Analysing Data

Analysing Data

Spec S4–S6 · AQA GCSE Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • The mean, median and mode are averages; the range and interquartile range measure spread.
  • The median and IQR are better than the mean and range when there are outliers.
  • Frequency density =frequencyclass width= \dfrac{\text{frequency}}{\text{class width}}; the area of a histogram bar is the frequency.
  • Cumulative frequency graphs estimate the median and quartiles; box plots compare distributions.
  • Correlation does not imply causation; interpolating is more reliable than extrapolating.

Averages and spread

Use an average for the typical value and a measure of spread for how varied the data is; the IQR is less affected by outliers.

To summarise and compare data sets, use a measure of central tendency and a measure of spread. These let you describe a whole population without looking at every value.

The interquartile range (Higher) is the upper quartile minus the lower quartile. It uses the middle 50% of the data, so an outlier barely changes it.

0102030405060Class AClass BClassTest mark
  • Median
  • Interquartile range
Illustrative values: two classes with the same median but different spread.

Central tendency

  • Mean
  • Median
  • Mode or modal class

Spread

  • Range
  • Interquartile range (Higher)
  • Consider outliers

Worked example

Find the median, range and interquartile range of: 2, 4, 5, 7, 8, 10, 11, 13, 15.

A data set has one extreme outlier. Which measure of spread is least affected?

Histograms

A histogram has frequency density on the y-axis, so the area of each bar, not its height, shows the frequency (Higher tier).

Histograms show grouped continuous data, especially when class widths are unequal.

The height of each bar is the frequency density. The area of each bar is the frequency for that class.

Never read the frequency from the height alone when classes are unequal.

fd 1.2fd 2.0fd 0.8ABCDA1B1B2C2C3D3
Illustrative histogram: bar widths are 10, 20 and 10 marks (drawn 1, 2 and 1 units wide). Area = frequency, so f = 12, 40 and 8.
  • Frequency densityfrequencyclass width\dfrac{\text{frequency}}{\text{class width}}
  • Frequencyfrequency density ×\times class width

Worked example

A class 10 < x ≤ 30 has frequency 30. Find the height of its histogram bar.

A bar has frequency density 4 and class width 5. What is the frequency?

Cumulative frequency and box plots

A cumulative frequency graph gives the median and quartiles; a box plot shows five numbers and lets you compare distributions (Higher tier).

A cumulative frequency graph plots the running total against the upper class boundary. Read off the median and quartiles.

A box plot shows the minimum, lower quartile, median, upper quartile and maximum, and clearly shows the IQR.

To compare, a higher median means a higher typical value and a wider IQR means more variability.

  1. 1

    Total

    find the total frequency, n

  2. 2

    Lower quartile

    read across at n/4, then down

  3. 3

    Median

    read across at n/2, then down

  4. 4

    Upper quartile

    read across at 3n/4, then down

Reading quartiles from a cumulative frequency graph

Which statement compares two box plots correctly?

Scatter graphs

A scatter graph shows the relationship between two variables as correlation; a line of best fit predicts values.

A scatter graph plots two related variables. Correlation can be positive, negative or none, and weak or strong.

A line of best fit is used to predict. Interpolation (inside the data range) is more reliable than extrapolation (outside it).

Correlation does not mean one variable causes the other.

12345678910246810xypositivenegativenone (flat)
Illustrative lines showing the three types of correlation.

Positive

Points show:
Both increase together
Example:
Height and shoe size

Negative

Points show:
One rises as the other falls
Example:
Age of a car and its value

None

Points show:
No pattern
Example:
Shoe size and test score

A line of best fit is drawn for data with x from 1 to 7. Which prediction is least reliable?

Try an exam question

Here are 7 numbers: 3, 5, 6, 8, 9, 12, 14. Work out the median, the range and the interquartile range.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Find the median time.

1 mark, marked against the mark scheme

Revision notes