Course contentsShow
← Data Science
Lesson 42 of 2,1452. Descriptive Statistics FoundationsFree lesson

Comparing Mean, Median, and Mode

Choosing the appropriate measure based on data type, distribution shape, and presence of outliers.

Comparing Mean, Median, and Mode

You'll learn how to choose the right measure of central tendency for your data.

When to Use Each Measure

Now that you know what mean, median, and mode are, the critical skill is knowing which one to use. Your choice depends on three key factors:

1. Data Type

  • Mode: Works with any data type (categorical, ordinal, numerical). It's your only option for nominal categories like "favorite color."
  • Median: Requires data you can order (ordinal or numerical).
  • Mean: Only makes sense for numerical data where arithmetic operations are meaningful.

2. Distribution Shape

  • Symmetric distributions (bell-shaped): Mean, median, and mode are roughly equal—use any, though mean is most common.
  • Skewed distributions (lopsided): The mean gets "pulled" toward extreme values. The median stays in the middle and better represents typical values.

3. Presence of Outliers

  • Outliers present: The median is robust—extreme values don't affect it. Mean gets distorted.
  • No outliers: Mean provides more information because it uses all data points.

Practical Example

Imagine house prices in a neighborhood: 200K,200K, 210K, 205K,205K, 198K, and one mansion at $5M.

  • Mean: ~$1.16M (misleading—no typical house costs this!)
  • Median: $205K (better represents the typical home)
  • Mode: Could be any exact price if repeated

The median wins here because outliers (the mansion) skew the mean.

Key Takeaway: Use the mode for categorical data, the median when you have outliers or skewed data, and the mean for symmetric numerical data without extreme values.