Skip to content

Median Absolute Deviation Calculator

Paste a list of numbers. The calculator finds the median, the absolute deviation of each value from the median, and the median of those deviations. For the default data the median is 13.5 and the median absolute deviation is 2.00.

Updated 2026-11-19 Editorial estimate Calccc Editorial Team Free · No sign-up

Separate values with commas, spaces, or semicolons. At least two numbers are required.

Scenarios A/BSave two input sets and compare the results. Stored only in this browser.
AEmpty
BEmpty

Your inputs stay in this browser. No account, no uploads.

Professional worksheet

Detailed schedule & assumptions

Inputs used, method, sensitivity, and any year-by-year table. The summary above stays compact; the audit lives here.

Professional worksheet

Inputs, scope, and sensitivity

Inputs used

Method and scope

Result audit

MetricValue

What this calculator does

The median absolute deviation, or MAD, is a measure of spread that is resistant to outliers. The standard deviation squares the deviations from the mean, so a single extreme value can dominate the result. The MAD instead takes the median of the absolute deviations from the median, which means half the data lies within one MAD of the median and half lies outside. That makes it useful for skewed data, contaminated samples, and any situation where a few unusual values should not set the scale. This calculator reports the MAD, the scaled MAD, the mean absolute deviation, the median, and the range. The scaled MAD multiplies by 1.4826 so that, for normally distributed data, it estimates the same spread as the standard deviation. The result is a descriptive statistic; it does not test a hypothesis or prove that a value is an outlier.

How to use it

  1. Enter at least two numbers, separated by commas, spaces, or semicolons.
  2. Check the median and the count before reading the spread.
  3. Read the MAD as the typical distance from the median.
  4. Use the scaled MAD when comparing with a standard deviation.
  5. Compare the mean absolute deviation to see how the two averages differ.
  6. Remove or keep outliers deliberately rather than assuming the calculator identifies them.

How it works

MAD = median(|x1 - median(x)|, |x2 - median(x)|, ..., |xn - median(x)|). The scaled MAD = 1.4826 x MAD. The mean absolute deviation uses the mean of the same absolute deviations instead of the median. The default data gives a median of 13.5 and a MAD of 2.00.

Example calculations

Example 1: A small symmetric data set

The default values have a median of 13.5 and a MAD of 2.00. The scaled MAD is 2.97, which is close to what a standard deviation would report for this nearly symmetric set. The mean absolute deviation is 2.13, slightly different because it uses the mean rather than the median.

Example 2: A data set with one extreme value

For 1, 2, 3, 4, 100, the median is 3.0 and the MAD is 1.00. The extreme value barely moves the result, while a standard deviation would be pulled upward by the 100. That resistance is the reason MAD is used in robust statistics. The mean absolute deviation is 20.20, larger than the MAD because the mean is pulled toward the outlier.

Example 3: A data set with no spread

For 5, 5, 5, 5, the median is 5.0 and the MAD is 0.00. Every value is identical, so there is no deviation from the median. A zero MAD is a valid result and tells you the data have no variation; it does not mean the calculation failed.

What the result means

The MAD is in the same units as the data, which makes it easy to explain: in the default case a typical value is about 2.00 away from the median. The scaled MAD of 2.97 is the version to compare with a standard deviation under normality. If the scaled MAD is much smaller than the standard deviation, the data likely contain outliers or a heavy tail. If the two are similar, the data are closer to symmetric. The calculator reports both the median and the mean absolute deviation so you can see how sensitive the summary is to the choice of center.

Common mistakes

  • Computing deviations from the mean and then taking the median; MAD uses deviations from the median.
  • Forgetting to take absolute values before the second median.
  • Comparing an unscaled MAD directly with a standard deviation without the 1.4826 factor.
  • Assuming a large MAD proves an outlier; it only describes spread.
  • Using MAD on categorical data, where a median may not be meaningful.
  • Dropping values before calculating without documenting the exclusion.

Assumptions and limits

  • The data are numeric and at least two values are provided.
  • The median is the appropriate measure of center for the data.
  • The absolute deviations are calculated from the same median.
  • The scaled MAD uses the normal-consistency constant 1.4826.
  • The data are treated as a sample or a full set; the calculator does not apply a sample correction.
  • The result is descriptive, not a significance test.

Frequently asked questions

What is the difference between MAD and standard deviation?

Both measure spread, but the standard deviation uses squared deviations from the mean and is sensitive to outliers. The MAD uses absolute deviations from the median and is resistant to outliers. The scaled MAD can be compared with a standard deviation for normally distributed data.

Why multiply by 1.4826?

For a normal distribution, the MAD is about 0.6745 standard deviations. Multiplying by 1.4826 makes the scaled MAD a consistent estimator of the standard deviation under normality. Without the scaling, the MAD is smaller than the standard deviation by that factor.

Can the MAD be zero?

Yes. If more than half the values are identical, the absolute deviations from the median can have a median of zero. In the flat example the MAD is 0.00. A zero MAD means there is no variation around the median in the majority of the data.

Does the calculator identify outliers?

No. It describes spread. A common outlier rule uses the median plus or minus a multiple of the MAD, but the choice of multiplier and the decision to remove a value are judgement calls. The calculator gives you the pieces, not the decision.

Should I use MAD or IQR?

Both are robust. The IQR is based on quartiles and is often used in box plots; the MAD is based on the median and is convenient for scale estimation. The choice depends on the convention in your field and on what you want to communicate.

Can I use this for a population instead of a sample?

The MAD is a descriptive statistic and does not require a sample correction. If you need an unbiased estimator of the population standard deviation, use the scaled MAD with the appropriate consistency constant for your distribution, or use the sample standard deviation.

Sources

This is an educational statistics tool, not a formal data analysis or a substitute for statistical software. Robust methods and outlier decisions depend on the study design and should be documented.

2026 update

How to use the professional output

This page is not just a formula box. The result is designed to be checked, compared, and taken away.

  • Professional worksheet. The result includes the inputs used, the method and scope, and a one-variable sensitivity check for finance calculators.
  • A/B scenarios. Save two input sets in the browser and compare the primary, secondary, and breakdown figures side by side.
  • Print / PDF and CSV. Print a branded worksheet, or export the year-by-year table when the calculator produces one.

The formulas and assumptions above still govern the number. The worksheet makes the estimate auditable; it does not make it a lender quote, tax return, or personalized advice.

Calccc Editorial Team · Updated 2026-11-19

Prepared by Calccc Editorial Team as an educational estimate. This page is not a licensed professional review and is not financial, medical, legal, or tax advice.

Editorial estimate prepared by the Calccc Editorial Team; not a licensed professional review.