• Foundation
  • About 16 min
  • 5 practice questions

Dispersion, range and mean deviation for ungrouped data

Explain why an average alone is not enough, find the range, and calculate the mean deviation about the mean and about the median for a list of observations.

If something looks wrong, please tell us.

Two cricketers can have exactly the same average and still be very different players: one scores steadily, the other swings between ducks and centuries. A measure of central tendency (mean, median, mode) tells us where data are centred; a measure of dispersion tells us how widely they are scattered. This chapter studies range, mean deviation, variance and standard deviation.

Same average, different spread

Here are made-up scores of two players in six matches. Player P: 10, 85, 40, 0, 95, 70. Player Q: 45, 55, 50, 48, 52, 50. Each total is 300, so each mean is 300 ÷ 6 = 50. Yet P's scores are spread from 0 to 95, while Q's stay between 45 and 55. A single number describing this spread would let us compare consistency.

Range

Range = maximum value − minimum value.

For P the range is 95 − 0 = 95; for Q it is 55 − 45 = 10. The range is quick, but it uses only the two extreme values and says nothing about how the rest of the data lie around the centre. Better measures look at the deviation of every observation from a central value.

Why we take absolute deviations

The deviation of an observation x from a number a is x − a. Some deviations are positive and some negative, so they cancel. In fact the deviations from the mean always add up to zero: Σ(xᵢ − x̄) = Σxᵢ − nx̄ = nx̄ − nx̄ = 0. So the plain average of deviations from the mean is useless as a measure of spread. Distance ignores direction, so we use absolute values |x − a|.

Mean deviation about a: M.D.(a) = Σ|xᵢ − a| / n. Usually a is the mean x̄ or the median M.

Finding the mean deviation of ungrouped data

  1. Find the centre a Mean x̄ or median M
  2. Deviations Work out xᵢ − a for each observation
  3. Absolute values Drop the minus signs: |xᵢ − a|
  4. Average them M.D.(a) = Σ|xᵢ − a| / n

Worked examples

Worked example 1: about the mean

Find the mean deviation about the mean of 4, 7, 8, 9, 10, 12, 13, 17. n = 8 and Σx = 80, so x̄ = 10. The absolute deviations are 6, 3, 2, 1, 0, 2, 3, 7, whose sum is 24. M.D.(x̄) = 24 ÷ 8 = 3.

Worked example 2: about the median

Find the mean deviation about the median of 11, 4, 15, 6, 9, 20, 8. Arrange in order: 4, 6, 8, 9, 11, 15, 20. With n = 7 (odd), the median is the ((7 + 1)/2)th = 4th observation, M = 9.

Absolute deviations from the median 9
xᵢ4689111520
|xᵢ − 9|53102611

Σ|xᵢ − M| = 5 + 3 + 1 + 0 + 2 + 6 + 11 = 28. M.D.(M) = 28 ÷ 7 = 4.

Worked example 3: comparing the two players

For P (mean 50), the absolute deviations are 40, 35, 10, 50, 45, 20, with sum 200, so M.D. = 200/6 ≈ 33.3 runs. For Q (mean 50) they are 5, 5, 0, 2, 2, 0, with sum 14, so M.D. = 14/6 ≈ 2.3 runs. On average P's scores are about 33 runs from the mean and Q's only about 2: Q is far more consistent.

Reminders on mean and median

  • Mean x̄ = Σxᵢ / n.
  • Median: arrange the data in order. If n is odd, take the ((n + 1)/2)th value; if n is even, take the mean of the (n/2)th and (n/2 + 1)th values.
  • Mean deviation can be taken about any central value, but the mean and the median are the usual choices.

Key takeaways

  • Dispersion measures how scattered data are; it complements the average.
  • Range = max − min, quick but crude.
  • Deviations from the mean sum to zero, so use absolute deviations.
  • M.D.(a) = Σ|xᵢ − a| / n.

Check your understanding

  1. 1. What is the range of 23, 8, 41, 15, 36, 12?

    1. 41
    2. 29
    3. 33
    4. 28
    Show answer

    Answer: C. Range = maximum − minimum = 41 − 8 = 33.

  2. 2. The sum of the deviations of 3, 5, 8, 12 from their mean is:

    1. 0
    2. 7
    3. 28
    4. −7
    Show answer

    Answer: A. The mean is 28/4 = 7, and the deviations are −4, −2, 1, 5, which add to 0. Deviations from the mean always sum to zero.

  3. 3. Find the mean deviation about the mean of 2, 6, 7, 9, 11.

    1. 2
    2. 2.4
    3. 3
    4. 7
    Show answer

    Answer: B. Mean = 35/5 = 7. Absolute deviations: 5, 1, 0, 2, 4, total 12. M.D. = 12/5 = 2.4.

  4. 4. Find the mean deviation about the median of 5, 1, 9, 3, 7, 13, 6.

    1. 2.5
    2. 3.2
    3. 6
    4. 20/7 ≈ 2.86
    Show answer

    Answer: D. In order: 1, 3, 5, 6, 7, 9, 13, so the median is 6. Absolute deviations: 5, 3, 1, 0, 1, 3, 7, total 20. M.D. = 20/7 ≈ 2.86.

  5. 5. Assertion (A): The mean of the deviations from the mean cannot be used as a measure of dispersion. Reason (R): The deviations from the mean always add up to zero.

    1. Both A and R are true, and R is the correct explanation of A
    2. Both A and R are true, but R is not the correct explanation of A
    3. A is true, but R is false
    4. A is false, but R is true
    Show answer

    Answer: A. Both are true. Since Σ(xᵢ − x̄) = 0 for every data set, the mean of these deviations is always 0, whatever the spread. R explains A, and is why absolute deviations are used.

Sign in to keep track of the lessons you finish.

Next lesson → Mean deviation for grouped data

Privacy · Terms · Refunds · Delivery