Mean Absolute Deviation (MAD) Calculator

What Is Mean Absolute Deviation?

Mean Absolute Deviation (MAD) measures how spread out values in a data set are from their mean. It answers the question: “On average, how far does each data point deviate from the mean?” Unlike variance or standard deviation, MAD uses absolute values rather than squared differences, which makes it more intuitive and easier to interpret in the original units of your data.

The Formula

Step 1: Calculate the mean: x̄ = (x₁ + x₂ + … + xₙ) / n

Step 2: Find the deviation of each value from the mean: |xᵢ - x̄|

Step 3: Average those absolute deviations: MAD = Σ|xᵢ - x̄| / n

MAD vs Standard Deviation

Standard deviation squares the deviations before averaging, which amplifies the effect of large outliers. MAD treats all deviations equally. This makes MAD more robust: a single extreme outlier will inflate standard deviation dramatically but only moderately affect MAD.

For normally distributed data, MAD ≈ 0.7979 × standard deviation. That constant is √(2/π), and it describes the true population values rather than any particular sample, so real data lands near 0.798 rather than on it.

The calculator reports the ratio, but resist reading too much into it on a small data set. It is a weak detector at low counts: the five values {1, 1, 1, 1, 1000} give a ratio of exactly 0.800, indistinguishable from well-behaved normal data, despite one value being a thousand times the others. The ratio needs a few dozen points before it says anything reliable. To spot an outlier, compare the largest deviation against how far the other values typically sit from the mean, which the calculator does directly and which works at any sample size.

Note that the standard deviation shown here divides by n, the population form. Some calculators divide by n − 1 to estimate a population from a sample, which gives a slightly larger figure. Neither is wrong; they answer different questions. Use n when your numbers are the entire group you care about, and n − 1 when they are a sample standing in for something bigger.

Median Absolute Deviation

A related measure is the Median Absolute Deviation (also abbreviated MAD): MAD_median = median(|xᵢ - median(x)|). It is even more robust than the mean-based version, because it shrugs off several outliers rather than just one, which is why robust statistics and anomaly detection reach for it.

Real-World Applications

• Quality control: A factory producing parts with MAD of 0.02mm has tighter tolerances than one with MAD of 0.15mm.
• Finance: Portfolio daily returns with a low MAD are more predictable than high-MAD portfolios.
• Meteorology: Temperature forecasts are evaluated by their MAD from actual temperatures (called Mean Absolute Error in forecasting).
• Education: MAD of test scores shows how spread out student performance is around the class average.

Worked Example

Data: {4, 7, 13, 2, 1}. Mean = 27/5 = 5.4. Deviations: |4-5.4|=1.4, |7-5.4|=1.6, |13-5.4|=7.6, |2-5.4|=3.4, |1-5.4|=4.4. MAD = (1.4+1.6+7.6+3.4+4.4)/5 = 18.4/5 = 3.68. This means values deviate from the mean by 3.68 units on average.