9.7 · Extended onlyPaper 2Paper 4

Histograms

In a minute

A shows grouped data as touching bars, where the area of each bar shows how many values are in its class. The height of a bar is its : the frequency divided by the class width. This stops a wide class looking bigger than it really is.

You'll be able to

  • Work out the frequency density of each class in a .
  • Draw a histogram for classes of different widths.
  • Read frequencies from a histogram, including an estimate for part of a class.
  • Find bar heights or a missing frequency when heights are given in centimetres or a bar is missing.
Before you start2 quick questions on what this builds on
Check yourself
A class in a grouped frequency table is 35<h≤5035 < h \le 50. What is its class width?
Check yourself
A grouped frequency table has the classes 10<t≤2010 < t \le 20 and 20<t≤3020 < t \le 30. Is a time of exactly 2020 in the class 20<t≤3020 < t \le 30?

Revise first: 9.3 Averages and measures of spread

Frequency density and drawing a histogram

A can have of different widths. Then a bar chart of the frequencies misleads you.

Say 1212 people wait between 00 and 55 minutes, and 1212 wait between 55 and 3030 minutes. The first 1212 are packed into 55 minutes. The second 1212 are spread over 2525 minutes. A fair picture makes the first bar taller.

Learn this
frequency density=frequencyclass width\text{frequency density} = \frac{\text{frequency}}{\text{class width}}

Why does dividing by the width work? Area == width ×\times height == class width ×frequencyclass width=\times \frac{\text{frequency}}{\text{class width}} = frequency.

Worked exampleThe table shows how long some students spent on homework one evening. Time (t minutes)0<t≤1010<t≤1515<t≤4545<t≤90Frequency7143336\begin{array}{c|cccc} \text{Time } (t \text{ minutes}) & 0 < t \le 10 & 10 < t \le 15 & 15 < t \le 45 & 45 < t \le 90 \\ \hline \text{Frequency} & 7 & 14 & 33 & 36 \end{array} Work out the frequency density of each class. Which class has the tallest bar on a histogram?

Try it first, then reveal one step at a time.

  1. Find each class width, upper boundary minus lower: 10−0=1010 - 0 = 10, 15−10=515 - 10 = 5, 45−15=3045 - 15 = 30 and 90−45=4590 - 45 = 45.

1. Work out each class width and frequency density. Two extra columns in the table keep this tidy.

2. Draw a continuous scale for the data along the horizontal axis. Label the vertical axis "Frequency density".

3. Draw each bar from its lower class boundary to its upper one, as tall as its frequency density. The bars touch, with no gaps.

Worked exampleThe table shows the heights of 3636 seedlings. Height (h cm)0<h≤44<h≤66<h≤77<h≤10Frequency61389\begin{array}{c|cccc} \text{Height } (h \text{ cm}) & 0 < h \le 4 & 4 < h \le 6 & 6 < h \le 7 & 7 < h \le 10 \\ \hline \text{Frequency} & 6 & 13 & 8 & 9 \end{array} Draw a histogram to show this information.

Try it first, then reveal one step at a time.

  1. Class widths: 4−0=44 - 0 = 4, 6−4=26 - 4 = 2, 7−6=17 - 6 = 1 and 10−7=310 - 7 = 3.
Histogram of seedling heights. Height in cm across, from 0 to 10. Frequency density up, from 0 to 8, with each small square 0.5 and a label every 1. Four touching bars: 0 to 4 cm at height 1.5, 4 to 6 cm at height 6.5, 6 to 7 cm at height 8, and 7 to 10 cm at height 3. 0 1 2 3 4 5 6 7 8 9 10 Height (h cm) 0 1 2 3 4 5 6 7 8 Frequency density
The 6<h≤76 < h \le 7 bar is the tallest, though its class holds only 88 seedlings.
Check yourself
A class 15<m≤4015 < m \le 40 has frequency 3535. Work out its frequency density.
Check yourself
A table has classes 0<x≤60 < x \le 6, 6<x≤216 < x \le 21 and 21<x≤2821 < x \le 28, with frequencies 1515, 3333 and 1414. How tall is the tallest bar on its histogram?

Reading frequencies from a histogram

Reading a histogram reverses the drawing. The area of a bar is the frequency of its class.

Learn this
frequency=frequency density×class width\text{frequency} = \text{frequency density} \times \text{class width}

Before you read any heights, work out what one small square on each axis is worth.

Histogram of apple masses. Mass in grams across, from 100 to 200. Frequency density up, from 0 to 3.5, with each small square 0.1 and a label every 0.5. Four touching bars: 100 to 120 g at height 0.8, 120 to 130 g at height 3.2, 130 to 150 g at height 2.1, and 150 to 200 g at height 0.4. 100 110 120 130 140 150 160 170 180 190 200 Mass (m grams) 0 0.5 1 1.5 2 2.5 3 3.5 Frequency density
The masses of the apples picked from one tree.
Worked exampleThe histogram shows the masses, mm grams, of the apples picked from one tree. Work out how many apples were picked.

Try it first, then reveal one step at a time.

  1. Each small square up is 0.10.1. Read the heights: 0.80.8, 3.23.2, 2.12.1 and 0.40.4.

For part of a class, assume its values are spread evenly across the class. Then use the width of just that part. The answer is an estimate.

Worked exampleUse the same histogram. Calculate an estimate of the number of apples with a mass of more than 140140 g.

Try it first, then reveal one step at a time.

  1. Only part of the class 130<m≤150130 < m \le 150 is above 140140. Assume its apples are spread evenly across the class.
Check yourself
A histogram bar goes from 2525 to 4545 on the horizontal axis. Its height is 1.81.8. How many values are in its class?
Check yourself
A histogram bar for 10<x≤3010 < x \le 30 has height 2.52.5. Estimate how many values lie between 1010 and 1616.

Heights in centimetres and missing bars

Some questions give bar heights in centimetres, not on a frequency density scale. The heights are still in proportion to the frequency densities.

1. Use the bar you know. Centimetres for 11 unit of frequency density == its height ÷\div its frequency density.

2. Work out the frequency density of each other class.

3. Multiply each frequency density by the centimetres for 11 unit.

Worked exampleThe table shows how long some people waited for a bus. Time (t minutes)0<t≤44<t≤1010<t≤1212<t≤16Frequency1433139\begin{array}{c|cccc} \text{Time } (t \text{ minutes}) & 0 < t \le 4 & 4 < t \le 10 & 10 < t \le 12 & 12 < t \le 16 \\ \hline \text{Frequency} & 14 & 33 & 13 & 9 \end{array} On a histogram, the bar for 0<t≤40 < t \le 4 is 4.24.2 cm tall. Calculate the heights, in cm, of the other three bars.

Try it first, then reveal one step at a time.

  1. Known bar: width 44, so frequency density =14÷4=3.5= 14 \div 4 = 3.5. It is drawn 4.24.2 cm tall.

If a bar is missing but you know the total frequency, find the other frequencies first. What is left belongs to the missing class.

Worked exampleA histogram shows the distances, dd km, that 6060 people cycled in a week. It has bars for 0<d≤60 < d \le 6 with height 1.51.5, 6<d≤146 < d \le 14 with height 2.252.25, and 17<d≤3317 < d \le 33 with height 0.750.75. The bar for 14<d≤1714 < d \le 17 is missing. Work out its height.

Try it first, then reveal one step at a time.

  1. Frequencies of the drawn bars: 1.5×6=91.5 \times 6 = 9, 2.25×8=182.25 \times 8 = 18 and 0.75×16=120.75 \times 16 = 12.
Check yourself
On a histogram, the class 0<x≤150 < x \le 15 has frequency 99 and its bar is 4.84.8 cm tall. The class 15<x≤2015 < x \le 20 has frequency 77. How tall is its bar, in cm?
Check yourself
On a histogram, a bar 44 units wide is 99 cm tall and stands for 2424 values. Another bar is 77 units wide and 66 cm tall. How many values does it stand for?

Common mistakes

  • Drawing each bar with the frequency as its height. Divide each frequency by its class width first, so wide classes don't look too big.
  • Reading a bar's height as its frequency. The frequency is the area: height ×\times class width.
  • Scaling bar heights in centimetres by the frequencies. Scale by the frequency densities, because the class widths differ.

Exam tips

  • When you complete a table or draw a histogram, write each frequency density, such as 13÷2=6.513 \div 2 = 6.5. Correct frequency densities can still earn marks if a bar is drawn wrongly.
  • "Calculate an estimate" for part of a class needs the width of that part times the bar's height, such as 2.1×102.1 \times 10. Show that product.
  • When heights are in centimetres, first work out the centimetres for 1 unit of frequency density from the bar you know. Show that step, as it is usually worth a method mark.
  • After you read frequencies from a histogram, check that they add to any total the question gives.

Key takeaways

Learn thisfrequency density=frequencyclass width\text{frequency density} = \frac{\text{frequency}}{\text{class width}}
Learn thisfrequency=frequency density×class width\text{frequency} = \text{frequency density} \times \text{class width}
Learn thisA histogram's bars touch, the horizontal axis is a continuous scale and the vertical axis is labelled "Frequency density". The area of a bar shows its frequency.
Learn thisPart of a class: assume the values are spread evenly, so use height ×\times width of that part. The answer is an estimate.
Learn thisHeights in cm: cm for 11 unit == known height ÷\div its frequency density. Multiply each frequency density by it.

Practise this

38 past-paper questions on 9.7 in the 0580 Extended bank, each with a mark scheme and a worked explanation.

Practise 9.7 →