Stacked and dodged bar plots — ggplot_bar_stacked

Utility functions for plotting stacked (on top of each other) and dodged (next to each other) bars in the same figure.

Usage

ggplot_bar_stacked_dodged(data, mapping, gap = 1)

add_stacked_dodged_xpos(data, ..., gap = 1)

calc_stacked_dodged_xlabels(data, ..., gap = 1)

Arguments

data: A data frame
mapping: An aesthetic mapping generated by aes(), containing the aesthetics x, y, fill, and dodge. The aesthetic x will form groups on the x-axis, while dodge will form individual bars within the groups.
gap: The width of the gap between bars, relative to the width of the bars themselves (default: 1).
...: A selection of two columns. Both will be combined to form x-axis coordinates. The first will form the outer iteration (groups), the second the inner iteration (bars within a group). If unnamed, the column with calculated positions will be called xpos.

Value

add_stacked_dodged_xpos() returns the input data frame with an additional column. Row and column order are preserved. calc_stacked_dodged_xlabels() returns a named character vector for use with scale_x_continuous(). ggplot_bar_stacked_dodged() returns a ggplot object.

Details

add_stacked_dodged_xpos() adds x-axis positions to a data frame for plotting two categorical variables within a bar plot. calc_stacked_dodged_xlabels() calculates matching label positions on the x-axis. ggplot_bar_stacked_dodged() uses both functions to generate a plot.

Author

Michaja Pehl

Examples

require(tidyverse)

set.seed(0)
(data <- crossing(a = factorise(c('left', 'center', 'right')),
                  b = factorise(c('top', 'middle', 'bottom')),
                  c = letters[1:4],
                  d = LETTERS[25:26]) %>%
        mutate(value = abs(rnorm(n())) + 0.2))
#> # A tibble: 72 × 5
#>    a     b      c     d     value
#>    <fct> <fct>  <chr> <chr> <dbl>
#>  1 left  top    a     Y     1.46 
#>  2 left  top    a     Z     0.526
#>  3 left  top    b     Y     1.53 
#>  4 left  top    b     Z     1.47 
#>  5 left  top    c     Y     0.615
#>  6 left  top    c     Z     1.74 
#>  7 left  top    d     Y     1.13 
#>  8 left  top    d     Z     0.495
#>  9 left  middle a     Y     0.206
#> 10 left  middle a     Z     2.60 
#> # ℹ 62 more rows

(plot.data <- add_stacked_dodged_xpos(data, c('c', 'a')))
#> # A tibble: 72 × 6
#>    a     b      c     d     value  xpos
#>    <fct> <fct>  <chr> <chr> <dbl> <dbl>
#>  1 left  top    a     Y     1.46      1
#>  2 left  top    a     Z     0.526     1
#>  3 left  top    b     Y     1.53      5
#>  4 left  top    b     Z     1.47      5
#>  5 left  top    c     Y     0.615     9
#>  6 left  top    c     Z     1.74      9
#>  7 left  top    d     Y     1.13     13
#>  8 left  top    d     Z     0.495    13
#>  9 left  middle a     Y     0.206     1
#> 10 left  middle a     Z     2.60      1
#> # ℹ 62 more rows

(xlabels <- calc_stacked_dodged_xlabels(data, c('c', 'a')))
#>  a  b  c  d 
#>  2  6 10 14 

ggplot(data = plot.data) +
    scale_x_continuous(breaks = xlabels) +
    facet_wrap(~ d, ncol = 1, scales = 'free_x')


ggplot_bar_stacked_dodged(data, aes(x = a, y = value, fill = b, dodge = c),
                          gap = 1/3) +
    facet_wrap(~ d, ncol = 1, scales = 'free_x')