• Mergers and Acquisitions ,
• 02 Mar

When thinking about data and how we work with it, it’s important to understand the difference between discrete and continuous variables as this is an important parameter in the questions that we ask. It is the question and the unknown that starts our journey. How we frame our question and the information we seek to know signals to us what we need to do.

This article is a quick definition reference guide intended to assist you with understanding the difference because it is important. If you want to learn more about these subjects, there are thousands of videos, charts, and pictures to explore as you go down the rabbit hole.

Discrete Variables

Discrete variables are like buckets. Think of them like sides of a regular 6-sided die. The only values that rolling this die will produce are 1, 2, 3, 4, 5, 6. We can’t roll a 2.6 nor can we roll a negative 4.  The same goes with flipping a coin. Now we can have a weighted die with different probabilities for each number and we can have different number of times that we roll this die or flip a coin, but the result of each event will always be one of these discrete values. It can be categorical as well as counts. For instance, the number of times a 6 is rolled, or the number of green haired students at a university. In Excel or a BI software like Tableau, you can filter and sort on discrete variables.

Continuous Variables

Continuous variables have an infinite number of possible values. It is the opposite of a discrete variable. We can look at data, such as with a histogram, where we group continuous variables, but the raw data itself is not limited to buckets to begin with. In this case, a 2.6 is possible because there are no buckets like we have with the die. How much a car weighs, the time it takes a piece of software to complete a task are examples. The values we see can be anything and depending on how many decimal places we want to go out to, there are potentially an infinite number of values continuous variables can be.

Why is this important?

Data is either Continuous or Discrete. How we statistically analyze data depends on if it’s discrete or continuous. You’ll typically see Continuous Distributions as smooth lined bell curves. Discrete distributions may have a similar normal distribution, but it is blocked into specific occurrences. You can look at some data in either way. Dates for instance you can block into months, or look at Sales over time on a continuous line.  In Tableau, you’d change the date dimension from discrete to continuous to get this effect.