Sign in

← All revision topics

English variety

Mean deviation, variance and standard deviation; coefficient of variation

Machine translation

Around age 14 · Free

Learn

  1. Gus records the packages he delivers over 5 days: 3, 5, 6, 7 and 9. The mean is 6. The deviation of a data value is how far it is from the mean: data value − mean. The 9 has deviation 9 − 6 = 3 and the 3 has deviation 3 − 6 = −3. If the data value is below the mean, its deviation is negative.

    024681012Gus
    The deviations are the distances from the mean, 6
  2. The deviations of all the data values always add up to 0, because the positive ones and the negative ones cancel out. That is why you cannot average them as they are. There are two ways to fix this: remove the sign or square them.

    Data value35679Sum
    Deviation−3−10130
    Without sign310138
    Square9101920
    Deviations, values without sign and squares
  3. The mean deviation is the mean of the deviations without sign: (3 + 1 + 0 + 1 + 3) ÷ 5 = 1.6. The variance is the mean of the squares of the deviations: (9 + 1 + 0 + 1 + 9) ÷ 5 = 4.

    The standard deviation is the square root of the variance: the square root of 4 is 2 packages. It is written σ and in Mexico it is also called desviación estándar. It has the same units as the data; the variance does not (packages squared). If the variance is not a square number, the root is estimated or rounded: the square root of 5 is 2.2, roughly.

    The further apart the data values are, the greater the standard deviation. It is 0 only if all the data values are equal. If you add the same number to all the data values, it does not change; if you multiply them by k, it is multiplied by k.

    With a frequency table, each square is multiplied by its frequency before dividing by the total number of data values.

    The standard deviation depends on the units and on the size of the data values. To compare the dispersion of data with very different means, or with different units, you use the coefficient of variation: standard deviation ÷ mean, as a percentage. A smaller coefficient means the data are more homogeneous.

     MeanStandard deviationCoefficient of variation
    Small2 kg0.5 kg25%
    Large20 kg2 kg10%
    The large ones have a greater standard deviation, but a smaller coefficient of variation

Worked examples

Example 1

Mara records the kilometres she cycles over 8 days: 2, 4, 4, 4, 5, 5, 7 and 9. Find the mean deviation, the variance and the standard deviation.

Data value24445579Sum
Deviation−3−1−1−100240
Without sign3111002412
Square91110041632
  1. The mean is (2 + 4 + 4 + 4 + 5 + 5 + 7 + 9) ÷ 8 = 40 ÷ 8 = 5.
  2. The deviations are −3, −1, −1, −1, 0, 0, 2 and 4. Without sign they add up to 3 + 1 + 1 + 1 + 0 + 0 + 2 + 4 = 12.
  3. Mean deviation: 12 ÷ 8 = 1.5. Squares: 9, 1, 1, 1, 0, 0, 4 and 16, which add up to 32. Variance: 32 ÷ 8 = 4.
  4. Standard deviation: the square root of 4 is 2.

Mean deviation 1.5 km, variance 4 km² and standard deviation 2 km.

Watch out! Here many people add up the deviations with their sign and say the mean deviation is 0, or give the variance, 4, as if it were the standard deviation.

Common mistake

Example 2

Ellie sells bread rolls over 5 days. Week A: 18, 20, 22, 20 and 20 rolls. Week B: 5, 15, 20, 25 and 35. Both have a mean of 20. What is the standard deviation of each one?

412202836AB
  1. Week A: deviations −2, 0, 2, 0 and 0; squares 4, 0, 4, 0 and 0, which add up to 8. Variance: 8 ÷ 5 = 1.6.
  2. The square root of 1.6 is not exact: it is between 1 and 2, and it is 1.3 to one decimal place.
  3. Week B: deviations −15, −5, 0, 5 and 15; squares 225, 25, 0, 25 and 225, which add up to 500. Variance: 500 ÷ 5 = 100. Standard deviation: 10.

Week A has a standard deviation of 1.3 rolls and week B, 10 rolls: week B is much more spread out.

Watch out! Here many people say they are the same, because both means are 20.

Common mistake

Example 3

Mara weighs small packages (mean 2 kg, standard deviation 0.5 kg) and large ones (mean 20 kg, standard deviation 2 kg). Which are more homogeneous?

 MeanStandard deviation
Small2 kg0.5 kg
Large20 kg2 kg
  1. The large ones have the greater standard deviation, but also the greater mean: you cannot compare them using only 2 kg and 0.5 kg.
  2. Coefficient of variation of the small ones: 0.5 ÷ 2 = 0.25, that is, 25%.
  3. Coefficient of variation of the large ones: 2 ÷ 20 = 0.10, that is, 10%.

The large ones are more homogeneous: their coefficient of variation, 10%, is smaller than the 25% of the small ones.

Watch out! Here many people compare 2 kg and 0.5 kg and say the small ones are more homogeneous.

Common mistake

A challenge to think about

Time yourself six times going from your house to the corner, and ask a friend to do the same. Find the mean and the standard deviation of each person's times. Who takes less time on average? Who is more regular? Is it always the same person?

Practise

Rounds of 10 questions. Choose a level:

Your grade in this topic:No grade yetDo a round at the Basic level to get a grade.

Next topic: Cumulative frequencies and cumulative frequency graphs →