Measures of dispersion

Measures of dispersion

Spec 2.3 · Edexcel International A Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • Range == largest −- smallest; IQR =Q3−Q1=Q_3-Q_1, the spread of the middle 50%.
  • Variance σ2=∑x2n−xˉ2\sigma^2=\dfrac{\sum x^2}{n}-\bar x^2 and standard deviation σ=variance\sigma=\sqrt{\text{variance}}; the divisor is nn.
  • For y=ax+by=ax+b, σy=∣a∣σx\sigma_y=|a|\sigma_x: adding a constant does not change the spread.
  • A smaller standard deviation or IQR means more consistent data.
  • Use the median and IQR for skewed data, and the mean and standard deviation otherwise.

Range and quartiles

The range uses only the extremes; the interquartile range is Q3−Q1Q_3-Q_1, the spread of the middle half, so extremes do not affect it.

A measure of dispersion describes how widely data vary. The range is the largest value minus the smallest, which depends on two values only.

The quartiles split ordered data into four equal parts: Q1Q_1 is the n4\frac n4th value, Q2Q_2 the n2\frac n2th and Q3Q_3 the 3n4\frac{3n}{4}th.

  • If the position is not a whole number, round up.
  • If it is a whole number, take the mean of that value and the next.

The interquartile range is not affected by extreme values.

  • Rangelargest −- smallest
  • Interquartile rangeIQR=Q3−Q1IQR=Q_3-Q_1

Worked example

Find Q1Q_1, Q3Q_3 and the IQR of 2, 4, 4, 4, 5, 5, 7, 9.

For n=11n=11 values, which value is Q1Q_1?

Percentiles and interpolation

Find the class containing the required position, then interpolate linearly within it.

An interpercentile range is the difference between two percentiles, such as P90−P10P_{90}-P_{10}, which ignores the most extreme 10% at each end.

For grouped continuous data, estimate any percentile by linear interpolation assuming even spread in each class. Locate the class containing the required position (for example n4\frac n4 for Q1Q_1).

Here LL is the lower boundary, FF the cumulative frequency before the class, ff its frequency and ww its width.

  • Value from interpolationL+position−Ff×wL+\dfrac{\text{position}-F}{f}\times w

Worked example

80 waiting times have cumulative frequencies 6, 30, 58, 74, 80 at 10, 20, 30, 40, 60 minutes. Estimate the IQR.

For grouped data with n=200n=200, which position is used for Q3Q_3?

Variance and standard deviation

Variance is the mean of the squares minus the square of the mean; the standard deviation is its square root.

The variance measures the mean squared distance from the mean: σ2=∑(x−xˉ)2n=∑x2n−xˉ2\sigma^2=\frac{\sum(x-\bar x)^2}{n}=\frac{\sum x^2}{n}-\bar x^2. Use the second form, which is quicker.

The standard deviation is its square root.

For a frequency table or grouped data (using midpoints), σ2=∑fx2∑f−xˉ2\sigma^2=\frac{\sum fx^2}{\sum f}-\bar x^2.

Variance is never negative. Standard deviation has the same units as the data; variance has the units squared. The divisor is nn.

  • Varianceσ2=∑x2n−xˉ2\sigma^2=\dfrac{\sum x^2}{n}-\bar x^2
  • Standard deviationσ=σ2\sigma=\sqrt{\sigma^2}

Worked example

n=20n=20, ∑x=140\sum x=140, ∑x2=1160\sum x^2=1160. Find the standard deviation.

A set has ∑x2n=30\frac{\sum x^2}{n}=30 and xˉ=5\bar x=5. What is the standard deviation?

Coding and combining data

For y=ax+by=ax+b the standard deviation is multiplied by ∣a∣|a|; adding a constant does not change it.

If y=ax+by=ax+b, then yˉ=axˉ+b\bar y=a\bar x+b but the standard deviation becomes σy=∣a∣σx\sigma_y=|a|\sigma_x and the variance becomes a2σx2a^2\sigma_x^2. Adding or subtracting a constant does not change the spread.

Example: σx=2\sigma_x=2 and y=3x+2y=3x+2 give σy=6\sigma_y=6.

To add or remove a value, update nn, ∑x\sum x and ∑x2\sum x^2 and recalculate. Adding a value equal to the mean leaves the mean unchanged but reduces the standard deviation.

  • Meanyˉ=axˉ+b\bar y=a\bar x+b
  • Standard deviationσy=∣a∣σx\sigma_y=|a|\sigma_x

Worked example

The sample above has n=20n=20, xˉ=7\bar x=7, variance 9. Add the value 7. Find the new variance.

Data have standard deviation 4. Each value is multiplied by 2, then 5 is added. What is the new standard deviation?

Interpreting and comparing spread

Compare using a measure of location and a measure of spread, in context. A smaller standard deviation means more consistent.

Compare data sets using a measure of location and a measure of spread, in context.

A smaller standard deviation (or IQR) means the data are more consistent.

Use the median and IQR when data are skewed or contain extreme values, and the mean and standard deviation otherwise. A value far from the mean increases the standard deviation a lot, because deviations are squared.

Example: machines with target 500 ml. A has mean 501.5 and σ=1.89\sigma=1.89; B has mean 500.6 and σ=0.8\sigma=0.8. B is better: its mean is closer to the target and it is more consistent.

01002003004005001.890.8ABMachineml
  • Mean
  • Standard deviation
Machines A and B: mean and standard deviation of fill volume (ml), target 500 ml.

Machine C has mean 500.2 and σ=3\sigma=3. Compared with B, which is more consistent?

Try an exam question

A sample of 1010 values of xx has ∑x=80\sum x=80 and ∑x2=700\sum x^2=700. (a) Find the mean and the standard deviation of xx. (b) The values are coded as y=3x+1y=3x+1. Find the standard deviation of yy.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Find the variance of the eight values.

1 mark, marked against the mark scheme

Revision notes