in the 'ggplot2' Framework, Extended
Version:	0.5.1
Date:	2026-09-15
Description:	Mosaic plots in the 'ggplot2' framework. Mosaic plot functionality is provided in a single 'ggplot2' layer by calling the geom 'mosaic'. This extends the now-defunct 'ggmosaic' package by allowing fitted loglinear models which can be used to show patterns of association among variables in frequency tables.
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
Language:	en-US
URL:	https://friendly.github.io/ggmosaic2/, https://github.com/friendly/ggmosaic2
BugReports:	https://github.com/friendly/ggmosaic2/issues
Depends:	ggplot2 (≥ 4.0.0), R (≥ 4.1.0)
Imports:	productplots, dplyr, plotly (≥ 4.9.3), purrr, rlang, tidyr, ggrepel, scales, withr (≥ 2.5.0)
Suggests:	gridExtra, knitr, rmarkdown, shiny (≥ 1.1.0), patchwork, vcdExtra
VignetteBuilder:	knitr
Encoding:	UTF-8
LazyData:	true
Config/roxygen2/version:	8.1.0
NeedsCompilation:	no
Packaged:	2026-09-15 14:27:34 UTC; friendly
Author:	Michael Friendly [aut, cre, cph], Gavin Klorfine [aut], Haley Jeppson [aut], Heike Hofmann [aut], Di Cook [aut], Hadley Wickham [ctb]
Maintainer:	Michael Friendly <friendly@yorku.ca>
Repository:	CRAN
Date/Publication:	2026-09-16 12:30:02 UTC

ggmosaic2: Mosaic Plots in the 'ggplot2' Framework, Extended

Description

Mosaic plots in the 'ggplot2' framework. Mosaic plot functionality is provided in a single 'ggplot2' layer by calling the geom 'mosaic'. This extends the now-defunct 'ggmosaic' package by allowing fitted loglinear models which can be used to show patterns of association among variables in frequency tables.

Author(s)

Maintainer: Michael Friendly friendly@yorku.ca (ORCID) [copyright holder]

Authors:

• Michael Friendly friendly@yorku.ca (ORCID) [copyright holder]

• Gavin Klorfine gklorfin@yorku.ca (ORCID)

• Haley Jeppson hjeppson@iastate.edu (ORCID)

• Heike Hofmann hofmann@iastate.edu (ORCID)

• Di Cook dicook@monash.edu (ORCID)

Other contributors:

• Hadley Wickham hadley@posit.co (ORCID) [contributor]

See Also

Useful links:

• https://friendly.github.io/ggmosaic2/

• https://github.com/friendly/ggmosaic2

• Report bugs at https://github.com/friendly/ggmosaic2/issues

Geom proto

Description

Geom proto

Value

A ggproto object inheriting from ggplot2::Geom, used internally by geom_mosaic() to draw the mosaic tile rectangles. Not intended to be called directly.

Geom proto

Description

Geom proto

Value

A ggproto object inheriting from ggplot2::Geom, used internally by geom_mosaic_jitter() to draw jittered points within mosaic plot cells. Not intended to be called directly.

Geom proto

Description

Geom proto

Value

A ggproto object inheriting from ggplot2::Geom, used internally by geom_mosaic_text() to draw text/label annotations on mosaic plot cells. Not intended to be called directly.

Geom proto

Description

Geom proto

Value

A ggproto object inheriting from ggplot2::Stat, used internally by stat_mosaic() (and geom_mosaic()) to compute the rectangle boundaries and cell statistics for a mosaic plot layer. Not intended to be called directly.

Geom proto

Description

Geom proto

Value

A ggproto object inheriting from ggplot2::Stat, used internally by stat_mosaic_jitter() (and geom_mosaic_jitter()) to compute jittered point positions for a mosaic plot layer. Not intended to be called directly.

Geom proto

Description

Geom proto

Value

A ggproto object inheriting from ggplot2::Stat, used internally by stat_mosaic_text() (and geom_mosaic_text()) to compute label positions for a mosaic plot layer. Not intended to be called directly.

Build model formula from user specification

Description

Build model formula from user specification

Usage

build_model_formula(expected, vars, conds = NULL, variable_labels = NULL)

Arguments

Value

Formula object or NULL

Template for a double decker plot. A double decker plot is composed of a sequence of spines in the same direction, with the final spine in the opposite direction.

Description

Template for a double decker plot. A double decker plot is composed of a sequence of spines in the same direction, with the final spine in the opposite direction.

Usage

ddecker(direction = "h")

Arguments

Value

A function of one argument, the number of splits n, that returns a character vector of divider function names ("hspine"/"vspine") to apply at each split – suitable for the divider argument of geom_mosaic() and related layers.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  geom_mosaic(divider = ddecker("v"))

Facet mosaic plots with panel-specific axes

Description

facet_mosaic_grid() lays panels out like ggplot2::facet_grid(), but gives every panel its own x and y position scales. Mosaic category breaks depend on the proportions calculated inside a panel, so sharing a position scale can place one panel's ticks and labels on another panel's mosaic.

Usage

facet_mosaic_grid(
  rows = NULL,
  cols = NULL,
  space = "fixed",
  shrink = TRUE,
  labeller = "label_value",
  as.table = TRUE,
  switch = NULL,
  drop = TRUE,
  margins = FALSE,
  axes = "all",
  axis.labels = "all"
)

Arguments

Details

Panel widths and heights remain fixed. By default, axes and their labels are drawn for every panel so that each panel's category positions are visible. The panel-specific label positions are retained with theme_mosaic(); that theme intentionally hides the tick marks themselves.

Value

A Facet object that can be added to a ggplot.

Author(s)

Gavin Klorfine

Examples

facet_data <- expand.grid(
  period = factor(c("Before", "After")),
  group = factor(c("A", "B")),
  response = factor(c("No", "Yes"))
)
facet_data$n <- c(30, 10, 20, 40, 10, 35, 40, 15)

ggplot(facet_data,
       aes(weight = n, x = product(response, group), fill = response)) +
  geom_mosaic() +
  facet_mosaic_grid(cols = vars(period))

Fit Poisson GLM and calculate Pearson residuals

Description

Fit Poisson GLM and calculate Pearson residuals

Usage

fit_loglinear_model(data, vars, model_formula)

Arguments

Value

Data frame with added .expected and .residual columns

Flying Etiquette Survey Data

Description

Data from the results of a SurveyMonkey survey commissioned by FiveThirtyEight for the story 41 Percent of Fliers Say It’s Rude To Recline Your Airplane Seat.

Usage

fly

Format

A data frame with 1040 rows and 27 variables:

id

Respondent ID

flight_freq

How often do you travel by plane?

do_you_recline

Do you ever recline your seat when you fly?

height

How tall are you?

has_child_under_18

Do you have any children under 18?

three_seats_two_arms

n a row of three seats, who should get to use the two arm rests?

two_seats_one_arm

In a row of two seats, who should get to use the middle arm rest?

window_shade

Who should have control over the window shade?

rude_to_move_to_unsold_seat

Is it rude to move to an unsold seat on a plane?

rude_to_talk_to_neighbor

Generally speaking, is it rude to say more than a few words to the stranger sitting next to you on a plane?

six_hr_flight_leave_seat

On a six hour flight from NYC to LA, how many times is it acceptable to get up if you're not in an aisle seat?

reclining_obligation_to_behind

Under normal circumstances, does a person who reclines their seat during a flight have any obligation to the person sitting behind them?

rude_to_recline

Is it rude to recline your seat on a plane?

eliminate_reclining

Given the opportunity, would you eliminate the possibility of reclining seats on planes entirely?

rude_to_switch_seats_friends

Is it rude to ask someone to switch seats with you in order to be closer to friends?

rude_to_switch_seats_family

Is it rude to ask someone to switch seats with you in order to be closer to family?

rude_to_wake_neighbor_bathroom

Is it rude to wake a passenger up if you are trying to go to the bathroom?

rude_to_wake_neighbor_walk

Is it rude to wake a passenger up if you are trying to walk around?

rude_to_bring_baby

In general, is it rude to bring a baby on a plane?

rude_to_bring_unruly_child

In general, is it rude to knowingly bring unruly children on a plane?

use_electronics_takeoff

Have you ever used personal electronics during take off or landing in violation of a flight attendant's direction?

smoked_inflight

Have you ever smoked a cigarette in an airplane bathroom when it was against the rules?

gender

Gender

age

Age

household_income

Household Income

education

Education

region

Region

Source

https://github.com/fivethirtyeight/data/tree/master/flying-etiquette-survey

Mosaic plots.

Description

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

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.

Value

A ggplot2 layer that can be added to a plot.

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(s)

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.

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

Jittered dots in Mosaic plots.

Description

A mosaic plat with jittered dots

Usage

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

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

Arguments

Details

Variables mapped only to fill, alpha, or colour remain available to the mosaic calculation and point aesthetics, but are not shown on automatic product axes. Position axes label only variables explicitly mapped through x or conds.

Value

A ggplot2 layer that can be added to a plot.

Computed variables

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(s)

Gavin Klorfine

Examples

data(titanic)

ggplot(data = titanic, aes(x = product(Class))) +
  geom_mosaic(aes(fill = Survived), alpha = 0.3) +
  geom_mosaic_jitter(aes(color = Survived))

ggplot(data = titanic, aes(x = product(Class))) +
  geom_mosaic(alpha = 0.1) +
  geom_mosaic_jitter(aes(color = Survived), drop_level = TRUE)

ggplot(data = titanic, aes(x = product(Class, Sex))) +
  mosaic_settings(divider = c("vspine", "hspine", "hspine")) +
  geom_mosaic(alpha = 0.3, aes(fill = Survived)) +
  geom_mosaic_jitter(aes(color = Survived))

 ggplot(data = titanic,
        aes(x = product(Class), conds = product(Sex), fill = Survived)) +
  geom_mosaic(alpha = 0.3,
              divider = c("vspine", "hspine", "hspine")) +
  geom_mosaic_jitter(
              divider = c("vspine", "hspine", "hspine"))

Labeling for Mosaic plots.

Description

A mosaic plot with text or labels

Usage

geom_mosaic_text(
  mapping = NULL,
  data = NULL,
  stat = "mosaic",
  position = "identity",
  na.rm = FALSE,
  divider = mosaic(),
  offset = 0.01,
  show.legend = NA,
  inherit.aes = TRUE,
  as.label = FALSE,
  repel = FALSE,
  repel_params = NULL,
  check_overlap = FALSE,
  display_values = "label",
  format_digits = 1,
  expected = NULL,
  ...
)

Arguments

Value

A ggplot2 layer that can be added to a plot.

Author(s)

Gavin Klorfine

Examples

data(titanic)

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

ggplot(data = titanic, aes(x = product(Class, Sex), fill = Survived)) +
  mosaic_settings(divider = c("vspine", "hspine", "hspine")) +
  geom_mosaic() +
  geom_mosaic_text(size = 2)

ggplot(data = happy, aes(x = product(happy, health), fill = happy)) +
  geom_mosaic(aes(x = product(health)), na.rm = TRUE, show.legend = FALSE) +
  geom_mosaic_text(na.rm = TRUE, show.legend = FALSE)

# avoid overlapping text
ggplot(data = happy, aes(x = product(happy, health), fill = happy)) +
  geom_mosaic(aes(x = product(health)), na.rm = TRUE, show.legend = FALSE) +
  geom_mosaic_text(na.rm = TRUE, check_overlap = TRUE, show.legend = FALSE)

# or use ggrepel
ggplot(data = happy, aes(x = product(happy, health), fill = happy)) +
  geom_mosaic(aes(x = product(health)), na.rm = TRUE, show.legend = FALSE) +
  geom_mosaic_text(na.rm = TRUE, repel = TRUE, show.legend = FALSE)

# and as a label
ggplot(data = happy, aes(x = product(happy, health), fill = happy)) +
  geom_mosaic(aes(x = product(health)), na.rm = TRUE, show.legend = FALSE) +
  geom_mosaic_text(
    na.rm = TRUE, repel = TRUE, as.label = TRUE,
    fill = "white", show.legend = FALSE
  )

# Display observed counts in cells
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  geom_mosaic(aes(fill = Survived)) +
  geom_mosaic_text(display_values = "observed")

# Display residuals with one shared model specification
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  mosaic_settings(expected = "independence") +
  geom_mosaic() +
  scale_fill_residual() +
  geom_mosaic_text(display_values = "residual",
                   format_digits = 2)

# Display expected values
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  mosaic_settings(expected = "independence") +
  geom_mosaic() +
  scale_fill_residual() +
  geom_mosaic_text(display_values = "expected",
                   format_digits = 1)

Launch shiny app (deprecated)

Description

Shiny app "EDA with Mosaic Plots" for interactive exploratory model building.

Usage

ggmosaic_app(example = c("mosaics", "models"), ...)

Arguments

Details

Deprecated. Inherited as-is from the original ggmosaic package and kept only for the historical record. It is not maintained, and currently cannot find its app directory because system.file() still looks up the old ggmosaic package name rather than ggmosaic2.

Value

Called for its side effect of launching a Shiny app; returns the result of shiny::runApp().

Examples

## Not run: 
# Deprecated and currently non-functional; see Details.
ggmosaic_app("mosaics")

## End(Not run)

Data related to happiness from the general social survey.

Description

The data is a small sample of variables related to happiness from the general social survey (GSS). The GSS is a yearly cross-sectional survey of Americans, run since 1972. We combine data for more than 25 years to yield over 60 thousand observations, and of the over 5,000 variables, we select some variables that are related to happiness:

Usage

data(happy)

Format

A data frame with 62466 rows and 11 variables

• year. year of the response, 1972 to 2018.

• age. age in years: 18–89 (89 stands for all 89 year olds and older).

• degree. highest education: lt high school, high school, junior college, bachelor, graduate.

• finrela. how is your financial status compared to others: far below, below average, average, above average, far above.

• happy. happiness: very happy, pretty happy, not too happy.

• health. health: excellent, good, fair, poor.

• marital. marital status: married, never married, divorced, widowed, separated.

• sex. sex: female, male.

• polviews. from extremely conservative to extremely liberal.

• partyid. party identification: strong republican, not str republican, ind near rep, independent, ind near dem, not str democrat, strong democrat, other party.

• wtssall. probability weight. 0.39–8.74

Horizontal bar partition: width constant, height varies.

Description

Horizontal bar partition: width constant, height varies.

Usage

hbar(data, bounds, offset = 0.02, max = NULL)

Arguments

Value

A data frame of rectangle boundaries (l, r, b, t), one row per level of data.

Examples

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

Horizontal spine partition: height constant, width varies.

Description

Horizontal spine partition: height constant, width varies.

Usage

hspine(data, bounds, offset = offset, max = NULL)

Arguments

Value

A data frame of rectangle boundaries (l, r, b, t), one row per level of data.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class))) +
  geom_mosaic(divider = "hspine")

Template for a mosaic plot. A mosaic plot is composed of spines in alternating directions.

Description

Template for a mosaic plot. A mosaic plot is composed of spines in alternating directions.

Usage

mosaic(direction = "h")

Arguments

Value

A function of one argument, the number of splits n, that returns a character vector of divider function names ("hspine"/"vspine") to apply at each split – suitable for the divider argument of geom_mosaic() and related layers.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  geom_mosaic(divider = mosaic("v"))

Settings for mosaic plot layers

Description

Set the divider, gap size, and model used to calculate expected frequencies for mosaic layers in a plot. A value set directly in a layer takes priority.

Usage

mosaic_settings(divider, offset, expected)

Arguments

Details

A layer inherits a setting only when its corresponding argument is omitted. Settings are resolved in this order: an explicitly supplied layer argument (including expected = NULL), the plot's mosaic_settings() value, and the layer default (mosaic(), 0.01, or NULL).

divider and offset are inherited by geom_mosaic(), stat_mosaic(), geom_mosaic_text(), stat_mosaic_text(), geom_mosaic_jitter(), and stat_mosaic_jitter(). expected is inherited by the same constructors except the mosaic-jitter geom and stat.

The position of mosaic_settings() among the layers in a plot has no effect. If it is added more than once, the last supplied value for each setting is used; omitted arguments do not change earlier settings. Settings are local to the plot and do not change package or session defaults. They share configuration, not computations or fitted model objects, between layers.

Value

An object of class "ggmosaic_settings" that can be added to a ggplot with +.

Author(s)

Gavin Klorfine

See Also

geom_mosaic(), stat_mosaic(), geom_mosaic_text(), stat_mosaic_text(), geom_mosaic_jitter(), stat_mosaic_jitter(), mosaic(), and ddecker()

Examples

data(titanic)

ggplot(titanic, aes(x = product(Class, Sex))) +
  mosaic_settings(
    expected = "independence",
    divider = c("vspine", "hspine"),
    offset = 0.005
  ) +
  geom_mosaic() +
  geom_mosaic_text(display_values = "residual") +
  scale_fill_residual() +
  theme_mosaic()

# An explicit layer argument overrides the plot setting. Here NULL turns
# model fitting off for the mosaic layer.
ggplot(titanic, aes(x = product(Class, Sex), fill = Survived)) +
  mosaic_settings(expected = "independence") +
  geom_mosaic(expected = NULL)

Calculate frequencies.

Description

Calculate frequencies.

Usage

prodcalc(
  data,
  formula,
  divider = mosaic(),
  cascade = 0,
  scale_max = TRUE,
  na.rm = FALSE,
  offset = 0.01,
  expected = NULL,
  variable_labels = NULL
)

Arguments

Value

A data frame giving rectangle boundaries (l, r, b, t) and computed frequencies for each partition/cell, plus .expected/.residual columns when expected is supplied.

Examples

data(happy)
prodcalc(happy, ~ happy, "hbar", offset = 0.005)
prodcalc(happy, ~ happy, "hspine", offset = 0.01)

Wrapper for a list

Description

Wrapper for a list

Usage

product(...)

Arguments

Value

A list of expressions (see rlang::exprs()), one per argument, used inside aes() to mark the variables that define the mosaic's product formula.

Examples

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

Diverging color scale for Pearson residuals

Description

Provides a red-white-blue color scale centered at 0 for visualizing Pearson residuals from loglinear models. Designed for use with geom_mosaic() when a model is supplied through the layer's expected argument or through mosaic_settings.

Usage

scale_fill_residual(
  ...,
  low = "darkred",
  mid = "white",
  high = "darkblue",
  midpoint = 0,
  limits = NULL,
  name = "Pearson\nResidual"
)

scale_fill_residuals(
  ...,
  low = "darkred",
  mid = "white",
  high = "darkblue",
  midpoint = 0,
  limits = NULL,
  name = "Pearson\nResidual"
)

Arguments

Details

The default legend always labels -4, 0, and 4. It also labels supplied limits and the observed minimum and maximum when those differ from the limits. The legend extends to every labeled value, with solid endpoint color beyond supplied limits. When the contributing mosaic cells have outlines, positive residuals have a solid dark blue outline, negative residuals have a dashed dark red outline, and an unoutlined midpoint band (white by default) separates them at zero. Setting colour = NA on every contributing mosaic layer removes these outlines from both the cells and the legend. Black ticks are drawn outside the color bar, which stretches with the mosaic panel. Nearby vertical labels are separated, and a thin elbow connects each displaced label to its exact tick. The neighboring label uses a longer straight tick so nearby text shares a common alignment. Automatically generated numeric labels are rounded to one decimal place. The legend can be hidden normally with theme(legend.position = "none").

Value

A ggplot2 fill scale that can be added to a plot.

Author(s)

Gavin Klorfine

Examples

data(titanic)

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

# Custom colors
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  geom_mosaic(expected = "independence") +
  scale_fill_residual(low = "red", high = "blue")

# Custom limits to highlight strong deviations
ggplot(data = titanic, aes(x = product(Class, Sex, Survived))) +
  geom_mosaic(expected = ~ Class + Sex) +
  scale_fill_residual(limits = c(-4, 4))

Helper function for determining scales

Description

Used internally to determine class of variable x

Usage

## S3 method for class 'productlist'
scale_type(x)

Arguments

Value

character string "productlist"

Determining scales for mosaics

Description

Determining scales for mosaics

Usage

scale_x_productlist(
  name = ggplot2::waiver(),
  breaks = product_breaks(),
  minor_breaks = NULL,
  labels = product_labels(),
  limits = NULL,
  expand = ggplot2::waiver(),
  oob = scales::censor,
  na.value = NA_real_,
  transform = "identity",
  position = "bottom",
  sec.axis = ggplot2::waiver()
)

scale_y_productlist(
  name = ggplot2::waiver(),
  breaks = product_breaks(),
  minor_breaks = NULL,
  labels = product_labels(),
  limits = NULL,
  expand = ggplot2::waiver(),
  oob = scales::censor,
  na.value = NA_real_,
  transform = "identity",
  position = "left",
  sec.axis = ggplot2::waiver()
)

ScaleContinuousProduct

Arguments

Value

A ggplot2 position scale that can be added to a plot.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class, Sex))) +
  geom_mosaic() +
  scale_x_productlist("Passenger class / Sex")

Translate shortcut strings to formulas

Description

Translate shortcut strings to formulas

Usage

shortcut_to_formula(shortcut, vars, conds = NULL)

Arguments

Value

Formula object

Spine partition: divide longest dimension.

Description

Spine partition: divide longest dimension.

Usage

spine(data, bounds, offset = offset, max = NULL)

Arguments

Value

A data frame of rectangle boundaries (l, r, b, t), one row per level of data.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class))) +
  geom_mosaic(divider = "spine")

Internal helper function

Description

Squeeze pieces to lie within specified bounds; directly copied from package productplots

Usage

squeeze(pieces, bounds = bound())

Arguments

Value

re-scaled values for piece according to boundaries given by bounds

Author(s)

Hadley Wickham

Theme for mosaic plots

Description

Themes set the general aspect of the plot such as the color of the background, gridlines, the size and color of fonts. theme_mosaic provides access to the regular ggplot2 theme, but removes any background, axes ticks, most of the gridlines, and ensures an aspect ratio of 1 for better viewing of the mosaics. This theme also applies a bold face to axes labels and allows for the convenient rotation of category labels.

Usage

theme_mosaic(base_size = 11, base_family = "", rot_labels = 0, ...)

Arguments

Details

As with other ggplot2 extensions, themes set the general look-and-feel of the plot such as the color of the background, gridlines, the size and color of fonts. theme_mosaic() provides access to the regular ggplot2 theme, but: removes any background, axes ticks, most of the gridlines, and ensures an aspect ratio of 1 for better viewing of the mosaics. This theme also applies a bold face to axes labels and allows for the convenient rotation of category labels to avoid overlap

Value

A ggplot2 theme object that can be added to a plot.

Author(s)

Gavin Klorfine

Examples

library(ggmosaic2)
data(happy)
ggplot(data = happy,
       aes(weight = wtssall, x = product(health), fill = happy)) +
  geom_mosaic(na.rm = TRUE) +
  theme_mosaic()

Passengers and crew on board the Titanic

Description

A dataset containing some demographics and survival of people on board the Titanic

Usage

titanic

Format

A data frame with 2201 rows and 4 variables:

Class

factor variable containing the class of a passenger (1st, 2nd, 3rd) or crew.

Sex

Male/Female.

Age

Child/Adult. This information is not very reliable, because it was inferred from boarding documents that did not state actual age in years.

Survived

Yes/No.

Vertical bar partition: height constant, width varies.

Description

Vertical bar partition: height constant, width varies.

Usage

vbar(data, bounds, offset = 0.02, max = NULL)

Arguments

Value

A data frame of rectangle boundaries (l, r, b, t), one row per level of data.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class))) +
  geom_mosaic(divider = "vbar")

Vertical spine partition: width constant, height varies.

Description

Vertical spine partition: width constant, height varies.

Usage

vspine(data, bounds, offset = offset, max = NULL)

Arguments

Value

A data frame of rectangle boundaries (l, r, b, t), one row per level of data.

Examples

data(titanic)
ggplot(data = titanic, aes(x = product(Class))) +
  geom_mosaic(divider = "vspine")