Stem-and-leaf display

The root - leaf plot (also branch leaves or stem - leaf plot and in English stem- and- leaf plot or stemplot ) is a graphical tool of descriptive and exploratory statistics. As a developer, this type of diagram applies John W. Tukey.

Similar to the histogram, or box plot the root - leaf plot of the visualization of frequency distributions used. Unlike those, the representation of the values ​​of each observation is obtained with the desired accuracy. From a root -leaf plot to statistical metrics such as modal, median and percentiles can be read. However, this type of representation encountered in a large number of features to their limits.

Construction

The diagram consists of two columns. The left column shows as " tribes " the equivalence classes into which the features on the right side shown as "leaves" are divided. Typical is a class formation after the decimal, but other divisions are possible, for example, to select the first two digits as the root. As a rule of thumb for the number of equivalence classes is considered, where n is the number of records.

Example

A series of measurements has the following, already ordered, data:

If you choose the natural numbers as the classification is as given below stem leaf plot. Thus, the diagram can be read correctly, the indication of the unit used to create is important:

3 | 5 7 2 | 5 5 6 7 8 1 | 0 | 3 4 Unit = 1 With a modified stem - leaf plot can visually compare two distributions. Suppose there is a second series of measurements with the following data:

Now you can write this on the left side of the " tribe ":

| 3 | 5 7 1 1 9 | 2 | 5 5 6 7 8    2 8 | 1 |      4 | 0 | 3 4 Unit = 1 References