Chapter 16
Cumulative frequency
This chapter will show you how to
✔ construct cumulative frequency tables
✔ find quartiles and interquartile ranges
✔ draw box plots
✔ compare and interpret distributions
You will need to know
l how to construct frequency tables
16.1 Cumulative frequency
You can construct a cumulative frequency table from a grouped
frequency table by calculating the running total of the frequency
up to the end of each class interval.
You can use the table to plot a cumulative frequency diagram.
You can draw cumulative frequency diagrams for both discrete and
continuous data sets.
EXAMPLE 1
The following grouped frequency distribution shows the time taken
Class intervals (such as 0 , t < 5)
for students to solve a puzzle. Using this data, draw a cumulative were discussed in Chapter 15.
frequency diagram.
Time taken (seconds) Frequency
0 , t < 5 2
5 , t < 10 9
10 , t < 15 9
15 , t < 20 8
20 , t < 25 3
25 , t < 30 1
continued .
288 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 288 28/8/09 12:37:16
Cumulative frequency
Time taken (seconds) Cumulative frequency Begin by creating a cumulative
0 , t < 5 2 frequency table. This cumulative
frequency column is a running total
5 , t < 10 2 1 9 5 11 of the frequency so far.
10 , t < 15 2 1 9 1 9 5 20 The figures in bold are the values of
15 , t < 20 2 1 9 1 9 1 8 5 28 the cumulative frequency. The class
intervals (t < ) are also changed to
20 , t < 25 2 1 9 1 9 1 8 1 3 5 31 show the values ‘up to and
25 , t < 30 2 1 9 1 9 1 8 1 3 1 1 5 32 including’ i.e. the upper class
boundary value.
30 You can now use this data to draw a
cumulative frequency diagram by
Cumulative plotting the cumulative frequency
20
frequency against the upper class boundary of
each class interval.
10 Cumulative frequency is always
plotted on the vertical axis. The
maximum value on the y-axis (the
0 10 20 30 cumulative frequency axis) should
be the total number of values in the
Time (s) data set, in this case 32.
Draw a smooth curve through the points (to give a cumulative frequency curve),
as shown.
The shape of the cumulative frequency diagram reflects the characteristics of the
data and how this data is spread or distributed within the range. This characteristic
S shape (called an ogive) appears in nearly all cumulative frequency diagrams.
Cumulative frequency graphs
l Choose a suitable scale for each axis.
l Always plot the cumulative frequencies on the vertical axis
(the y-axis).
l Plot the points using the upper class boundary e.g. (5, 2), (10, 11),
(15, 20) and so on.
l Draw a smooth curve through the points (points can also be joined
with straight lines) and include the origin.
l Check that you have plotted your points correctly.
l Check that your graph is an S-shape.
Handling data 289
M16_CME_SB_IGCSE_6867_U16.indd 289 28/8/09 12:37:17
Estimating values using a cumulative frequency graph
Once you have drawn your cumulative frequency diagram you can use
it to estimate values for the data.
Cumulative
frequency
Number of values
less than or
equal to x (�x)
0
Values of x
Value x
For example, given a value of x you can estimate the number of values
less than or greater than x by drawing a straight line that meets the
cumulative frequency curve and then drawing a corresponding line to
meet the other axis.
EXAMPLE 2
Use the cumulative frequency curve from Example 1 to estimate Always draw lines on your graph to
(a) how many students solved the puzzle show how you obtained your
answers. Do not rub them out.
(i) within 12 seconds (ii) in more than 18 seconds
(b) the time by which 28 students had solved the puzzle. (a) Draw a vertical line for x 5 12
to intersect the cumulative
frequency curve. Now draw a line to
30
meet the cumulative frequency axis.
The value obtained is 15.
Cumulative
20
frequency
(b) The line at x 5 18 seconds gives
10 a value of 26. This means that
32 2 26 5 6 students took longer
than 18 seconds to solve the puzzle.
0 10 20 30
(c) Draw a line from the cumulative
Time (s) frequency value of 28 to intersect
the curve. Now draw a vertical line
to meet the x-axis at the point
x 5 20. This is the time by which 28
(a) (i) 15 students solved the puzzle within 12 seconds.
students solved the puzzle.
(ii) 32 2 26 5 6
6 students took longer than 18 seconds.
(b) x 5 20
28 students solved the puzzle within 20 seconds.
290 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 290 28/8/09 12:37:18
Cumulative frequency
EXERCISE 16A
1 (a) Copy and complete the cumulative frequency table showing
the distribution of test marks for a group of 32 students.
Mark (%) Number of Mark (%) Cumulative
students frequency
1–10 1 <10 1
11–20 2 <20 3
21–30 4
31–40 7
41–50 5
51–60 8
61–70 2
71–80 2
Your x-axis (mark) should run
81–90 1 from 0 to 100, and your y-axis from
0 to 35.
91–100 0
(b) Plot the results in a cumulative frequency graph.
2 The table below shows the frequency distribution of test marks for
120 students.
Mark (%) Number of students
1–10 1
11–20 6
21–30 8
31–40 15
41–50 17
51–60 24
61–70 22
71–80 15
81–90 9
91–100 3
Construct a cumulative frequency table (taking the first class
interval to be <10 and the last interval to be <100). Draw the
corresponding cumulative frequency graph for this distribution.
Handling data 291
M16_CME_SB_IGCSE_6867_U16.indd 291 28/8/09 12:37:18
3 The results for the long jump at a school sports day are shown Distance x (m) Frequency
in the table.
1.70 , x < 1.80 2
Draw a cumulative frequency diagram for this distribution.
1.80 , x < 1.90 6
Estimate how many students jumped over 2.35 m.
1.90 , x < 2.00 9
2.00 , x < 2.10 7
2.10 , x < 2.20 15
2.20 , x < 2.30 8
2.30 , x < 2.40 8
2.40 , x < 2.50 2
4 The temperature in °C recorded over a 66-day period is shown Temperature t (°C) Number of
in the table. days
Draw a cumulative frequency diagram and estimate the number 0 , t < 3 1
of days that the temperature was above 18°C.
3 , t < 7 7
7 , t < 11 18
11 , t < 15 20
15 , t < 19 17
19 , t < 23 2
23 , t < 27 1
2.40 , x < 2.50 2
5 The cumulative frequency curve for the amount of time spent on a
homework task is shown below.
Construct a cumulative frequency table and use this to estimate
the number of students who spent more than 1 hour on the task.
Time spent on homework
50
45
40
35
Cumulative 30
frequency 25
20
15
10
5
0
0 20 40 60 80 100 120
Time (minutes)
292 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 292 28/8/09 12:37:19
Cumulative frequency
16.2 The median, quartiles and
inter-quartile range
You can obtain important statistical measures from a cumulative
frequency graph. These are useful when you want to compare two or
more data sets.
l The median is the middle value of the distribution. n is the total frequency.
This is at 12 (n 1 1) on the cumulative frequency axis.
1
l The lower quartile (LQ) is at 4 (n11) on the cumulative
frequency axis.
3
The answers for the median, the
l The upper quartile (UQ) is at 4 (n11) on the cumulative lower quartile and the upper
frequency axis. quartile are read off on the
horizontal axis.
l The interquartile range (IQR) gives an improved measure of the
The first step in this process is to
spread of the data and is given by locate the appropriate points on the
Inter-quartile range 5 upper quartile 2 lower quartile cumulative frequency axis.
The IQR is a measure of spread of the middle 50% of the data, so it
excludes extreme values.
EXAMPLE 3
Use the cumulative frequency graph of Example 1 to find an
n = 32
estimate for The middle value is
1 1
(a) the median (b) the lower quartile 2 (32 1 1) 5 16 2th value
(c) the upper quartile (d) the interquartile range.
30 Draw a horizontal line from this
point to meet the curve. The median
value is where this vertical line
Cumulative
20 meets the x-axis, as shown.
frequency 16.5
10
0 5 Median 30
Time (s)
continued .
Handling data 293
M16_CME_SB_IGCSE_6867_U16.indd 293 28/8/09 12:37:20
(a) The median 5 13 seconds.
30
(b) To find the lower quartile
24.75 (LQ), first locate a point at
1
Cumulative 20 4 (32 1 1) = 8.25 on the frequency
frequency axis and draw a horizontal line to
intersect the curve. Draw a vertical
10
8.25 line to meet the x-axis. This gives
the lower quartile.
0 5 IQR 30 (c) Find the upper quartile (UQ) in a
similar way, but this time use a
Time (s)
point at 34 (32 1 1) = 24.75 on the
LQ UQ cumulative frequency axis.
(b) Total 5 32
1
4 of (32 1 1) 5 8.25
(d) Remember IQR 5 UQ 2 LQ
LQ 5 9.5 seconds
3
(c) 4 of (32 1 1) 5 24.75
UQ = 24.5 seconds
(d) IQR 5 17.5 2 9.5 5 8 seconds
The cumulative graph can be divided up into other fractions, not just
quarters.
Tenths are quite common and these are called deciles.
Percentiles divide the cumulative frequency into hundredths.
70
The 70th percentile will be at 100 (n 1 1) on the cumulative
frequency (cf) axis.
EXAMPLE 4
29.7
Cumulative
20
frequency
10
6.6
0 5 Median 30
Time (s)
Use the cumulative frequency curve to find an estimate for these.
(a) the 20th percentile (b) the 90th percentile
continued .
294 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 294 28/8/09 12:37:20
Cumulative frequency
20
(a) The 20th percentile will be at 100 (32 1 1) 5 6.6
Draw a horizontal line at 6.6 to the curve and then
a vertical line down to the time axis.
20th percentile 5 9 seconds
90
(b) The 90th percentile will be at 100 (32 1 1) 5 29.7
Repeat the process of (a) at 29.7
90th percentile 5 23 seconds
EXERCISE 16B
1 The cumulative frequency graph for the distances travelled to
school by 50 students is shown below.
50
40
Cumulative 30
frequency
20
10
0
0 1 2 3 4 5 6 7 8 9 10
Distance travelled (km)
From the graph, estimate
(a) the median distance (b) the lower quartile distance
(c) the upper quartile distance (d) the inter-quartile range.
2 A survey was conducted to find the
amount of time students spent on 100
homework each week. The results are 90
shown in the cumulative frequency graph. 80
From the graph, estimate 70
(a) the median time Cumulative 60
frequency 50
(b) the lower quartile time
40
(c) the upper quartile time
30
(d) the IQR.
20
Give your answers in hours and minutes. 10
(e) Estimate what percentage of students spent 0
5 hours or less on homework each week. 0 2 4 6 8 10 12 14 16
Time (hours)
Handling data 295
M16_CME_SB_IGCSE_6867_U16.indd 295 28/8/09 12:37:21
3 The cumulative frequency diagram shows
the marks gained in a geography exam. 160
140
Use the graph to find mark estimates for
120
(a) the median
Cumulative 100
(b) LQ frequency 80
(c) UQ 60
(d) IQR 40
(e) the 70th percentile. 20
0
0 10 20 30 40 50 60 70 80 90
Mark (%)
4 A tyre company carried out a survey
to find how far cars travel before they 100
need new tyres. The results for 100 cars 90
is shown in the cumulative frequency 80
diagram. 70
Use the graph to estimate Cumulative 60
(a) the median distance travelled frequency 50
(b) the lower quartile distance 40
travelled 30
(c) the upper quartile distance 20
travelled 10
(d) the inter-quartile range. 0
10 15 20 25 30 35 40
(e) what percentage of cars travelled Distance travelled (� 1000 km)
a distance of more than 30 000
kilometres before the tyres needed to
be changed?
5 During a busy day at a doctor’s surgery the amount of waiting
time for patients was recorded as shown in the table.
The shortest waiting time was 1 12 mins. and the longest time
was 9 12 mins.
(a) Construct a cumulative Waiting time, t Frequency
frequency table. (minutes) (f )
(b) Draw a cumulative 0 , t < 1 0
frequency curve.
1 , t < 2 4
(c) Give an estimate of the
median waiting time. 2 , t < 3 23
(d) Find the 60th percentile. 3 , t < 4 43
4 , t < 5 58
5 , t < 6 37
6 , t < 7 11
7 , t < 8 3
8 , t < 9 0
9 , t < 10 1
296 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 296 28/8/09 12:37:22
Cumulative frequency
6 The local sports centre conducted a survey on the age distribution of
its 800 members and the results are shown in the table.
The youngest person was 6 years old and the oldest was 78.
Age, a (years) Frequency (f )
0 , a < 10 41
10 , a < 20 138
20 , a < 30 168
30 , a < 40 192
40 , a < 50 126
50 , a < 60 85
60 , a < 70 39
70 , a < 80 11
(a) Construct a cumulative frequency table for this data.
(b) Draw a cumulative frequency graph.
(c) Use the curve to estimate the median age of the leisure centre
members.
(d) Find an estimate of the lower and upper quartiles.
(e) Find the 95th percentile.
Handling data 297
M16_CME_SB_IGCSE_6867_U16.indd 297 28/8/09 12:37:22
EXAMINATION QUESTIONS
1 500 eggs were sorted by mass into five different sizes.
Mass (m grams) Frequency
Small 35 < m , 40 20
Medium 40 < m , 50 60
Standard 50 < m , 60 200
Large 60 < m , 75 180
Extra large 75 < m , 80 40
y
8
6
W T
5
4
V
3
1
U
–5 –4 –3 –2 –1 0 1 2 3 4 5 6 7 8 9 10 x
This cumulative frequency curve has been drawn using the information in the table above.
(a) Explain why the point (60, 280) is on the curve. [2]
(b) Estimate the median mass of the eggs. [1]
(c) Estimate the inter-quartile range of the masses of the eggs. [2]
(CIE Paper 4, Jun 2000)
298 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 298 28/8/09 12:37:24
Cumulative frequency
2
y
5
B
4
2
A
2
0 1 2 3 4 5 6 x
The cumulative frequency diagram shows the heights of plants measured in an experiment.
From the diagram, estimate
(a) (i) the lower quartile, [1]
(ii) the inter-quartile range. [1]
(b) the number of plants with a height greater than 25cm. [1]
(CIE Paper 2, Nov 2000)
3 In a survey, 200 shoppers were asked how much they had just spent at a supermarket.
The results are shown in the table.
Amount
0 , x < 20 20 , x < 40 40 , x < 60 60 , x < 80 80 , x < 100 100 , x < 140
($x)
Number of
10 32 48 54 36 20
shoppers
(a) (i) Make a cumulative frequency table for these 200 shoppers. [2]
(ii) Using a scale of 2 cm to represent $20 on the horizontal axis and 2 cm to
represent 20 shoppers on the vertical axis, draw a cumulative frequency diagram
for this data. [4]
(b) Use your cumulative frequency diagram to find
(i) the median amount, [1]
(ii) the upper quartile, [1]
(iii) the inter-quartile range, [1]
(iv) how many shoppers spent at least $75. [2]
(CIE Paper 4, Jun 2003)
Handling data 299
M16_CME_SB_IGCSE_6867_U16.indd 299 28/8/09 12:37:26
4 The depth, d centimetres, of a river was recorded each day during a period of one year
(365 days). The results are shown by the cumulative frequency curve.
y
8
–2 –1 0 1 2 3 4 5 6 7 8 9 10 x
(a) Use the cumulative frequency curve to find
(i) the median depth, [1]
(ii) the inter-quartile range, [2]
(iii) the depth at the 40th percentile, [2]
(iv) the number of days when the depth of the river was at least 25 cm. [2]
(b)
d 0 , d < 10 10 , d < 20 20 , d < 30 30 , d < 40 40 , d < 50 50 , d < 60 60 , d < 70
Number
17 41 62 98 85 p q
of days
Show that p 5 47 and q 5 15. [2]
(CIE Paper 4, Jun 2004)
300 Handling data
M16_CME_SB_IGCSE_6867_U16.indd 300 28/8/09 12:37:28
