Skip to contents

Draws boxplots with configurable quantile type

Source: R/geom_j.R
geom_boxplot_j.Rd

geom_boxplot_j() draws boxplots whose statistics can follow different percentile definitions for quartiles. By default it uses quantile(type = 2) to follow PCTLDEF = 5 (SAS default). You can override the quantile definition via quantile_type. It uses StatBoxplotQuantile then renders with GeomBoxplot.

Usage

geom_boxplot_j(
  mapping = NULL,
  data = NULL,
  position = "dodge2",
  ...,
  coef = 1.5,
  quantile_type = 2,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

mapping, data, position, ...

Standard ggplot2 layer arguments passed to layer()/GeomBoxplot.

coef

Numeric multiplier for the IQR to compute fences. Defaults to 1.5 (Tukey).

quantile_type

Integer in 1:9 passed to stats::quantile(type = ...) to compute quartiles. Defaults to 2 (follows PCTLDEF = 5).

na.rm

Logical; if TRUE, silently removes NA values.

show.legend, inherit.aes

See ggplot2::layer().

Value

A ggplot2 layer that produces boxplots using the chosen quantile type.

Examples

library(ggplot2)
library(pharmaverseadamjnj)
#> Loading required package: pharmaverseadam
#> 
#> Attaching package: ‘pharmaverseadamjnj’
#> The following objects are masked from ‘package:pharmaverseadam’:
#> 
#>     adae, adcm, adeg, adex, adlb, adpc, adsl, advs
ggplot(advs, aes(AVISIT, AVAL, fill = TRT01A)) +
  geom_boxplot_j(position = position_dodge2(preserve = "single"), na.rm = TRUE) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))