Mean Absolute Deviation- Calculate MAD Step-by-Step
What Is Mean Absolute Deviation?
Mean Absolute Deviation (MAD) measures how spread out values are in a dataset. Unlike variance or standard deviation, it gives you the average distance between each data point and the mean.
It's simple: find how far each number is from the average, ignore whether it's above or below, then average those distances.
Statisticians use MAD when they want a measure of spread that's less sensitive to outliers than standard deviation. It's also easier to explain to non-statisticians.
Mean Absolute Deviation Formula
The formula looks like this:
MAD = Σ|x - μ| / n
Where:
- Σ means "sum of"
- x is each individual value
- μ (mu) is the mean of all values
- n is the count of values
- The bars | | mean "absolute value" — drop any negative signs
How to Calculate MAD: Step-by-Step
Step 1: Find the Mean
Add all values together and divide by how many values you have.
Example dataset: 4, 7, 2, 9, 5
Mean = (4 + 7 + 2 + 9 + 5) / 5 = 27 / 5 = 5.4
Step 2: Find Each Deviation
Subtract the mean from each value. Don't worry about negative signs yet.
- 4 - 5.4 = -1.4
- 7 - 5.4 = 1.6
- 2 - 5.4 = -3.4
- 9 - 5.4 = 3.6
- 5 - 5.4 = -0.4
Step 3: Take Absolute Values
Convert every deviation to a positive number.
- |-1.4| = 1.4
- |1.6| = 1.6
- |-3.4| = 3.4
- |3.6| = 3.6
- |-0.4| = 0.4
Step 4: Sum the Absolute Deviations
1.4 + 1.6 + 3.4 + 3.6 + 0.4 = 10.4
Step 5: Divide by the Number of Values
MAD = 10.4 / 5 = 2.08
The mean absolute deviation is 2.08. On average, values in this dataset sit 2.08 units away from the mean.
MAD vs. Standard Deviation: What's the Difference?
Both measure spread, but they treat outliers differently.
| Measure | How It Handles Deviations | Sensitivity to Outliers | Use When |
|---|---|---|---|
| MAD | Uses absolute values (keeps all deviations positive) | Lower — outliers have less impact | Data has extreme values or you want a robust measure |
| Standard Deviation | Squares deviations (makes outliers count more) | Higher — outliers heavily influence the result | Data is normally distributed and you want traditional statistics |
Standard deviation squares each deviation before averaging. A value that's 10 units from the mean contributes 100 to the calculation. MAD only counts it as 10.
This makes MAD more resistant to distortion from extreme values.
When to Use Mean Absolute Deviation
- Forecasting accuracy — Common in supply chain and demand planning. MAD tells you the average error size in your predictions.
- Budget variance analysis — Shows how actual spending typically deviates from budgeted amounts.
- Quality control — Measures consistency in manufacturing processes.
- Any dataset with outliers — When standard deviation would give a misleading picture.
Practical Example: Forecasting Demand
You forecast weekly sales of 100, 120, 95, 110, 105 units. Actual sales were 95, 140, 90, 115, 100.
Errors: -5, +20, -5, +5, -5
Absolute errors: 5, 20, 5, 5, 5
MAD = (5 + 20 + 5 + 5 + 5) / 5 = 8
Your forecasts are off by 8 units on average. That's a useful, interpretable number for planning inventory buffers.
Common Mistakes to Avoid
- Forgetting to use absolute values — If you average raw deviations, you'll always get zero. That's useless.
- Confusing MAD with mean deviation — Same thing, but some textbooks call it "average absolute deviation."
- Using MAD when standard deviation is expected — Industries have conventions. Finance defaults to standard deviation. Supply chain often uses MAD.
Quick Reference
| Term | Symbol | What It Means |
|---|---|---|
| Mean | μ | Average of all values |
| Deviation | x - μ | Distance from the mean |
| Absolute value | | | | Positive version of a number |
| MAD | — | Average of all absolute deviations |
Bottom Line
Mean Absolute Deviation gives you the average distance from the mean. It's straightforward to calculate and interpret.
Use it when you need a robust measure of spread that won't be skewed by outliers. Use standard deviation when your data is well-behaved and follows a normal distribution.
The calculation is five steps: find the mean, subtract it from each value, take absolute values, sum them, then divide by the count. That's it.