Boxplot with Summarized and Grouped Data in R

Question

I have the following pre-summarized cost data:

MeanCost	Std	MedianCost	LowerIQR	UpperIQR	StatusGroup	AgeGroup
700	500	650	510	780	Dead	Young
800	600	810	666	1000	Alive	Young
500	200	657	450	890	Comatose	Young
300	400	560	467	670	Dead	Old
570	600	500	450	600	Alive	Old
555	500	677	475	780	Comatose	Old
333	455	300	200	400	Dead	Middle
678	256	600	445	787	Alive	Middle
1500	877	980	870	1200	Comatose	Middle

I wish to create a boxplot with this information - similar to the one below. Where each Color represents Status Group (blue=dead, read=alive, green=comatose). And each grouped cluster represents an age group (left cluster=young, middle cluster=middle, right cluster=old).

I know that I don't have min and max, so whiskers are not necessary.
I want to code this in R, and any help would be appreciated! Thank you.

Here is the code I have tried:

 dattest<- data.frame(
  Mean_Cost = c(700,800,500,300,570,555,333,678,1500), 
  Std = c(500,600,200,400,600,500,455,256,877), 
  Median_Cost = c(650,810,657,560,500,677,300,600,980), 
  LowerIQR = c(510,666,450,467,450,475,200,445,870), 
  UpperIQR = c(780,1000,890,670,600,780,400,787,1200), 
  StatusGroup = c(1,2,3,1,2,3,1,2,3),
  AgeGroup = c(1,1,1,2,2,2,3,3,3))

where for StatusGroup 1=dead, 2=alive, 3-comatose
and for AgeGroup 1=young, 2=old, 3=middle

 ggplot(dattest, aes(xmin = AgeGroup-.25, xmax=AgeGroup+.25, ymin=LowerIQR, ymax=UpperIQR)) + 
    geom_rect(fill="transparent", col = "blue") + 
    geom_segment(aes(y=Median_Cost, yend=Median_Cost, x=AgeGroup-.25, xend=AgeGroup+.25), col="blue") + 
    geom_point(mapping=aes(x = StatusGroup, y = Mean_Cost), col="red") +
    scale_x_continuous(breaks=1:3, labels=c("Young","Old","Middle")) + 
    theme_classic()

And this code is definitely not giving me what I want

possible duplicate: https://stackoverflow.com/questions/22212885/producing-a-boxplot-in-ggplot2-using-summary-statistics. Please don't post data as an image. It's easier to help you if you include samples data in a [reproducible format](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) such as `dput()` so we can copy/paste the data into R rather than retyping it to test possible solutions. — MrFlick, Apr 26 '21 at 02:58
Thank you - I edited it so that it a table that you can copy and paste into a CSV. I also do not think this is a duplicate because the previous post does not address how to plot the data with multiple groupings like this data. — KKolo, Apr 26 '21 at 03:11
The "grouping" you are looking for is just a standard `fill=` aesthetic. I think it would work if you just tried it. Add the code you tried to your question. — MrFlick, Apr 26 '21 at 03:12
I am able to do it with one group - but not two, that's where I get stuck. — KKolo, Apr 26 '21 at 03:14
Then please add the code you tried for one group. It's much easier to edit and fix what you start then begin from scratch. — MrFlick, Apr 26 '21 at 03:15

pete_999999999 · Accepted Answer · 2021-04-26T04:37:12.780

Is this what you are trying to do?

library(tidyverse)
df <- tibble::tribble(
  ~MeanCost, ~Std, ~MedianCost, ~LowerIQR, ~UpperIQR, ~StatusGroup, ~AgeGroup,
       700L, 500L,        650L,      510L,      780L,       "Dead",   "Young",
       800L, 600L,        810L,      666L,     1000L,      "Alive",   "Young",
       500L, 200L,        657L,      450L,      890L,   "Comatose",   "Young",
       300L, 400L,        560L,      467L,      670L,       "Dead",     "Old",
       570L, 600L,        500L,      450L,      600L,      "Alive",     "Old",
       555L, 500L,        677L,      475L,      780L,   "Comatose",     "Old",
       333L, 455L,        300L,      200L,      400L,       "Dead",  "Middle",
       678L, 256L,        600L,      445L,      787L,      "Alive",  "Middle",
      1500L, 877L,        980L,      870L,     1200L,   "Comatose",  "Middle"
  )

df %>% 
  mutate(AgeGroup = factor(AgeGroup, levels = c("Young", "Middle", "Old"))) %>% 
  ggplot(aes(x = AgeGroup, fill = StatusGroup)) +
  geom_boxplot(aes(
    lower = LowerIQR, 
    upper = UpperIQR, 
    middle = MedianCost, 
    ymin = MedianCost - Std, 
    ymax = MedianCost + Std),
    stat = "identity", width = 0.5)

Edit

To add an "x" at the mean you can adjust the position:

df %>% 
  mutate(AgeGroup = factor(AgeGroup, levels = c("Young", "Middle", "Old"))) %>% 
  ggplot(aes(x = AgeGroup, fill = StatusGroup)) +
  geom_boxplot(aes(
    lower = LowerIQR, 
    upper = UpperIQR, 
    middle = MedianCost, 
    ymin = MedianCost - Std, 
    ymax = MedianCost + Std),
    stat = "identity", width = 0.5) +
  geom_point(aes(y = MeanCost),
             position = position_dodge(width = 0.5),
             shape = 4)

Yes - just about! I want to add a point for the mean, but when I add In the 'geom_point' code , it adds the mean markers next to the boxplots, and not on top of them. — KKolo, Apr 26 '21 at 03:48

Boxplot with Summarized and Grouped Data in R

1 Answers1

Edit