Mean Deviation, Characteristics, Example, Uses

• Based on All Observations

The Mean Deviation is calculated using every single value in the dataset. It finds the absolute deviation of each observation from a central point (mean or median) and incorporates all of them into the final average. This ensures that the measure is comprehensive and reflects the variability of the entire dataset, not just a summary or a few extreme values. It provides a more complete picture of dispersion than a range-based measure.

• Simple to Understand and Calculate

The concept behind Mean Deviation is very intuitive: it is the average of the absolute “distances” of data points from the center. The calculation process is straightforward, involving simple steps of finding deviations, taking their absolute values, and then averaging. This simplicity makes it relatively easy for students and practitioners to compute and interpret without needing advanced mathematical knowledge, unlike the standard deviation which involves squaring.

• Not Amenable to Algebraic Treatment

A major limitation of Mean Deviation is that it is not mathematically tractable for further algebraic operations. Because it uses absolute values, which are not differentiable at zero, it is difficult to manipulate in calculus-based statistics. This property makes it unsuitable for advanced statistical theory, such as optimization in regression analysis or complex probability functions, where the variance and standard deviation are preferred.

• Less Affected by Extreme Values Than Standard Deviation

Since Mean Deviation does not square the deviations, it does not amplify the effect of extreme values (outliers) to the same degree as variance or standard deviation. While an outlier will still increase the MD, its influence is proportional to its distance from the center, not the square of that distance. This makes MD a more robust measure in datasets with significant outliers.

• Uses Absolute Values to Avoid Cancellation

The key to its calculation is taking the absolute value of each deviation from the mean or median. This is done because the sum of the simple deviations (without absolute values) from the mean is always zero. Using absolute values prevents positive and negative deviations from canceling each other out, allowing for a meaningful calculation of the average “spread” of the data.

• Can be Calculated from Mean or Median

Mean Deviation offers flexibility in its choice of central tendency. It can be calculated from either the mean or the median. While the mean is most common, using the median can be advantageous when dealing with skewed distributions, as the median itself is a more robust center than the mean in such cases, leading to a more representative measure of dispersion.

Data set: 12, 15, 18, 22, 25

Number of observations n = 5

1) Mean deviation about the mean

Step 1 — Compute the mean

Add the observations: 12+15+18+22+25 = 92

Divide by n=5 mean xˉ=92÷5=18.4

Step 2 — Compute absolute deviations from the mean

∣12−18.4∣ = 6.4
∣15−18.4∣ = 3.4
∣18−18.4∣ = 0.4
∣22−18.4∣ = 3.6
∣25−18.4∣ = 6.6

Step 3 — Sum the absolute deviations

6.4 + 3.4 + 0.4 + 3.6 + 6.6 = 20.4

Step 4 — Divide by n to get mean deviation

Mean deviation about the mean =20.4 ÷ 5 = 4.08

So Mean Deviation (about mean) = 4.08.

2) Mean deviation about the median

Step 1 — Find the median

Ordered data: 12, 15, 18, 22, 25 → median is the middle value = 18.

Step 2 — Absolute deviations from the median

∣12−18∣ = 6
∣15−18∣ = 3
∣18−18∣ = 0
∣22−18∣ = 4
∣25−18∣ = 7

Step 3 — Sum them

6+3+0+4+7 = 20

Step 4 — Divide by n

Mean deviation about the median = 20 ÷ 5 = 4.0

So Mean Deviation (about median) = 4.00.