Formula for Average Deviation- Statistical Measurement

What Average Deviation Actually Is

Average deviation (also called mean absolute deviation) tells you how spread out numbers are in a dataset. That's it. It's the average distance between each data point and the mean.

Most people confuse it with standard deviation. They're related, but average deviation is simpler—it ignores whether values are above or below the mean. You just measure the distance.

Statisticians don't use this as often as standard deviation, but it's still useful when you want a straightforward measure of variability without complex math.

The Formula

Here's the formula for average deviation:

AD = (Σ|x - μ|) / n

Where:

The absolute value signs are the key here. They force all distances to be positive, so you don't get negatives canceling out positives.

How to Calculate Average Deviation: Step by Step

Let's work through a real example. Say you have test scores: 70, 75, 80, 85, 90

Step 1: Find the mean

Add them up: 70 + 75 + 80 + 85 + 90 = 400

Divide by 5: 400 / 5 = 80

Step 2: Find each distance from the mean (ignore negatives)

Step 3: Add those distances

10 + 5 + 0 + 5 + 10 = 30

Step 4: Divide by the number of values

30 / 5 = 6

The average deviation is 6. On average, each score sits 6 points away from the mean of 80.

Average Deviation vs Standard Deviation

These two measure the same thing—spread—but differently.

Feature Average Deviation Standard Deviation
Formula complexity Simpler More complex (squares, square roots)
Sensitivity to outliers Less sensitive More sensitive (squares large values)
Mathematical properties Weaker for advanced stats Stronger (used in regression, ANOVA)
Interpretability Directly interpretable Slightly less intuitive
Common usage Less common Industry standard

Standard deviation wins in most statistical applications because it has better mathematical properties. Average deviation is easier to explain to non-statisticians.

When to Use Average Deviation

Average deviation makes sense when:

Skip it when:

Quick Reference: The Process

  1. Calculate the mean of your dataset
  2. Subtract the mean from each data point
  3. Take the absolute value of each result
  4. Sum all absolute values
  5. Divide by the number of data points

That's the entire process. No squaring, no square roots, no complications.

Common Mistakes

People mess this up in two ways:

Forgetting the absolute value. Without it, deviations above and below the mean cancel out, giving you zero every time. That's useless.

Confusing it with variance. Variance is the average of squared deviations. Average deviation doesn't square anything. Different formulas, different results.