grouped_ggbarstats
Grouped bar (column) charts with statistical tests
Helper function for ggstatsplot::ggbarstats
to apply this
function across multiple levels of a given factor and combining the
resulting plots using ggstatsplot::combine_plots
.
Usage
grouped_ggbarstats(data, main, condition, counts = NULL, grouping.var,
title.prefix = NULL, ratio = NULL, paired = FALSE,
results.subtitle = TRUE, labels.legend = NULL, stat.title = NULL,
sample.size.label = TRUE, label.separator = " ",
label.text.size = 4, label.fill.color = "white",
label.fill.alpha = 1, bar.outline.color = "black",
bf.message = TRUE, sampling.plan = "indepMulti",
fixed.margin = "rows", prior.concentration = 1, subtitle = NULL,
caption = NULL, legend.position = "right",
x.axis.orientation = NULL, conf.level = 0.95, nboot = 100,
simulate.p.value = FALSE, B = 2000, legend.title = NULL,
xlab = NULL, ylab = "Percent", k = 2, perc.k = 0,
bar.label = "percentage", data.label = NULL, bar.proptest = TRUE,
ggtheme = ggplot2::theme_bw(), ggstatsplot.layer = TRUE,
package = "RColorBrewer", palette = "Dark2", direction = 1,
ggplot.component = NULL, return = "plot", messages = TRUE, ...)
Arguments
- data
A dataframe (or a tibble) from which variables specified are to be taken. A matrix or tables will not be accepted.
- main
The variable to use as the rows in the contingency table.
- condition
The variable to use as the columns in the contingency table.
- counts
A string naming a variable in data containing counts, or
NULL
if each row represents a single observation (Default).- grouping.var
A single grouping variable (can be entered either as a bare name
x
or as a string"x"
).- title.prefix
Character string specifying the prefix text for the fixed plot title (name of each factor level) (Default:
NULL
). IfNULL
, the variable name entered forgrouping.var
will be used.- ratio
A vector of numbers: the expected proportions for the proportion test. Default is
NULL
, which means if there are two levelsratio = c(1,1)
, etc.- paired
Logical indicating whether data came from a within-subjects design study (Default:
FALSE
). IfTRUE
, McNemar test subtitle will be returned. IfFALSE
, Pearson's chi-square test will be returned.- results.subtitle
Decides whether the results of statistical tests are to be displayed as a subtitle (Default:
TRUE
). If set toFALSE
, only the plot will be returned.- labels.legend
A character vector with custom labels for levels of the
main
variable displayed in the legend.- stat.title
Title for the effect being investigated with the chi-square test. The default is
NULL
, i.e. no title will be added to describe the effect being shown. An example of astat.title
argument will be something like"main x condition"
or"interaction"
.- sample.size.label
Logical that decides whether sample size information should be displayed for each level of the grouping variable
condition
(Default:TRUE
).- label.separator
If
"both"
counts and proportion information is to be displayed in a label, this argument decides whether these two pieces of information are going to be on the same line (" "
) or on separate lines ("\n"
).- label.text.size
Numeric that decides text size for slice/bar labels (Default:
4
).- label.fill.color
Character that specifies fill color for slice/bar labels (Default:
white
).- label.fill.alpha
Numeric that specifies fill color transparency or
"alpha"
for slice/bar labels (Default:1
range0
to1
).- bar.outline.color
Character specifying color for bars (default:
"black"
).- bf.message
Logical that decides whether to display a caption with results from bayes factor test in favor of the null hypothesis (default:
FALSE
).- sampling.plan
Character describing the sampling plan. Possible options are
"indepMulti"
(independent multinomial; default),"poisson"
,"jointMulti"
(joint multinomial),"hypergeom"
(hypergeometric). For more, see?BayesFactor::contingencyTableBF()
.- fixed.margin
For the independent multinomial sampling plan, which margin is fixed (
"rows"
or"cols"
). Defaults to"rows"
.- prior.concentration
Specifies the prior concentration parameter, set to
1
by default. It indexes the expected deviation from the null hypothesis under the alternative, and corresponds to Gunel and Dickey's (1974)"a"
parameter.- subtitle
The text for the plot subtitle. Will work only if
results.subtitle = FALSE
.- caption
The text for the plot caption.
- legend.position
The position of the legend
"none"
,"left"
,"right"
,"bottom"
,"top"
(Default:"right"
).- x.axis.orientation
The orientation of the
x
axis labels one of "slant" or "vertical" to change from the default horizontal orientation (Default:NULL
which is horizontal).- conf.level
Scalar between 0 and 1. If unspecified, the defaults return
95%
lower and upper confidence intervals (0.95
).- nboot
Number of bootstrap samples for computing confidence interval for the effect size (Default:
100
).- simulate.p.value
a logical indicating whether to compute p-values by Monte Carlo simulation.
- B
an integer specifying the number of replicates used in the Monte Carlo test.
- legend.title
Title text for the legend.
- xlab
Custom text for the
x
axis label (Default:NULL
, which will cause thex
axis label to be themain
variable).- ylab
Custom text for the
y
axis label (Default:"percent"
).- k
Number of digits after decimal point (should be an integer) (Default:
k = 2
).- perc.k
Numeric that decides number of decimal places for percentage labels (Default:
0
).- bar.label
Character decides what information needs to be displayed on the label in each pie slice. Possible options are
"percentage"
(default),"counts"
,"both"
.- data.label
Character decides what information needs to be displayed on the label in each pie slice. Possible options are
"percentage"
(default),"counts"
,"both"
.- bar.proptest
Decides whether proportion test for
main
variable is to be carried out for each level ofcondition
(Default:TRUE
).- ggtheme
A function,
ggplot2
theme name. Default value isggplot2::theme_bw()
. Any of theggplot2
themes, or themes from extension packages are allowed (e.g.,ggthemes::theme_fivethirtyeight()
,hrbrthemes::theme_ipsum_ps()
, etc.).- ggstatsplot.layer
Logical that decides whether
theme_ggstatsplot
theme elements are to be displayed along with the selectedggtheme
(Default:TRUE
).- package
Name of package from which the palette is desired as string or symbol.
- palette
If a character string (e.g.,
"Set1"
), will use that named palette. If a number, will index into the list of palettes of appropriate type. Default palette is"Dark2"
.- direction
Either
1
or-1
. If-1
the palette will be reversed.- ggplot.component
A
ggplot
component to be added to the plot prepared byggstatsplot
. This argument is primarily helpful forgrouped_
variant of the current function. Default isNULL
. The argument should be entered as a function. If the given function has an argumentaxes.range.restrict
and if it has been set toTRUE
, the added ggplot component might not work as expected.- return
Character that describes what is to be returned: can be
"plot"
(default) or"subtitle"
or"caption"
. Setting this to"subtitle"
will return the expression containing statistical results, which will be aNULL
if you setresults.subtitle = FALSE
. Setting this to"caption"
will return the expression containing details about Bayes Factor analysis, but valid only whentype = "p"
andbf.message = TRUE
, otherwise this will return aNULL
.- messages
Decides whether messages references, notes, and warnings are to be displayed (Default:
TRUE
).- ...
Arguments passed on to
combine_plots
- title.text
String or plotmath expression to be drawn as title for the combined plot.
- title.color
Text color for title.
- title.size
Point size of title text.
- title.vjust
Vertical justification for title. Default =
0.5
(centered ony
).0
= baseline aty
,1
= ascender aty
.- title.hjust
Horizontal justification for title. Default =
0.5
(centered onx
).0
= flush-left at x,1
= flush-right.- title.fontface
The font face (
"plain"
,"bold"
(default),"italic"
,"bold.italic"
) for title.- caption.text
String or plotmath expression to be drawn as the caption for the combined plot.
- caption.color
Text color for caption.
- caption.size
Point size of title text.
- caption.vjust
Vertical justification for caption. Default =
0.5
(centered on y).0
= baseline at y,1
= ascender at y.- caption.hjust
Horizontal justification for caption. Default =
0.5
(centered on x).0
= flush-left at x,1
= flush-right.- caption.fontface
The font face (
"plain"
(default),"bold"
,"italic"
,"bold.italic"
) for caption.- sub.text
The label with which the combined plot should be annotated. Can be a plotmath expression.
- sub.color
Text color for annotation label (Default:
"black"
).- sub.size
Point size of annotation text (Default:
12
).- sub.x
The x position of annotation label (Default:
0.5
).- sub.y
The y position of annotation label (Default:
0.5
).- sub.hjust
Horizontal justification for annotation label (Default:
0.5
).- sub.vjust
Vertical justification for annotation label (Default:
0.5
).- sub.vpadding
Vertical padding. The total vertical space added to the label, given in grid units. By default, this is added equally above and below the label. However, by changing the y and vjust parameters, this can be changed (Default:
grid::unit(1, "lines")
).- sub.fontface
The font face (
"plain"
(default),"bold"
,"italic"
,"bold.italic"
) for the annotation label.- sub.angle
Angle at which annotation label is to be drawn (Default:
0
).- sub.lineheight
Line height of annotation label.
- title.caption.rel.heights
Numerical vector of relative columns heights while combining (title, plot, caption).
- title.rel.heights
Numerical vector of relative columns heights while combining (title, plot).
- caption.rel.heights
Numerical vector of relative columns heights while combining (plot, caption).
Value
Unlike a number of statistical softwares, ggstatsplot
doesn't
provide the option for Yates' correction for the Pearson's chi-squared
statistic. This is due to compelling amount of Monte-Carlo simulation
research which suggests that the Yates' correction is overly conservative,
even in small sample sizes. As such it is recommended that it should not
ever be applied in practice (Camilli & Hopkins, 1978, 1979; Feinberg, 1980;
Larntz, 1978; Thompson, 1988).
For more about how the effect size measures and their confidence intervals
are computed, see ?rcompanion::cohenG
, ?rcompanion::cramerV
, and
?rcompanion::cramerVFit
.
See Also
Examples
# NOT RUN {
# }
# NOT RUN {
# with condition and with count data
library(jmv)
ggstatsplot::grouped_ggbarstats(
data = as.data.frame(HairEyeColor),
main = Hair,
condition = Eye,
counts = Freq,
grouping.var = Sex
)
# the following will take slightly more amount of time
# for reproducibility
set.seed(123)
# let's create a smaller dataframe
diamonds_short <- ggplot2::diamonds %>%
dplyr::filter(.data = ., cut %in% c("Very Good", "Ideal")) %>%
dplyr::filter(.data = ., clarity %in% c("SI1", "SI2", "VS1", "VS2")) %>%
dplyr::sample_frac(tbl = ., size = 0.05)
# plot
ggstatsplot::grouped_ggbarstats(
data = diamonds_short,
main = color,
condition = clarity,
grouping.var = cut,
sampling.plan = "poisson",
title.prefix = "Quality",
bar.label = "both",
messages = FALSE,
perc.k = 1,
nrow = 2
)
# }