- Practice
- About 17 min
- 5 practice questions
Measures of dispersion: range, variance and standard deviation
See why an average is not enough, find the range and coefficient of range, and calculate the variance and standard deviation of ungrouped data by the mean method and the direct method.
If something looks wrong, please tell us.
By the end of this lesson you can
- Explain what dispersion means and why two data sets with the same mean can differ.
- Find the range L − S and the coefficient of range (L − S)/(L + S).
- Define deviations from the mean, variance σ² and standard deviation σ.
- Calculate σ for ungrouped data by the mean method and by the direct method.
In earlier classes you summarised data with a measure of central tendency: the mean, median or mode. But one number in the middle does not tell the whole story. Two batsmen can have the same average score while one is steady and the other swings between big scores and ducks. A measure of dispersion tells us how spread out, or scattered, the values are around the centre. The Tamil Nadu textbook lists range, mean deviation, quartile deviation, standard deviation, variance and coefficient of variation; this chapter studies range, variance, standard deviation and the coefficient of variation.
Same mean, different spread
Consider two made-up sets of marks. Set P: 18, 20, 22. Set Q: 5, 20, 35. Both have mean 20. But in P every value is within 2 of the mean, while in Q the values are 15 away. We need a number that captures this difference.
Range and coefficient of range
The simplest measure is the range: the difference between the largest value L and the smallest value S. Dividing by L + S gives a pure number, the coefficient of range, which lets us compare data in different units.
Range R = L − S. Coefficient of range = (L − S)/(L + S).
Worked example 1: range
For the made-up data 34, 58, 21, 47, 66, 39: L = 66 and S = 21. Range = 66 − 21 = 45. Coefficient of range = 45/(66 + 21) = 45/87 ≈ 0.52. The range is quick, but it uses only two values and ignores everything in between, so it can mislead.
Deviations, variance and standard deviation
For observations x₁, x₂, ..., xₙ with mean x̄, the deviations from the mean are x₁ − x̄, x₂ − x̄, ..., xₙ − x̄. Their sum is always 0 (positives and negatives cancel), so we square them first. The variance σ² (sigma squared) is the mean of the squared deviations. The standard deviation σ is the positive square root of the variance; it has the same units as the data.
Variance σ² = Σ(xᵢ − x̄)²/n. Standard deviation σ = √[Σ(xᵢ − x̄)²/n] = √[Σxᵢ²/n − (Σxᵢ/n)²].
The second form, the direct method, comes from expanding the square, and is handy when the raw values are given and the mean is awkward. Variance can never be negative, since it is an average of squares.
Standard deviation by the mean method
- Find the mean x̄ = Σx/n.
- Deviations Subtract the mean from each value: d = x − x̄.
- Square Work out d² for every value.
- Average Variance σ² = Σd²/n.
- Square root σ = √(σ²), in the units of the data.
Worked example 2: mean method
Find the standard deviation of 2, 4, 6, 8, 10. Mean = 30/5 = 6. Deviations: −4, −2, 0, 2, 4. Squares: 16, 4, 0, 4, 16, with sum 40. Variance = 40/5 = 8. σ = √8 ≈ 2.83.
Worked example 3: direct method
A shop's (made-up) daily sales of a book over five days are 12, 15, 9, 18, 6. Σx = 60 and Σx² = 144 + 225 + 81 + 324 + 36 = 810. σ² = 810/5 − (60/5)² = 162 − 144 = 18, so σ = √18 ≈ 4.24. Check by the mean method: mean 12, deviations 0, 3, −3, 6, −6, squares 0, 9, 9, 36, 36, sum 90, and 90/5 = 18.
Worked example 4: back to sets P and Q
Set P (18, 20, 22): squared deviations 4, 0, 4, so σ² = 8/3 and σ ≈ 1.63. Set Q (5, 20, 35): squared deviations 225, 0, 225, so σ² = 450/3 = 150 and σ ≈ 12.25. The standard deviation confirms what we saw: Q is far more spread out.
Key takeaways: dispersion measures spread. Range = L − S is easy but crude. Variance is the mean of the squared deviations from the mean, and standard deviation is its positive square root, in the same units as the data. Use the mean method when the mean is a whole number and the direct method otherwise.
Check your understanding
-
1. What is the range of 12, 7, 25, 18, 9?
- 13
- 25
- 18
- 16
Show answer
Answer: C. Largest L = 25, smallest S = 7. Range = 25 − 7 = 18.
-
2. What is the coefficient of range of the data 15, 20, 25?
- 0.25
- 10
- 0.5
- 0.4
Show answer
Answer: A. L = 25 and S = 15. Coefficient = (25 − 15)/(25 + 15) = 10/40 = 0.25.
-
3. What is the variance of 1, 3, 5, 7, 9?
- 2.83
- 40
- 10
- 8
Show answer
Answer: D. Mean = 25/5 = 5. Squared deviations: 16, 4, 0, 4, 16, sum 40. Variance = 40/5 = 8. (2.83 is the standard deviation, √8.)
-
4. The variance of a data set is 0.64. What is its standard deviation?
- 0.4096
- 0.8
- 0.32
- 0.08
Show answer
Answer: B. Standard deviation = √variance = √0.64 = 0.8, because 0.8 × 0.8 = 0.64.
-
5. For 5 observations, Σx = 30 and Σx² = 200. What is the standard deviation?
- 2
- 4
- 6
- 1.6
Show answer
Answer: A. σ² = Σx²/n − (Σx/n)² = 200/5 − (30/5)² = 40 − 36 = 4. So σ = √4 = 2. (4 is the variance.)
Sign in to keep track of the lessons you finish.