CODEWITHSUNDEEP
rboxplot
Manish
Manish

Reputation: 3521

How to put values on a boxplot for median, 1st quartile and last quartile?

How to put values on boxplot and control its width? 

X<-c(1,2,,3,4,4,5,5,6,6,6,6,6,7)

I need to write values for min, max, 1st quartile, median and last quartile. How can I put it there?

Upvotes: 8

Views: 64506

Answers (4)

Tuong Nguyen
Tuong Nguyen

Reputation: 143

boxplot(X, axes = FALSE, staplewex = 1)
text(y = boxplot.stats(X)$stats, labels = boxplot.stats(X)$stats, x = 1.25)

How to do this but in multiple Boxplot

Example data:

a <- data.frame(
    Rank = c(1,2,3,1,2,3,1,2,3,3,2,1,3,2,1,2,1,3,3,3,1,1,2,2,1,3 ),
    Time = c(1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2,3,3,3,3,3,3,3,3), 
    Value = c(5,10,15,20,30,50,70,80,100,5,7,9,11,15,17,19,17,19,100,200,300,400,500,700,1000,200))

boxplot(a$Value[a$Rank==1] ~ a$Time[a$Rank==1])

Upvotes: 0

Fernando Fernandes
Fernando Fernandes

Reputation: 23

I had some problems trying to understand the boxplot quartiles labels and compare to summary function values, so i´d like to share with you.

Sometimes there will be differences in quartiles labels in boxplot using fivenum or stats comparing to r summary function values. This occurs only on even datasets.

e.g.(using text(x = boxplot.stats(X)$stats, labels = boxplot.stats(X)$stats, y = 1.25)):

product<-c(3,12,20,25,30,35,70,70,80,150)

summary(product)
Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
3.00   21.25   32.50   49.50   70.00  150.00 

boxplot(product,horizontal=TRUE,col="grey",staplewex=1,axes=FALSE)
text(x = boxplot.stats(product)$stats, labels = boxplot.stats(product)$stats, y = 1.25)

BoxplotFiveNum

As you can see in the picture, the values do not match.

In this case you can use quantile function:

text(x=quantile(produto),labels=quantile(produto),y=1.25)

BoxplotQuantile

Now you should get the same values as listed on summary function. Otherwise just ignore summary function and use fiveNum values instead. The differences occurs due to lack of universal agreement among statisticians.

A google search will show you the differences on quantiles calculation.

Upvotes: 1

mpalanco
mpalanco

Reputation: 13570

The answer provided by mnel is perfect, assuming —as in the example— that there are not outliers. In that case, the lower and upper whiskers are then equal to the minimum and maximum. But, if there are outliers and we don't specify range = 0 in the boxplot function, we don't get the right values for the labels. Instead, we could use:

text(x=boxplot.stats(X)$stats, labels = boxplot.stats(X)$stats, y = 1.25)

Let's see an example:

Horizontally

X <- c(1,2,3,3,4,4,5,5,6,6,6,6,10,15)
boxplot(X, horizontal = TRUE, axes = FALSE, staplewex = 1)
  1. Incorrect valuestext(x = fivenum(X), labels = fivenum(X), y = 1.25)

enter image description here

  1. Correct values: text(x = boxplot.stats(X)$stats, labels = boxplot.stats(X)$stats, y = 1.25)

enter image description here

Vertically

Switching the arguments for x and y inside text.

boxplot(X, axes = FALSE, staplewex = 1)
text(y = boxplot.stats(X)$stats, labels = boxplot.stats(X)$stats, x = 1.25)

enter image description here

Upvotes: 17

mnel
mnel

Reputation: 115390

You can use horizontal = TRUE get a horizontal boxplot and axes = FALSE to remove the axes. staplewex = 1 sets the staple width the same as the box width

Then you can use fivenum to return the statistics used to create the boxplot and use these as text labels, fiddling with the y value until you have what you want

boxplot(X, horizontal = TRUE, axes = FALSE, staplewex = 1)
text(x=fivenum(X), labels =fivenum(X), y=1.25)

enter image description here

Note that i've inserted a 3 for the value missing in your example data X

Upvotes: 20

Related Questions