Stem and Leaf plots

Stem-and-leaf plots are a form of diagnostic graphic.  They give you good summary information about a given measurement variable.  However, they are rather ugly and difficult to understand if you are not familiar with them.  Therefore people typically avoid them if their audience is not familiar with statistics.

Stem-and-leaf plots are simple tallies.  All cases are tallied according to their two largest digits.  For example, if we asked 8 people how much money they currently had on their person, we might get the following answers:  $15, $8, $59, $22, $40, $40, $73, $4

A stem-and-leaf plot categorizes these answers by the tens place.  People who had less than $10 would be in the 0 tens place.  People who had between $11 and $19 would be in the 1 tens place.  People who had between $21 and $29 would be in the 2 tens place.  And so on.  So if I just tallied the 8 responses in this way, I would get:

0 | 1 1
1 | 1 
2 | 1
3 | 
4 | 1 1 
5 | 1
6 | 
7 | 1 

A stem-and-leaf plot complicates this tally just slightly.  Instead of putting in hash marks for each case, I actually put in the second digit of the case.  If I do this, I would get:

0 | 4 8
1 | 5 
2 | 2
3 | 
4 | 0 0 
5 | 9
6 | 
7 | 3 

This display is like a histogram tipped on its side.  It is actually more informative than a histogram in that you can see the actual value of each case up to two digits.  However, as I said above, this plot is ugly and not something you would want to use to communicate to an unsophisticated audience.

SPSS instructions