Statistics 101 – Medians

A median value is a way to represent what is typical or common in a set of data, by identifying the value in the middle of the set. Property values, population ages, and company salaries are just a few examples of numbers we can calculate a median for when looking at statistics.

The median is simply found by arranging all numbers in a set of data in order and identifying the value in the exact middle of the list. Sometimes, a set of data might have some extreme values or “outliers”. Think of the property values of a residential street, where one property is significantly more expensive than the others. Calculating the mean average of this dataset would be skewed by the really expensive property. This result might appear greater than what is actually typical on that street. Using the median value instead, offers a figure that better represents what is typical without being influenced by some unusually high or low values.

In the Correctional Service of Canada, we use medians quite often to observe trends in offender data. For example, let’s look at the duration of offender employment in the community. In a particular year, some offenders may not have been employed for very long, while others may have been employed for a full year. When trying to identify what is the typical duration of employment, these extreme values can skew results if the mean average is used. Instead, the results are calculated with the median, as it provides a better representation of how many days most offenders in the community were employed.

In CSC reporting, medians are often used to show what is typical in our offender data. This type of calculation helps decision-makers monitor the organization’s performance, identify shifts in results over time, and make better, informed decisions in relation to corporate objectives. When observing reports, be sure to watch for this icon so you know when you are looking at median values.