Measures of Dispersion

It is important in the practice of statistics, Six Sigma projects, and analysis of data. First understand the central tendency of your data as does the data tend to be a symmetrical bell curve shape once you put it into the form of a histogram? or Is it severely skewed left or right?

From there, you want to understand the dispersion.

• How far does the data stray from the central point that you’re looking for?
• How far out to the left or right?
• You also need to know, within one standard deviation, how much dispersion is in there that would take in 66 to 67% of the entire data population for your project.

You can see some characteristics of dispersion when you compare two example histograms. The histograms display the different call handle times for a technical support center. They have approximately the same mean value but the second histogram has a greater range than the first one. You could have essentially the same number of observations and maybe even the same mean value at the center, but you could have a big difference in the dispersion between them.

There are three important measures of dispersion:

• standard deviation, which measures the amount of dispersion from the average, left or right, from the center value in your data
• range, which is essentially the spread of the data, or how far to the left and to the right of your central measurement it goes
• variance, which is the square of the standard deviation and characterizes the difference between the individual measurements in the entire study