Using the ‘Median’ as a Measure of Central Tendency

A measure of central tendency is a single value that describes the way in which a set of data cluster around a central value. In other words, it is a way to describe the center of a data set.

The three main measures of central tendency are:

Median of the Distribution

In very simple terms; when your set of data is arranged in order from smallest to largest, the median is the literal middle of the data set. For an odd sized sample, the median is equal to the literal middle number. To find the median in an even sized sample, you have to add the two middle numbers and then divide the result by two.

median measure of central tendency

The median of a distribution of data is determined as the 50th percentile point once the distribution has been defined.

For normal distributions, the median value occurs at or close to the mean of the distribution. This is also where the mode, or high point (most frequently occurring value), of the distribution occurs.

For distributions that are skewed to the left, both the median and mode occur to the left of (are smaller than) the average of the distribution.

median measure of central tendency

For distributions that are skewed to the right, both the median and mode occur to the right of (are larger than) the average of the distribution.

median measure of central tendency

Finding skewed distributions in a Lean Six Sigma project?

I normally find skewed distributions in data representing time. For example, a call center’s time to resolve a level one help desk ticket. The call center is continuously trying to decrease the time to resolve a level one help desk ticket. The data set should show that most of the data should cluster closer to zero. This would represent skewed data that would have a central tendency best described by the median.

Let me know in the comments below, if you have ever encountered skewed distributions in your Lean Six Sigma Projects and how did you analyze them?