Measures of central tendency and spread

Measures of central tendency and spread

Spec L3 · AQA A-Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • Mean xˉ=∑xn\bar x=\frac{\sum x}{n} (or ∑fx∑f\frac{\sum fx}{\sum f}); median is the middle value; mode is the most common.
  • The median is not affected by extreme values, so it suits skewed data.
  • IQR =Q3−Q1=Q_3-Q_1 is unaffected by extremes; range uses only two values.
  • Variance =∑x2n−xˉ2=\frac{\sum x^2}{n}-\bar x^2; standard deviation is its square root.
  • Comment on both average and spread, in context. Never average two standard deviations.

Averages

The mean, median and mode each describe the centre. Choose the one that suits the data.

The mean uses every value, so extreme values pull it. The median is the middle value of the ordered data (the mean of the middle two if nn is even) and is not affected by extremes. The mode is the most common value (the modal class for grouped data) and is the only average for categorical data.

From a frequency table, the mean is ∑fx∑f\frac{\sum fx}{\sum f}.

051015MeanMedianModeAverageValue
Averages of 12, 15, 15, 18, 22, 26: mean 18, median 16.5, mode 15.

Mean

  • ∑xn\frac{\sum x}{n}
  • Uses every value
  • Pulled by extremes

Median

  • Middle of ordered data
  • Not affected by extremes
  • Suits skewed data

Mode

  • Most common value
  • Modal class if grouped
  • Only average for categories

Worked example

Find the mean, median and mode of 12, 15, 15, 18, 22, 26.

Which average is least affected by one very large value?

Measures of spread

Range, interquartile range and standard deviation each measure spread. Match the spread to the average you quote.

The range is largest minus smallest, so one extreme value can distort it.

The interquartile range (IQR) is Q3−Q1Q_3-Q_1, the spread of the middle half, and is not affected by extremes.

The standard deviation uses every value and measures typical distance from the mean. It is in the same units as the data; larger means more spread.

10121416182022242628min 12Q1 15median 16.5Q3 22max 26
The same data: range 26 − 12 = 14 (whole line); interquartile range 22 − 15 = 7 (shaded, the middle half).

Mean

Matching spread:
Standard deviation
Use when:
Data roughly symmetric

Median

Matching spread:
Interquartile range
Use when:
Data skewed or has extremes

Q1=12Q_1=12 and Q3=20Q_3=20. What is the interquartile range?

Variance and standard deviation

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

For a frequency table use ∑fx2∑f−xˉ2\frac{\sum fx^2}{\sum f}-\bar x^2. The second form is quicker when you are given ∑x\sum x and ∑x2\sum x^2.

This course uses divisor nn (calculators may label it σ\sigma or σn\sigma_n). A version with divisor n−1n-1 gives a slightly larger value, so check which one a question requires.

510152025300.020.040.060.080.1xymean 1616 − 4.916 + 4.9illustrative distribution
Illustrative bell curve with mean 16 and standard deviation √24 = 4.90: a typical value lies about 4.9 from the mean.
  • Variance∑x2n−xˉ2\dfrac{\sum x^2}{n}-\bar x^2
  • Standard deviationvariance\sqrt{\text{variance}}

Worked example

n=25n=25, ∑x=400\sum x=400 and ∑x2=7000\sum x^2=7000. Find the standard deviation.

n=10n=10, ∑x=50\sum x=50 and ∑x2=300\sum x^2=300. What is the variance?

Summary statistics

Given nn, ∑x\sum x and ∑x2\sum x^2 you can find the mean and standard deviation without the raw data, and correct or combine sets.

To correct a mistake, remove the wrong value from the sums and add the right one, then recalculate.

To combine two sets, add the nn, ∑x\sum x and ∑x2\sum x^2 values, then use the formulae on the totals. Never average two standard deviations.

Use the statistics mode on your calculator, but know the formulae: you may need to work from ∑x\sum x and ∑x2\sum x^2.

  1. 1

    Adjust the sums

    subtract the wrong value from ∑x\sum x, add the right one

  2. 2

    Adjust the squares

    do the same with the squares in ∑x2\sum x^2

  3. 3

    Recalculate

    find the mean, then the variance, then the standard deviation

Correcting a recorded value

Worked example

For n=25n=25, ∑x=400\sum x=400, ∑x2=7000\sum x^2=7000, one value recorded as 40 should have been 20. Find the corrected mean and standard deviation.

Two sets have standard deviations 3 and 5. Is the combined standard deviation 4?

Comparing data sets

Make two comments in context: one about the average (centre) and one about the spread.

Adding a value equal to the mean leaves the mean unchanged and decreases the standard deviation. It adds no squared deviation while increasing nn.

Standard deviation measures spread, not size.

99499699810001002100410061008101010120.050.10.150.20.250.30.350.4xylabel 1000 gMachine AMachine B
Illustrative bag masses: A (mean 1004 g, s.d. 3.2 g) is wider and further from 1000 g; B (1001 g, s.d. 1.1 g) is narrower and closer.

Machine A

Mean (g):
1004
Standard deviation (g):
3.2

Machine B

Mean (g):
1001
Standard deviation (g):
1.1

B is closer to the 1000 g label, and its bags are more consistent.

Machine A has the larger standard deviation. What does that tell you?

Try an exam question

A sample of 2020 values has ∑x=300\sum x=300 and ∑x2=5000\sum x^2=5000. Calculate the mean and the standard deviation.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

What is the median time?

1 mark, marked against the mark scheme

Revision notes