How to Make a Frequency Table
Make a frequency table with data from Exercise 2.3.8.
Class Interval Frequency Rel. Frequence ------------------------------------------- 10 - 19 2 .07 20 - 29 4 .13 30 - 39 9 .30 40 - 49 7 .23 50 - 59 6 .20 60 - 69 2 .07 ------------------------------------------- 30 1.00The relative frequencies are determined by dividing the frequencies by the total number of observations.
For each class interval, graph a bar that spans the entire class interval whose height is equal to the frequency (or the relative frequency).
The area of a bar over a class interval relative to the total area corresponds to the proportion of the data in that class interval.
Make a histogram with the data from Exercise 2.3.8. The boundaries should be at 9.5, 19.5, 29.5, 39.5, 49.5, 59.5, and 69.5 for the frequency table listed above.
A histogram is approximately symmetric is the left and right sides are approximately mirror-images of one another.
A histogram is skewed to the right if the right half is stretched out much farther than the left half.
A histogram is skewed to the left if the left half is stretched out much farther than the right half.
The histogram from this example is approximately symmetric. Describing it with a very mild skewness to the right is also acceptable.
The last significant digit of each observation is called the leaf. There is exactly one leaf per observation written on the right side of the display. For each row (stem) the leaves are ordered from smallest to largest. The leaves should line up exactly, and be flush with the vertical line, to give a visual description of how many observations are in each row.
The remaining portion of each observation is called the stem. Each stem is written only once, on the left side of the vertical line. Even if there are no observations for a stem, the stem should be written to show that there is a gap in the data.
An example is the easiest way to learn how to do it. Do data from Exercise 2.3.8 in class.
1|02 2|1223 3|245566789 4|3455556 5|345567 6|04There are 30 observations. The fourth smallest is 22, while the largest is 64.
When the data has too many significant digits, better results come if the data is rounded first. Look at the first two columns of data from Exercise 2.3.9. Here is a stem-and-leaf display after rounding.
17|1 18| 19|5 20|13 21|04 22|2 23| 24|089If the data will all fall onto a very small number of stems, stems can be split. The options are splitting each stem into two (with leaves of 0-4 on the top stem and leaves of 5-9 on the bottom stem) or into five (with leaves of 0-1, 2-3, 4-5, 6-7, 8-9). Here are the first three columns of Exercise 2.3.5 shown both ways.
0|11334 0|5578 1|01223 1|7or
0|11 0|33 0|455 0|7 0|8 1|01 1|223 1| 1|7
Bret Larget, larget@mathcs.duq.edu