Add Correlation Coefficients with P-values to a Scatter Plot

Add correlation coefficients with p-values to a scatter plot. Can be also used to add `R2`.

Usage

stat_cor(
  mapping = NULL,
  data = NULL,
  method = "pearson",
  alternative = "two.sided",
  cor.coef.name = c("R", "rho", "tau"),
  label.sep = ", ",
  label.x.npc = "left",
  label.y.npc = "top",
  label.x = NULL,
  label.y = NULL,
  label.y.step = 1.4,
  label.anchor = c("data", "panel"),
  output.type = "expression",
  digits = 2,
  r.digits = digits,
  p.digits = digits,
  r.accuracy = NULL,
  p.accuracy = NULL,
  r.leading.zero = NULL,
  p.format.style = "default",
  p.leading.zero = NULL,
  p.decimal.mark = NULL,
  p.coef.name = "p",
  conf.level = 0.95,
  geom = "text",
  position = "identity",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE,
  ...
)

Arguments

mapping

Set of aesthetic mappings created by aes(). If specified and inherit.aes = TRUE (the default), it is combined with the default mapping at the top level of the plot. You must supply mapping if there is no plot mapping.

data

The data to be displayed in this layer. There are three options:

If NULL, the default, the data is inherited from the plot data as specified in the call to ggplot().

A data.frame, or other object, will override the plot data. All objects will be fortified to produce a data frame. See fortify() for which variables will be created.

A function will be called with a single argument, the plot data. The return value must be a data.frame, and will be used as the layer data. A function can be created from a formula (e.g. ~ head(.x, 10)).

method

a character string indicating which correlation coefficient (or covariance) is to be computed. One of "pearson" (default), "kendall", or "spearman".

alternative

a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater" or "less". You can specify just the initial letter.

cor.coef.name

character. Can be one of "R" (pearson coef), "rho" (spearman coef) and "tau" (kendall coef). Uppercase and lowercase are allowed.

label.sep

a character string to separate the terms. Default is ", ", to separate the correlation coefficient and the p.value.

label.x.npc, label.y.npc

can be numeric or character vector of the same length as the number of groups and/or panels. If too short they will be recycled.

If numeric, value should be between 0 and 1. Coordinates to be used for positioning the label, expressed in "normalized parent coordinates".
If character, allowed values include: i) one of c('right', 'left', 'center', 'centre', 'middle') for x-axis; ii) and one of c( 'bottom', 'top', 'center', 'centre', 'middle') for y-axis.

If too short they will be recycled.

label.x, label.y

numeric Coordinates (in data units) to be used for absolute positioning of the label. If too short they will be recycled.

label.y.step

numeric value giving the vertical spacing (in text-line units) between the labels of successive groups when several groups are present. Default is 1.4 (unchanged behavior). Set label.y.step = 0 to stop the per-group vertical shift, so that labels align across facet panels when a factor is mapped to an aesthetic that also defines the facets. This places the labels flush with the top of each panel; it is similar in spirit to the aes(group = 1) workaround, which instead leaves them one text line lower.

label.anchor

character. How label.x.npc/label.y.npc are interpreted. "data" (default) converts them to data coordinates using the data range, so the label follows the data; the output is unchanged from previous versions. "panel" places the label at the true panel-relative position (npc), so the labels stay aligned across panels/facets whose axis ranges differ - e.g. with facet_wrap(scales = "free_y"), with geom_smooth() extending each panel by a different amount, or across separate plots combined with ggarrange(). An explicit label.x/label.y always stays in data units regardless of label.anchor. Requires ggplot2 >= 3.5.0 (already a dependency).

output.type

character One of "expression", "latex", "tex" or "text".

digits, r.digits, p.digits

integer indicating the number of decimal places (round) or significant digits (signif) to be used for the correlation coefficient and the p-value, respectively. In the default p-value label style, the displayed p-value keeps the legacy raw value unless p.accuracy is set; p.digits is used by non-default p.format.style outputs and the computed p variable.

r.accuracy

a real value specifying the number of decimal places of precision for the correlation coefficient. Default is NULL. Use (e.g.) 0.01 to show 2 decimal places of precision. If specified, then r.digits is ignored.

p.accuracy

a real value specifying the number of decimal places of precision for the p-value. Default is NULL. Use (e.g.) 0.0001 to show 4 decimal places of precision. If specified, then p.digits is ignored.

r.leading.zero

logical. Whether to include the leading zero before the decimal point in the correlation coefficient (e.g., "0.73" vs ".73"). Default (NULL) keeps the leading zero; set to FALSE for APA-style reporting.

p.format.style

character specifying the p-value formatting style. One of "default", "apa", "nejm", "lancet", "ama", "graphpad", "scientific". Default is "default" for backward compatibility.

p.leading.zero

logical. Whether to include leading zero before decimal point (e.g., "0.05" vs ".05"). If NULL, uses the style's default setting.

p.decimal.mark

character string to use as the decimal mark. If NULL, uses getOption("OutDec").

p.coef.name

character. Symbol used for the p-value label. Default is "p"; use "P" for an uppercase p-value label. For output.type = "expression" this should be a single valid plotmath symbol (e.g. "P"), since it is parsed as an expression.

conf.level

confidence level for the confidence interval of the correlation coefficient, used to compute the conf.int.low, conf.int.high and conf.int.label computed variables (see the Computed variables section). Default is 0.95. A confidence interval is only available for method = "pearson" (with at least 4 complete observations); for "spearman"/"kendall" the confidence-interval variables are NA.

geom

The geometric object to use to display the data for this layer. When using a stat_*() function to construct a layer, the geom argument can be used to override the default coupling between stats and geoms. The geom argument accepts the following:

A Geom ggproto subclass, for example GeomPoint.
A string naming the geom. To give the geom as a string, strip the function name of the geom_ prefix. For example, to use geom_point(), give the geom as "point".
For more information and other ways to specify the geom, see the layer geom documentation.

position

A position adjustment to use on the data for this layer. This can be used in various ways, including to prevent overplotting and improving the display. The position argument accepts the following:

The result of calling a position function, such as position_jitter(). This method allows for passing extra arguments to the position.
A string naming the position adjustment. To give the position as a string, strip the function name of the position_ prefix. For example, to use position_jitter(), give the position as "jitter".
For more information and other ways to specify the position, see the layer position documentation.

na.rm

If FALSE (the default), removes missing values with a warning. If TRUE silently removes missing values.

show.legend

logical. Should this layer be included in the legends? NA, the default, includes if any aesthetics are mapped. FALSE never includes, and TRUE always includes. It can also be a named logical vector to finely select the aesthetics to display. To include legend keys for all levels, even when no data exists, use TRUE. If NA, all levels are shown in legend, but unobserved levels are omitted.

inherit.aes

If FALSE (the default for most ggpubr functions), overrides the default aesthetics, rather than combining with them. This is most useful for helper functions that define both data and aesthetics and shouldn't inherit behaviour from the default plot specification. Set to TRUE to inherit aesthetics from the parent ggplot layer.

...

other arguments to pass to geom_text or geom_label.

Computed variables

r: correlation coefficient
rr: correlation coefficient squared
rmse: root mean square deviation (RMSE/RMSD) between x and y, computed on the complete pairs; meaningful when x and y are on the same scale (e.g. predicted vs. reference values). Rounded and formatted with the coefficient settings (r.digits / r.accuracy).
r.label: formatted label for the correlation coefficient
rr.label: formatted label for the squared correlation coefficient
p.label: label for the p-value
rmse.label: formatted label for the RMSE/RMSD
conf.int.low, conf.int.high: lower and upper bounds of the confidence interval of the correlation coefficient (Pearson only; NA for Spearman/Kendall), at the level given by conf.level
conf.int.label: formatted label for the confidence interval, e.g. "95% CI [0.21, 0.75]" (NA when the interval is unavailable)
label: default label displayed by stat_cor()

Examples

# Load data
data("mtcars")
df <- mtcars
df$cyl <- as.factor(df$cyl)

# Scatter plot with correlation coefficient
# :::::::::::::::::::::::::::::::::::::::::::::::::
sp <- ggscatter(df,
  x = "wt", y = "mpg",
  add = "reg.line", # Add regressin line
  add.params = list(color = "blue", fill = "lightgray"), # Customize reg. line
  conf.int = TRUE # Add confidence interval
)
# Add correlation coefficient
sp + stat_cor(method = "pearson", label.x = 3, label.y = 30)


# Specify the number of decimal places of precision for p and r
# Using 3 decimal places for the p-value and
# 2 decimal places for the correlation coefficient (r)
sp + stat_cor(p.accuracy = 0.001, r.accuracy = 0.01)


# Show only the r.label but not the p.label
sp + stat_cor(aes(label = after_stat(r.label)), label.x = 3)


# Use R2 instead of R
ggscatter(df, x = "wt", y = "mpg", add = "reg.line") +
  stat_cor(
    aes(label = paste(after_stat(rr.label), after_stat(p.label), sep = "~`,`~")),
    label.x = 3
  )


# Show the RMSE/RMSD (root mean square deviation) between x and y
# (useful for agreement between paired measurements on the same scale)
sp + stat_cor(aes(label = after_stat(rmse.label)), label.x = 3)


# Combine the correlation coefficient and the RMSE (comma-separated)
sp + stat_cor(
  aes(label = paste(after_stat(r.label), after_stat(rmse.label), sep = "~`,`~")),
  label.x = 3
)


# Show the confidence interval of the correlation coefficient (Pearson)
sp + stat_cor(aes(label = after_stat(conf.int.label)), label.x = 3)


# Correlation coefficient with its confidence interval
sp + stat_cor(
  aes(label = paste(after_stat(r.label), after_stat(conf.int.label), sep = "~`,`~")),
  label.x = 1.5
)


# Color by groups and facet
# ::::::::::::::::::::::::::::::::::::::::::::::::::::
sp <- ggscatter(df,
  x = "wt", y = "mpg",
  color = "cyl", palette = "jco",
  add = "reg.line", conf.int = TRUE
)
sp + stat_cor(aes(color = cyl), label.x = 3)

Add Correlation Coefficients with P-values to a Scatter Plot

Usage

Arguments

Computed variables

See also

Examples