Grouped box plot with multiple variables

[guybrettrobertson]

Is there a way to create a grouped box plot with multiple grouping variables using {echarts4r}?

I would like to create a grouped box plot with multiple variables comparable to this {ggplot2} plot: https://www.r-graph-gallery.com/265-grouped-boxplot-with-ggplot2.html

I have tried the following:

variety = rep(LETTERS[1:7], each=40)
treatment = rep(c("high","low"),each=20)
note = seq(1:280)+sample(1:150, 280, replace=T)

data.frame(variety, treatment ,  note) |>
  echarts4r::group_by(variety, treatment) |>
  echarts4r::e_charts() |>
  echarts4r::e_boxplot(note) |>
  echarts4r::e_legend()

But it gives me this:

I have also tried using e_add_nested but it doesn't seem to work for boxplots.

[guybrettrobertson]

I just noticed that has already come up in issue 253 but no clean solution was available at the time. Have any developments been made on this?

[sandra-ab]

I would love this feature too. I've been using bar charts as a workaround (where the multiple grouping works as expected), but often boxplots are just so much more appropriate. :)

[JohnCoene]

#253

[munoztd0]

Created a gist function to do it -> https://gist.github.com/munoztd0/c860717d1f29e8a1b9114c46e945202e

#253 (comment)

[sharp-lilac]

Hi, any news on this? Boxplots are so important for data visualizations and not being able to group by multiple variables is really surprising. Any workarounds?

[munoztd0]

@sharp-lilac I agree with you is just that boxplots are not "normal" plots they are constituted of various diverse source of information (aggregated statistics as well a raw data points, i.e. outliers.)

I started devloping a new branch that will implement a new approach for boxplots but it might get ugly at first, the commit 06c2e5d shows the first steps.

On the branch grouped_boxplot:

library(echarts4r)

data <- data.frame(
  category = rep(LETTERS[1:5], each = 100),
  value = rnorm(500, mean = rep(1:5, each = 100), sd = 1:5),
  gender = sample(c("M", "F"), 500, replace = TRUE)
)

plot <- data |>
  group_by(gender) |>
  e_charts(category) |>
  e_boxplot_group_(serie = "value") |>
  e_x_axis(name = "Category") |>
  e_y_axis(name = "Value") |>
  e_tooltip(trigger = "axis") |>
  e_legend(show = TRUE)

plot

^{Created on 2024-11-25 with reprex v2.0.2}

[munoztd0]

Hey, glad it kinda works.
Yeah I know, I haven't finished implementing the outliers. I have to refactor it somehow, work in progress.

[sharp-lilac]

@munoztd0 Thank you! Exciting