Measures of location and spread

Measures of location and spread

Spec 2.3 · Edexcel A-Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • Mean xˉ=∑xn\bar x=\dfrac{\sum x}{n} (or ∑fx∑f\dfrac{\sum fx}{\sum f}); median is the middle value; mode is the most frequent.
  • For grouped data use class midpoints, so results are estimates.
  • Variance =∑x2n−xˉ2=Sxxn=\dfrac{\sum x^2}{n}-\bar x^2=\dfrac{S_{xx}}{n}; standard deviation is its square root.
  • Coding y=x−aby=\dfrac{x-a}{b}: xˉ=byˉ+a\bar x=b\bar y+a, sx=b sys_x=b\,s_y.
  • Compare data sets with one comment on location and one on spread, in context.

Measures of central tendency

The mean uses every value, the median is the middle value and the mode is the most frequent.

The mean is xˉ=∑xn\bar x=\frac{\sum x}{n}, the median is the middle value of the ordered data and the mode is the most frequent value (or modal class). For grouped data use class midpoints, so the mean is an estimate.

00.511.522348ValueFrequency
Frequencies for the data 2, 3, 3, 4, 8: mean = 4, median = 3, mode = 3.

Mean

Uses every value:
Yes
Affected by outliers:
Yes
Works for non-numerical data:
No

Median

Uses every value:
No
Affected by outliers:
No
Works for non-numerical data:
No

Mode

Uses every value:
No
Affected by outliers:
No
Works for non-numerical data:
Yes

Worked example

Find the mean, median and mode of 2, 4, 4, 6, 9.

Which measure is least affected by one extreme value?

Measures of variation

The range, interquartile range and interpercentile ranges all measure spread, in different ways.

The range is largest minus smallest. A percentile has a given percentage of values below it. The IQR is Q3−Q1Q_3-Q_1, and an interpercentile range such as 10th to 90th covers the middle 80%80\%.

For grouped data estimate a percentile by linear interpolation: find the class containing position p100n\frac{p}{100}n. Percentile-based ranges ignore extreme values; the range does not.

02468101214MinimumQ1MedianQ3MaximumSummary valueValue
Five-number summary of 1, 2, 4, 5, 7, 8, 10, 13: range = 13 − 1 = 12 and IQR = 9 − 3 = 6.
  • Rangelargest − smallest
  • IQRQ3−Q1Q_3-Q_1
  • 10th to 90thmiddle 80%80\%

What percentage of the data does the 10th to 90th interpercentile range cover?

Variance and standard deviation

Use Sxx=∑x2−(∑x)2nS_{xx}=\sum x^2-\frac{(\sum x)^2}{n}: variance is Sxxn\frac{S_{xx}}{n} and the standard deviation is its square root.

The standard deviation measures the typical distance of values from the mean. The variance is ∑x2n−xˉ2\frac{\sum x^2}{n}-\bar x^2. The version with Sxxn−1\frac{S_{xx}}{n-1} is also accepted.

For frequency tables use ∑fx2\sum fx^2 and ∑f\sum f in place of ∑x2\sum x^2 and nn.

−8−6−4−224680.10.20.30.40.5xystandard deviation s
Normal curves with the same mean: a larger standard deviation gives a wider, flatter curve.
  • SxxS_{xx}∑x2−(∑x)2n\sum x^2-\dfrac{(\sum x)^2}{n}
  • Standard deviationSxxn\sqrt{\dfrac{S_{xx}}{n}}

Worked example

Given ∑x=240\sum x=240, ∑x2=6100\sum x^2=6100 and n=10n=10, find the standard deviation.

The variance of a data set is 49. What is the standard deviation?

Grouped data

Use class midpoints: xˉ≈∑fx∑f\bar x\approx\frac{\sum fx}{\sum f}, and ∑fx2\sum fx^2 for the standard deviation.

Both results are estimates because the real values within classes are unknown. Say ‘estimate’ in your answer.

024680–1010–2020–3030–40ClassFrequency
Illustrative grouped data: midpoints 5, 15, 25, 35 give Σfx = 360 and Σf = 20, so the estimated mean is 18.

Midpoint x

  • 5
  • 15
  • 25
  • 40
  • Total

Frequency f

  • 4
  • 12
  • 20
  • 14
  • 50

fx

  • 20
  • 180
  • 500
  • 560
  • 1260

fx²

  • 100
  • 2700
  • 12500
  • 22400
  • 37700

Worked example

Use the table to estimate the mean and standard deviation.

Why is a mean from grouped data only an estimate?

Coding

With y=x−aby=\frac{x-a}{b}, the mean is multiplied by bb and shifted by aa; the spread is only multiplied by bb.

Coding replaces xx by y=x−aby=\frac{x-a}{b}, so x=by+ax=by+a. Shifting does not change spread.

The variance is multiplied by b2b^2.

102030405060708090100−55101520xyy = (x − a)/b
Coding y = (x − a)/b is a straight line: a shifts it along and b changes its slope (illustrative a and b).
  • Meanxˉ=byˉ+a\bar x=b\bar y+a
  • Standard deviationsx=b sys_x=b\,s_y
  • Varianceb2×b^2\times variance of yy

Worked example

With y=x−40010y=\frac{x-400}{10}, the coded data have mean 33 and variance 7.257.25. Find the mean and variance of xx.

With y=x−505y=\frac{x-50}{5}, yˉ=4\bar y=4 and the standard deviation of yy is 2. What is the standard deviation of xx?

Large data set and interpretation

Make one comment on location and one on spread, each in context.

Pearson’s large data set gives weather data. Questions may use its terminology but need no knowledge of the actual data.

Comment on a measure of location (mean or median) and a measure of spread (SD, IQR), each in context.

  1. 1

    Location

    compare the mean or median, in context

  2. 2

    Spread

    compare the standard deviation or IQR, in context

  3. 3

    Conclusion

    say what it means for the situation

Commenting on data in context.

Location

  • Mean or median
  • A higher mean means warmer on average

Spread

  • SD or IQR
  • A smaller SD means more consistent temperatures

Town A has a lower standard deviation of daily temperature than town B. What does this tell you?

Try an exam question

For 10 values, ∑x=150\sum x=150 and ∑x2=2400\sum x^2=2400. Find the mean and standard deviation.

[3 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

What is the mean delivery time?

1 mark, marked against the mark scheme

Revision notes