Scatterplots from ggplot2 combined with marginal histograms/boxplots/density plots with statistical details added as a subtitle.

ggscatterstats(
  data,
  x,
  y,
  type = "parametric",
  conf.level = 0.95,
  bf.prior = 0.707,
  bf.message = TRUE,
  beta = 0.1,
  k = 2L,
  label.var = NULL,
  label.expression = NULL,
  point.label.args = list(size = 3),
  formula = y ~ x,
  smooth.line.args = list(size = 1.5, color = "blue"),
  method = "lm",
  method.args = list(),
  point.args = list(size = 3, alpha = 0.4),
  point.width.jitter = 0,
  point.height.jitter = 0,
  marginal = TRUE,
  marginal.type = "histogram",
  margins = "both",
  marginal.size = 5,
  xfill = "#009E73",
  yfill = "#D55E00",
  xparams = list(fill = xfill),
  yparams = list(fill = yfill),
  centrality.parameter = "none",
  centrality.label.args = list(size = 3),
  vline.args = list(color = xfill, size = 1, linetype = "dashed"),
  hline.args = list(color = yfill, size = 1, linetype = "dashed"),
  results.subtitle = TRUE,
  xlab = NULL,
  ylab = NULL,
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ggtheme = ggplot2::theme_bw(),
  ggstatsplot.layer = TRUE,
  ggplot.component = NULL,
  output = "plot",
  ...
)

Arguments

data

A dataframe (or a tibble) from which variables specified are to be taken. A matrix or tables will not be accepted.

x

The column in data containing the explanatory variable to be plotted on the x-axis. Can be entered either as a character string (e.g., "x") or as a bare expression (e.g, x).

y

The column in data containing the response (outcome) variable to be plotted on the y-axis. Can be entered either as a character string (e.g., "y") or as a bare expression (e.g, y).

type

Type of association between paired samples required (""parametric": Pearson's product moment correlation coefficient" or ""nonparametric": Spearman's rho" or ""robust": percentage bend correlation coefficient" or ""bayes": Bayes Factor for Pearson's r"). Corresponding abbreviations are also accepted: "p" (for parametric/pearson), "np" (nonparametric/spearman), "r" (robust), "bf" (for bayes factor), resp.

conf.level

Scalar between 0 and 1. If unspecified, the defaults return 95% lower and upper confidence intervals (0.95).

bf.prior

A number between 0.5 and 2 (default 0.707), the prior width to use in calculating Bayes factors.

bf.message

Logical that decides whether to display Bayes Factor in favor of the null hypothesis. This argument is relevant only for parametric test (Default: TRUE).

beta

bending constant (Default: 0.1). For more, see WRS2::pbcor().

k

Number of digits after decimal point (should be an integer) (Default: k = 2L).

label.var

Variable to use for points labels. Can be entered either as a character string (e.g., "var1") or as a bare expression (e.g, var1).

label.expression

An expression evaluating to a logical vector that determines the subset of data points to label. This argument can be entered either as a character string (e.g., "y < 4 & z < 20") or as a bare expression (e.g., y < 4 & z < 20).

point.label.args

A list of additional aesthetic arguments to be passed to ggrepel::geom_label_repel geom used to display the labels.

formula

Formula to use in smoothing function, eg. y ~ x, y ~ poly(x, 2), y ~ log(x). NULL by default, in which case method = NULL implies formula = y ~ x when there are fewer than 1,000 observations and formula = y ~ s(x, bs = "cs") otherwise.

smooth.line.args

A list of additional aesthetic arguments to be passed to ggplot2::geom_smooth geom used to display the regression line.

method

Smoothing method (function) to use, accepts either NULL or a character vector, e.g. "lm", "glm", "gam", "loess" or a function, e.g. MASS::rlm or mgcv::gam, stats::lm, or stats::loess. "auto" is also accepted for backwards compatibility. It is equivalent to NULL.

For method = NULL the smoothing method is chosen based on the size of the largest group (across all panels). stats::loess() is used for less than 1,000 observations; otherwise mgcv::gam() is used with formula = y ~ s(x, bs = "cs") with method = "REML". Somewhat anecdotally, loess gives a better appearance, but is \(O(N^{2})\) in memory, so does not work for larger datasets.

If you have fewer than 1,000 observations but want to use the same gam() model that method = NULL would use, then set method = "gam", formula = y ~ s(x, bs = "cs").

method.args

List of additional arguments passed on to the modelling function defined by method.

point.args

A list of additional aesthetic arguments to be passed to ggplot2::geom_point geom used to display the raw data points.

point.width.jitter, point.height.jitter

Degree of jitter in x and y direction, respectively. Defaults to 0 (0%) of the resolution of the data. Note that the jitter should not be specified in the point.args because this information will be passed to two different geoms: one displaying the points and the other displaying the labels for these points.

marginal

Decides whether ggExtra::ggMarginal() plots will be displayed; the default is TRUE.

marginal.type

Type of marginal distribution to be plotted on the axes ("histogram", "boxplot", "density", "violin", "densigram").

margins

Along which margins to show the plots. One of: [both, x, y].

marginal.size

Integer describing the relative size of the marginal plots compared to the main plot. A size of 5 means that the main plot is 5x wider and 5x taller than the marginal plots.

xfill, yfill

Character describing color fill for x and y axes marginal distributions (default: "#009E73" (for x) and "#D55E00" (for y)). The same colors will also be used for the lines denoting centrality parameters if centrality.parameter argument is set to TRUE. Note that the defaults are colorblind-friendly.

xparams

List of extra parameters to use only for the marginal plot along the x axis.

yparams

List of extra parameters to use only for the marginal plot along the y axis.

centrality.parameter

Decides which measure of central tendency ("mean" or "median") is to be displayed as vertical (for x) and horizontal (for y) lines. Note that mean values corresponds to arithmetic mean and not geometric mean.

centrality.label.args

A list of additional aesthetic arguments to be passed to the geom_label used to display the label corresponding to the centrality parameter and test value.

vline.args, hline.args

A list of additional aesthetic arguments to be passed to ggplot2::geom_vline and ggplot2::geom_hline geoms used to display the centrality parameter labels on vertical and horizontal lines.

results.subtitle

Decides whether the results of statistical tests are to be displayed as a subtitle (Default: TRUE). If set to FALSE, only the plot will be returned.

xlab

Labels for x and y axis variables. If NULL (default), variable names for x and y will be used.

ylab

Labels for x and y axis variables. If NULL (default), variable names for x and y will be used.

title

The text for the plot title.

subtitle

The text for the plot subtitle. Will work only if results.subtitle = FALSE.

caption

The text for the plot caption.

ggtheme

A function, ggplot2 theme name. Default value is ggplot2::theme_bw(). Any of the ggplot2 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 selected ggtheme (Default: TRUE). theme_ggstatsplot is an opinionated theme layer that override some aspects of the selected ggtheme.

ggplot.component

A ggplot component to be added to the plot prepared by ggstatsplot. This argument is primarily helpful for grouped_ variant of the current function. Default is NULL. The argument should be entered as a function.

output

If "expression", will return expression with statistical details, while "dataframe" will return a dataframe containing the results.

...

Currently ignored.

Note

  • If you set marginal = TRUE, the resulting plot can't be further modified with ggplot2 functions since it is no longer a ggplot object. In case you want a ggplot object, set marginal = FALSE. Also have a look at the ggplot.component argument.

  • The plot uses ggrepel::geom_label_repel to attempt to keep labels from over-lapping to the largest degree possible. As a consequence plot times will slow down massively (and the plot file will grow in size) if you have a lot of labels that overlap.

References

https://indrajeetpatil.github.io/ggstatsplot/articles/web_only/ggscatterstats.html

See also

Examples

# \donttest{ # to get reproducible results from bootstrapping set.seed(123) library(ggstatsplot) # creating dataframe with rownames converted to a new column mtcars_new <- as_tibble(mtcars, rownames = "car") # simple function call with the defaults ggstatsplot::ggscatterstats( data = mtcars_new, x = wt, y = mpg, type = "np", label.var = car, label.expression = wt < 4 & mpg < 20, centrality.parameter = "median" )
# }