The function ggstatsplot::ggwithinstats is designed to facilitate data exploration, and for making highly customizable publication-ready plots, with relevant statistical details included in the plot itself if desired. We will see examples of how to use this function in this vignette.

To begin with, here are some instances where you would want to use ggwithinstats-

  • to check if a continuous variable differs across multiple groups/conditions

  • to compare distributions visually and check for outliers

Note: This vignette uses the pipe operator (%>%), if you are not familiar with this operator, here is a good explanation:

Comparisons between groups with ggwithistats

To illustrate how this function can be used, we will use the bugs dataset throughout this vignette. This data set, “Bugs”, provides the extent to which men and women want to kill arthropods that vary in freighteningness (low, high) and disgustingness (low, high). Each participant rates their attitudes towards all anthropods. Subset of the data reported by Ryan et al. (2013). Note that this is a repeated measures design because the same participant gave four different ratings across four different conditions (LDLF, LDHF, HDLF, HDHF).

Let’s have a look at the data-

Suppose the first thing we want to inspect is the distribution of desire to kill across all conditions (disregarding the factorial structure of the experiment). We also want to know if the mean differences in this desire across conditions is statistically significant.

The simplest form of the function call is-


  • The function automatically decides whether a dependent samples test is preferred (for 2 groups) or an ANOVA (3 or more groups). based on the number of levels in the grouping variable.
  • The output of the function is a ggplot object which means that it can be further modified with ggplot2 functions.

We can make the output much more aesthetically pleasing as well as informative by making use of the many optional parameters in ggwithinstats. We’ll add a title and caption, better x and y axis labels, and tag and label the outliers in the data. We can and will change the overall theme as well as the color palette in use.

As can be appreciated from the effect size (partial eta squared) of 0.18, there are small differences in the mean desire to kill across conditions. Importantly, this plot also helps us appreciate the distributions within any given condition.

So far we have only used a classic parametric test, but we can also use other available options: The type (of test) argument also accepts the following abbreviations: "p" (for parametric), "np" (for nonparametric), "r" (for robust), "bf" (for Bayes Factor).

Let’s use the combine_plots function to make one plot from four separate plots that demonstrates all of these options. Let’s compare desire to kill bugs only for low versus high disgust conditions to see how much of a difference whether a bug is disgusting-looking or not makes to the desire to kill that bug. We will generate the plots one by one and then use combine_plots to merge them into one plot with some common labeling. It is possible, but not necessarily recommended, to make each plot have different colors or themes.

For example,

# for reproducibility

# selecting subset of the data
df_disgust <- dplyr::filter(.data = data_bugs, key %in% c("LDHF", "HDHF"))

# parametric t-test
p1 <- ggstatsplot::ggwithinstats(
  data = df_disgust,
  x = key,
  y = value,
  type = "p",
  effsize.type = "d",
  conf.level = 0.99,
  title = "Parametric test",
  package = "ggsci",
  palette = "nrc_npg",
  ggtheme = ggthemr::ggthemr(palette = "light"),
  messages = FALSE

# Mann-Whitney U test (nonparametric test)
p2 <- ggstatsplot::ggwithinstats(
  data = df_disgust,
  x = key,
  y = value,
  xlab = "Condition",
  ylab = "Desire to kill bugs",
  type = "np",
  conf.level = 0.99,
  sort = "ascending",
  title = "Non-parametric Test",
  package = "ggsci",
  palette = "uniform_startrek",
  ggtheme = ggthemes::theme_map(),
  ggstatsplot.layer = FALSE,
  messages = FALSE

# robust t-test
p3 <- ggstatsplot::ggwithinstats(
  data = df_disgust,
  x = key,
  y = value,
  xlab = "Condition",
  ylab = "Desire to kill bugs",
  type = "r",
  conf.level = 0.99,
  sort = "descending",
  title = "Robust Test",
  tr = 0.2,
  package = "wesanderson",
  palette = "Royal2",
  ggtheme = hrbrthemes::theme_ipsum_tw(),
  ggstatsplot.layer = FALSE,
  messages = FALSE

# Bayes Factor for parametric t-test
p4 <- ggstatsplot::ggwithinstats(
  data = df_disgust,
  x = key,
  y = value,
  xlab = "Condition",
  ylab = "Desire to kill bugs",
  type = "bf",
  title = "Bayesian Test",
  package = "ggsci",
  palette = "nrc_npg",
  ggtheme = ggthemes::theme_fivethirtyeight(),
  messages = FALSE

# combining the individual plots into a single plot
  p1, p2, p3, p4,
  nrow = 2,
  title.text = "Effect of disgust on desire to kill bugs ",
  caption.text = "Source: Bugs dataset from `jmv` R package",
  title.size = 14,
  caption.size = 12

Grouped analysis with grouped_ggwithinstats

What if we want to carry out this same analysis but for each region (or gender)?

ggstatsplot provides a special helper function for such instances: grouped_ggwithinstats. This is merely a wrapper function around ggstatsplot::combine_plots. It applies ggwithinstats across all levels of a specified grouping variable and then combines list of individual plots into a single plot. Note that the grouping variable can be anything: conditions in a given study, groups in a study sample, different studies, etc.

Let’s focus on the two regions and for years: 1967, 1987, 2007. Also, let’s carry out pairwise comparisons to see if there differences between every pair of continents.

Grouped analysis with ggwithinstats + purrr

Although this grouping function provides a quick way to explore the data, it leaves much to be desired. For example, the same type of test and theme is applied for all genders, but maybe we want to change this for different genders, or maybe we want to gave different effect sizes for different years. This type of customization for different levels of a grouping variable is not possible with grouped_ggwithinstats, but this can be easily achieved using the purrr package.

See the associated vignette here:

Between-subjects designs

For independent measures designs, ggbetweenstats function can be used:

Summary of tests

Following (within-subjects) tests are carried out for each type of analyses-

Type No. of groups Test
Parametric > 2 One-way repeated measures ANOVA
Non-parametric > 2 Friedman test
Robust > 2 Heteroscedastic one-way repeated measures ANOVA for trimmed means
Bayes Factor > 2 One-way repeated measures ANOVA
Parametric 2 Student’s t-test
Non-parametric 2 Wilcoxon signed-rank test
Robust 2 Yuen’s test on trimmed means for dependent samples
Bayes Factor 2 Student’s t-test

Following effect sizes (and confidence intervals/CI) are available for each type of test-

Type No. of groups Effect size CI?
Parametric > 2 \(\eta_{p}^2\), \(\omega^2\) Yes
Non-parametric > 2 \(W_{Kendall}\) (Kendall’s coefficient of concordance) Yes
Robust > 2 No No
Bayes Factor > 2 No No
Parametric 2 Cohen’s d, Hedge’s g (central-and noncentral-t distribution based) Yes
Non-parametric 2 r (computed as \(Z/\sqrt{N}\)) Yes
Robust 2 \(\xi\) (Explanatory measure of effect size) Yes
Bayes Factor 2 No No

Here is a summary of multiple pairwise comparison tests supported in ggwithinstats-

Type Test p-value adjustment?
Parametric Student’s t-test Yes
Non-parametric Durbin-Conover test Yes
Robust Yuen’s trimmed means test Yes
Bayes Factor No No

Effect size interpretation

To see how the effect sizes displayed in these tests can be interpreted, see:


If you find any bugs or have any suggestions/remarks, please file an issue on GitHub: