# ggstatsplot v0.0.3

0

0th

Percentile

## 'ggplot2' Based Plots with Statistical Details

Extension of 'ggplot2', 'ggstatsplot' creates graphics with details from statistical tests (parametric, non-parametric, or robust) included in the plots themselves. It is targeted primarily at behavioral sciences community to provide a one-line code to generate information-rich plots for statistical analysis of continuous (violin plots, scatterplots, histograms) or categorical (pie charts) data.

# ggstatsplot: ggplot2 Based Plots with Statistical Details

## Overview

ggstatsplot is an extension of ggplot2 package for creating graphics with details from statistical tests included in the plots themselves and targeted primarily at behavioral sciences community to provide a one-line code to produce information-rich plots. Currently, it supports only the most common types of statisticla tests (parametric, nonparametric, and robust versions of t-tets/anova, correlation, and contingency tables analyses). Accordingly, it produces limited kinds of plots: violin plots (for comparisons between groups or conditions), pie charts (for categorical data), scatterplots (for correlations between variables), and histograms (for hypothesis about distributions).

Future versions will include other types of analyses and plots as well.

## Installation

To get the latest, stable CRAN release:

utils::install.packages(pkgs = "ggstatsplot")


You can get the development version from GitHub. If you are in hurry and want to reduce the time of installation, prefer-

# needed package to download from GitHub repo
utils::install.packages(pkgs = "devtools")

devtools::install_github(
repo = "IndrajeetPatil/ggstatsplot", # package path on GitHub
quick = TRUE                         # skips docs, demos, and vignettes
)


If time is not a constraint-

devtools::install_github(
repo = "IndrajeetPatil/ggstatsplot", # package path on GitHub
dependencies = TRUE,                 # installs packages which ggstatsplot depends on
)


If you are not using the RStudio IDE and you get an error related to “pandoc” you will either need to remove the argument build_vignettes = TRUE (to avoid building the vignettes) or install pandoc. If you have the rmarkdown R package installed then you can check if you have pandoc by running the following in R:

rmarkdown::pandoc_available()
#> [1] TRUE


## Citation

If you want to cite this package in a scientific journal or in any other context, run the following code in your R console:

utils::citation(package = "ggstatsplot")


## Help

Documentation for any function can be accessed with the standard help command-

?ggbetweenstats
?ggscatterstats
?gghistostats
?ggpiestats
?ggcorrmat
?combine_plots
?grouped_ggbetweenstats
?grouped_ggscatterstats
?grouped_gghistostats
?grouped_ggpiestats
?grouped_ggcorrmat


## Usage

ggstatsplot relies on non-standard evaluation, which means you can’t enter arguments in the following manner: x = data$x, y = data$y. This may work well for most of the functions most of the time, but is highly discouraged. You should always specify data argument for all functions.

Additionally, ggstatsplot is a very chatty package and will by default output information about references for tests, notes on assumptions about linear models, and warnings. If you don’t want your console to be cluttered with such messages, they can be turned off by setting messages = FALSE.

Here are examples of the main functions currently supported in ggstatsplot:

• ggbetweenstats

This function creates a violin plot for between-group or between-condition comparisons with results from statistical tests in the subtitle. The simplest function call looks like this-

ggstatsplot::ggbetweenstats(
data = datasets::iris,
x = Species,
y = Sepal.Length,
messages = FALSE
)


Number of other arguments can be specified to make this plot even more informative and, additionally, this function returns a ggplot2 object and thus any of the graphics layers can be further modified:

library(ggplot2)

ggstatsplot::ggbetweenstats(
data = datasets::iris,
x = Species,
y = Sepal.Length,
notch = TRUE,                                   # show notched box plot
mean.plotting = TRUE,                           # whether mean for each group id to be displayed
type = "parametric",                            # which type of test is to be run
outlier.tagging = TRUE,                         # whether outliers need to be tagged
outlier.label = Sepal.Width,                    # variable to be used for the outlier tag
xlab = "Type of Species",                       # label for the x-axis variable
ylab = "Attribute: Sepal Length",               # label for the y-axis variable
title = "Dataset: Iris flower data set",        # title text for the plot
caption = expression(                           # caption text for the plot
paste(italic("Note"), ": this is a demo")
),
messages = FALSE
) +                                               # further modification outside of ggstatsplot
ggplot2::coord_cartesian(ylim = c(3, 8)) +
ggplot2::scale_y_continuous(breaks = seq(3, 8, by = 1))


The type (of test) argument also accepts the following abbreviations: "p" (for parametric), "np" (for nonparametric), "r" (for robust). Additionally, the type of plot to be displayed can also be modified ("box", "violin", or "boxviolin").

Variant of this function ggwithinstats is currently under work. You can still use this function just to prepare the plot for exploratory data analysis, but the statistical details displayed in the subtitle will be incorrect. You can remove them by adding + ggplot2::labs(subtitle = NULL).

For more, see the ggbetweenstats vignette: https://indrajeetpatil.github.io/ggstatsplot/articles/ggbetweenstats.html

• ggscatterstats

This function creates a scatterplot with marginal histograms/boxplots/density/violin plots from and results from statistical tests in the subtitle:

ggstatsplot::ggscatterstats(
data = datasets::iris,
x = Sepal.Length,
y = Petal.Length,
title = "Dataset: Iris flower data set",
messages = FALSE
)


Number of other arguments can be specified to modify this basic plot-

library(datasets)

ggstatsplot::ggscatterstats(
data = subset(datasets::iris, iris\$Species == "setosa"),
x = Sepal.Length,
y = Petal.Length,
type = "robust",                               # type of test that needs to be run
xlab = "Attribute: Sepal Length",              # label for x axis
ylab = "Attribute: Petal Length",              # label for y axis
line.color = "black",                         # changing regression line color line
title = "Dataset: Iris flower data set",       # title text for the plot
caption = expression(                          # caption text for the plot
paste(italic("Note"), ": this is a demo")
),
marginal.type = "density",                     # type of marginal distribution to be displayed
xfill = "blue",                                # color fill for x-axis marginal distribution
yfill = "red",                                 # color fill for y-axis marginal distribution
centrality.para = "median",                    # which type of central tendency lines are to be displayed
width.jitter = 0.2,                            # amount of horizontal jitter for data points
height.jitter = 0.4,                           # amount of vertical jitter for data points
messages = FALSE                               # turn off messages and notes
)


For more, see the ggscatterstats vignette: https://indrajeetpatil.github.io/ggstatsplot/articles/ggscatterstats.html

• ggpiestats

This function creates a pie chart for categorical variables with results from contingency table analysis included in the subtitle of the plot. If only one categorical variable is entered, proportion test will be carried out.

ggstatsplot::ggpiestats(
data = datasets::iris,
main = Species,
messages = FALSE
)


This function can also be used to study an interaction between two categorical variables. Additionally, as with the other functions in ggstatsplot, this function returns a ggplot2 object and can further be modified with ggplot2 syntax (e.g., we can change the color palette after ggstatsplot has produced the plot)-

library(ggplot2)

ggstatsplot::ggpiestats(
data = datasets::mtcars,
main = cyl,
condition = am,
title = "Dataset: Motor Trend Car Road Tests",
messages = FALSE
) +
ggplot2::scale_fill_brewer(palette = "Dark2")   # further modification outside of ggstatsplot


As with the other functions, this basic plot can further be modified with additional arguments:

library(ggplot2)

ggstatsplot::ggpiestats(
data = datasets::mtcars,
main = am,
condition = cyl,
title = "Dataset: Motor Trend Car Road Tests",      # title for the plot
stat.title = "interaction effect",                  # title for the results from Pearson's chi-squared test
legend.title = "Transmission",                      # title for the legend
factor.levels = c("0 = automatic", "1 = manual"),   # renaming the factor level names for main variable
facet.wrap.name = "No. of cylinders",               # name for the facetting variable
facet.proptest = FALSE,                             # turning of facetted proportion test results
caption = expression(                               # text for the caption
paste(italic("Note"), ": this is a demo")
),
messages = FALSE                                    # turn off messages and notes
)


For more, including information about the variant of this function grouped_ggpiestats, see the ggpiestats vignette: https://indrajeetpatil.github.io/ggstatsplot/articles/ggpiestats.html

• gghistostats

In case you would like to see the distribution of one variable and check if it is significantly different from a specified value with a one sample test, this function will let you do that.

library(datasets)

ggstatsplot::gghistostats(
data = datasets::iris,
x = Sepal.Length,
title = "Distribution of Iris sepal length",
type = "parametric",               # one sample t-test
test.value = 3,                    # default value is 0
centrality.para = "mean",          # which measure of central tendency is to be plotted
centrality.color = "darkred",     # decides color of vertical line representing central tendency
binwidth = 0.10,                   # binwidth value (needs to be toyed around with until you find the best one)
messages = FALSE                   # turn off the messages
)


The type (of test) argument also accepts the following abbreviations: "p" (for parametric) or "np" (for nonparametric) or "bf" (for Bayes Factor).

ggstatsplot::gghistostats(
data = NULL,
title = "Distribution of variable x",
x = stats::rnorm(n = 1000, mean = 0, sd = 1),
test.value = 1,
test.value.line = TRUE,
test.value.color = "black",
centrality.para = "mean",
type = "bf",
bf.prior = 0.8,
messages = FALSE,
caption = expression(
paste(italic("Note"), ": black line - test value; blue line - observed mean")
)
)


As seen here, by default, Bayes Factor quantifies the support for the alternative hypothesis (H1) over the null hypothesis (H0) (i.e., BF10 is displayed). In case you run parametric t-test and the effect is not significant, caption will be displayed containing information about evidence in favor of the null hypothesis (H0). This is not recommended, but if you want to turn off this behavior, you can use the argument bf.message = FALSE.

ggstatsplot::gghistostats(
data = datasets::ToothGrowth,
x = len,
title = "Distribution of tooth length",
centrality.para = "mean",
test.value = 20,
test.value.line = TRUE,
xlab = "Tooth length",
caption = expression(
paste(italic("Note"), ": black line - test value; blue line - observed mean")
),
messages = FALSE
)


For more, including information about the variant of this function grouped_gghistostats, see the gghistostats vignette: https://indrajeetpatil.github.io/ggstatsplot/articles/gghistostats.html

• ggcorrmat

ggcorrmat makes correlalograms with minimal amount of code. Just sticking to the defaults itself produces publication-ready correlation matrices. (Wrapper around ggcorrplot)

# as a default this function outputs a correlalogram plot
ggstatsplot::ggcorrmat(
data = datasets::iris,
corr.method = "spearman",                # correlation method
sig.level = 0.005,                       # threshold of significance
cor.vars = Sepal.Length:Petal.Width,     # a range of variables can be selected
cor.vars.names = c("Sepal Length", "Sepal Width", "Petal Length", "Petal Width"),
title = "Correlalogram for length measures for Iris species",
subtitle = "Iris dataset by Anderson",
caption = expression(
paste(
italic("Note"),
": X denotes correlation non-significant at ",
italic("p "),
)
)
)


Multiple arguments can be modified to change the appearance of the correlation matrix.

Alternatively, you can use it just to get the correlation matrices and their corresponding p-values (in a tibble format). This is especially useful for robust correlation coefficient, which is not currently supported in ggcorrmat plot.

# getting the correlation coefficient matrix
ggstatsplot::ggcorrmat(
data = datasets::iris,
cor.vars = Sepal.Length:Petal.Width,
corr.method = "robust",
output = "correlations",             # specifying the needed output
digits = 3                           # number of digits to be dispayed for correlation coefficient
)
#> # A tibble: 4 x 5
#>   variable     Sepal.Length Sepal.Width Petal.Length Petal.Width
#>   <chr>               <dbl>       <dbl>        <dbl>       <dbl>
#> 1 Sepal.Length        1          -0.193        0.878       0.846
#> 2 Sepal.Width        -0.193       1           -0.452      -0.392
#> 3 Petal.Length        0.878      -0.452        1           0.966
#> 4 Petal.Width         0.846      -0.392        0.966       1

# getting the p-value matrix
ggstatsplot::ggcorrmat(
data = datasets::iris,
cor.vars = Sepal.Length:Petal.Width,
corr.method = "robust",
output = "p-values"
)
#> # A tibble: 4 x 5
#>   variable     Sepal.Length   Sepal.Width  Petal.Length Petal.Width
#>   <chr>               <dbl>         <dbl>         <dbl>       <dbl>
#> 1 Sepal.Length       0      0.0177        0             0
#> 2 Sepal.Width        0.0177 0             0.00000000636 0.000000686
#> 3 Petal.Length       0      0.00000000636 0             0
#> 4 Petal.Width        0      0.000000686   0             0


For examples and more information, see the ggcorrmat vignette: https://indrajeetpatil.github.io/ggstatsplot/articles/ggcorrmat.html

• combine_plots

ggstatsplot also contains a helper function combine_plots to combine multiple plots. This is a wrapper around and lets you combine multiple plots and add combination of title, caption, and annotation texts with suitable default parameters.

The full power of ggstatsplot can be leveraged with a functional programming package like purrr that replaces many for loops with code that is both more succinct and easier to read and, therefore, purrr should be preferrred.

For more, see the associated vignette- https://indrajeetpatil.github.io/ggstatsplot/articles/theme_mprl.html

• theme_mprl

All plots from ggstatsplot have a default theme: theme_mprl. For more, see the associated vignette- https://indrajeetpatil.github.io/ggstatsplot/articles/theme_mprl.html

## Functions in ggstatsplot

 Name Description legend_title_margin Custom function to set upper and lower margins to legend title in ggplot2 theme_mprl Default theme used in all ggstatsplot package plots theme_corrmat Default theme used for correlation matrix ggstatsplot-package ggstatsplot ggcorrmat Visualization of a correlalogram (or correlation matrix) using 'ggplot2'/'ggcorrplot' gghistostats Histogram for distribution of a numeric variable ggpiestats Pie charts with statistical tests ggscatterstats Scatterplot with marginal distributions grouped_ggcorrmat Visualization of a correlalogram (or correlation matrix) using 'ggplot2'/'ggcorrplot' for all levels of a grouping variable grouped_ggscatterstats Scatterplot with marginal distributions for all levels of a grouping variable specify_decimal_p Custom function for getting specified number of decimal places in results for p-value grouped_ggbetweenstats Violin plots for group or condition comparisons repeated across all levels of a grouping variable. check_outlier Finding the outliers in the dataframe using Tukey's interquartile range rule ggbetweenstats violin plots for group or condition comparisons grouped_proptest Function to run proportion test on grouped data. combine_plots Combining multiple plots using cowplot::plot_grid() with a combination of title, caption, and annotation label signif_column Creating a new character type column with significance labels grouped_gghistostats Grouped histograms for distribution of a numeric variable grouped_ggpiestats Grouped pie charts with statistical tests %>% Pipe operator theme_pie Default theme used for pie chart No Results!