A stemandleaf plot shows all the data values from a sample
set. It was invented by John Tukey in the
1960's. John Tukey also invented the boxandwhisker plot in the
1970's.
Example of a stemandleaf plot:
Quality Ratings for Natural Peanut Butter:
Stem 
Leaves 
3 
4 
4 
0 
5 
2 7 7 
6 
0 0 3 7 9 9 9 
7 
1 
8 
9 
Key:
means 34.
All the data can be recovered from this plot by
putting the stems and leaves together:
34, 40, 52, 57, 57, 60, 60, 63, 67, 69, 69, 69, 71, 89.
Example of sidebyside stemandleaf plots: Suppose we wanted to compare the quality ratings of natural and regular peanut butter. Since the quality ratings were made using the same scale, we can create sidebyside stemandleaf plots. This means that the stem will be the same for each type of peanut butter, and so it will be put in the middle of our plot. The leaves will go to the left and right of the stems. The leaves to the left will help represent the quality ratings of the natural peanut butter, and the leaves to the right will help represent the quality ratings of the regular peanut butter.
Natural PB Leaves 
Stems 
Regular PB Leaves 

1 
1 

2 
3 3 6 9 
3 
3 
1 1 3 4 4 5 
0 
4 
0 0 3 5 6 9 
2 7 7 
5 
4 4 
0 0 3 7 9 9 9 
6 
0 
1 
7 
6 
9 
8 
3 3 
Key:
means 11
All the data can be recovered from this plot by
putting the stems and leaves together: Natural peanut butter quality ratings: 33, 40, 52, 57, 57, 60, 60, 63, 67, 69, 69, 69, 71, 89 Regular peanut butter quality ratings: 11, 23, 23, 26, 29, 31, 31, 33, 34, 34, 35, 40, 40, 43, 45, 46, 49, 54, 54, 60, 76, 83, 83
It may be hard to create a stemandleaf plot if you have a
large amount of data. A histogram would then
be an easier choice. The histogram, like the stemandleaf plot, shows the shape
of the data, but unlike a stemandleaf plot, you can not recover the data from
a histogram. To see how to create the above
stemandleaf plots, Click
Here.
