Websink Prep UKGCSE MathsTextbook contentsMaths textbooksYear 1Year 2Year 4Year 5Year 6GCSEBrowse allPast papersFind a tutorBecome a tutor
GCSE Maths · EdexcelChapter 24 · Statistics
GCSE Maths · Statistics · Higher · Lesson 24.13

GCSE Maths: Cumulative frequency graphs (Higher)

Part of the GCSE Maths statistics chapter (Higher). Chapter overview · whole chapter on one page · Download worksheet (PDF) · Answers (PDF)

Download worksheet (PDF)Answers (PDF)
24.13

Cumulative frequency graphs

Edexcel 1MA1: S3, S4
Grade 5-8

Cumulative frequency is a running total of the frequencies. It tells us how many values are up to and including the end of each class. A cumulative frequency graph plots each cumulative frequency against the upper boundary of its class and joins the points with a smooth S-shaped curve. It rises from 0 at the start and ends at the total number of values.

0204060800204060Time taken (minutes)Cumulative frequency
A cumulative frequency curve for the times taken by 80 workers: 4 took up to 10 minutes, 14 took up to 20 minutes, and so on.

Reading the graph. To find the median, go to half of the total (n ÷ 2) on the vertical axis, across to the curve and down to the horizontal axis. For the lower quartile use n ÷ 4 and for the upper quartile use 3n ÷ 4. The interquartile range is the upper quartile minus the lower quartile. A percentile works in the same way: the 90th percentile is at 90% of n. These readings are estimates. Notice that on a cumulative frequency graph we use n ÷ 2, n ÷ 4 and 3n ÷ 4, not (n + 1) ÷ 2 and so on.

To find how many values are less than a given value, go up from that value to the curve and across to the cumulative frequency axis. To find how many are more than a value, subtract that reading from the total.

Key ideas

  • Cumulative frequency = running total of the frequencies.
  • Plot (upper class boundary, cumulative frequency), starting with (lowest boundary, 0), and join with a smooth curve.
  • Median at n ÷ 2, lower quartile at n ÷ 4, upper quartile at 3n ÷ 4. IQR = Q3 − Q1.
  • Percentiles: the p-th percentile is at p% of n.
  • To find the number above a value, read the cumulative frequency and subtract it from the total.
  • Readings from a graph are estimates.
Worked example 24.13.1
The table shows the times taken by 80 workers to complete a task.
TimeFrequency
0 < t ≤ 10 min4
10 < t ≤ 20 min10
20 < t ≤ 30 min22
30 < t ≤ 40 min25
40 < t ≤ 50 min13
50 < t ≤ 60 min6
Complete the cumulative frequency column.

Key words. Cumulative frequency is a running total of the frequencies. A cumulative frequency graph plots each running total against the upper end of its class.

Method. Cumulative frequency is a running total: add each frequency to the total of all the earlier ones.

  1. Start the running total with the first frequency. Cumulative frequency means “how many values so far”.
    First class: 4
  2. Add the next frequency to the total so far (10 < t ≤ 20 min). Each cumulative frequency is the previous one plus this class’s frequency.
    4 + 10 = 14
  3. Add the next frequency to the total so far (20 < t ≤ 30 min). Each cumulative frequency is the previous one plus this class’s frequency.
    14 + 22 = 36
  4. Add the next frequency to the total so far (30 < t ≤ 40 min). Each cumulative frequency is the previous one plus this class’s frequency.
    36 + 25 = 61
  5. Add the next frequency to the total so far (40 < t ≤ 50 min). Each cumulative frequency is the previous one plus this class’s frequency.
    61 + 13 = 74
  6. Add the next frequency to the total so far (50 < t ≤ 60 min). Each cumulative frequency is the previous one plus this class’s frequency.
    74 + 6 = 80
  7. Check the last number. The last cumulative frequency must equal the total number of values.
    Total of frequencies = 80 = last cumulative frequency 80

Check. The last running total (80) equals the sum of all the frequencies. ✓

Watch out. The cumulative frequency never goes down, and the last one is the total number of data values.

Worked example 24.13.2
Use the table in the last example to draw a cumulative frequency graph.

Key words. Cumulative frequency is a running total of the frequencies. A cumulative frequency graph plots each running total against the upper end of its class.

Method. Make a cumulative frequency table, then plot each cumulative frequency against the upper end of its class, and join the points with a smooth curve.

  1. Work out the cumulative frequencies. Add each frequency to the total so far.
    TimeFrequencyCumulative frequency
    0 < t ≤ 10 min44
    10 < t ≤ 20 min1014
    20 < t ≤ 30 min2236
    30 < t ≤ 40 min2561
    40 < t ≤ 50 min1374
    50 < t ≤ 60 min680
  2. Write the points to plot. Plot each cumulative frequency at the upper class boundary, because that is the point by which that many values have been counted. Start at (0, 0): nothing is below the first lower boundary.
    (0, 0), (10, 4), (20, 14), (30, 36), (40, 61), (50, 74), (60, 80)
  3. Choose scales and plot the points. The horizontal axis goes from the lowest to the highest class boundary. The vertical axis goes from 0 to the total frequency.
    Horizontal axis: Time taken (minutes). Vertical axis: cumulative frequency up to 80.
  4. Join the points with a smooth curve. The curve is S-shaped and never goes down.
    0204060800204060Time taken (minutes)Cumulative frequency

Check. The curve starts at 0 and ends at the total frequency, 80. ✓

Watch out. Plot at the upper boundary of each class, not the midpoint and not the lower boundary.

Worked example 24.13.3
Use the cumulative frequency graph above to estimate the median, the lower quartile, the upper quartile and the interquartile range.

Key words. Cumulative frequency is a running total of the frequencies. A cumulative frequency graph plots each running total against the upper end of its class.

Method. On a cumulative frequency graph, the median is at height n ÷ 2, the lower quartile at n ÷ 4 and the upper quartile at 3n ÷ 4. Read across to the curve and down to the axis.

  1. Find the total number of values. The last cumulative frequency is the total, n.
    n = 80
  2. Work out the heights to read at. Lower quartile at n ÷ 4, median at n ÷ 2, upper quartile at 3n ÷ 4.
    n ÷ 4 = 20
    n ÷ 2 = 40
    3n ÷ 4 = 60
  3. Draw a horizontal line across from each height to the curve, then straight down to the axis. The value on the horizontal axis is the quartile.
    0204060800204060Time taken (minutes)Cumulative frequency23.531.539.5
  4. Read the three values. Read the scale carefully. The answers are estimates.
    Q1 ≈ 23.5 min
    Median ≈ 31.5 min
    Q3 ≈ 39.5 min
  5. Subtract to find the interquartile range. IQR = Q3 − Q1.
    39.5 − 23.5 = 16 min

Check. Going up from each reading to the curve and across should give the heights again. ✓

Watch out. Read heights from the cumulative frequency axis and answers from the data axis. Do not use n + 1 here.

Worked example 24.13.4
Estimate the number of workers who took more than 45 minutes.

Key words. Cumulative frequency is a running total of the frequencies. A cumulative frequency graph plots each running total against the upper end of its class.

Method. Read the cumulative frequency at the value. This is the number of values up to that value; subtract from the total to get the number above it.

  1. Find 45 minutes on the horizontal axis and go up to the curve. The curve shows how many values are at or below each value.
    0204060800204060Time taken (minutes)Cumulative frequency69
  2. Read across to the cumulative frequency axis. This gives the number of values up to 45.
    Cumulative frequency = 69
  3. Subtract from the total to find how many are more than that. The last cumulative frequency is the total.
    80 − 69 = 11

Check. 69 below plus 11 above makes 80, the total. ✓

Watch out. The reading gives the number below the value. For “more than” you must subtract from the total.

Worked example 24.13.5
Estimate the 90th percentile of the times.

Key words. Cumulative frequency is a running total of the frequencies. A cumulative frequency graph plots each running total against the upper end of its class.

Method. Cumulative frequency graphs are read by going across from the cumulative frequency to the curve, then down to the axis.

  1. Find the position on the cumulative frequency axis. The total cumulative frequency is n = 80. The 90th percentile is 90% of the way through the data: 90 ÷ 100 × 80 = 72.
    Position = 72
  2. Draw a horizontal line across from that height to the curve. We want the data value that has this many values at or below it.
    0204060800204060Time taken (minutes)Cumulative frequency48
  3. Draw a vertical line down to the horizontal axis and read the value. Read the scale carefully. The result is an estimate.
    90th percentile ≈ 48 min

Check. Going up from your answer to the curve and across should give the same height again. ✓

Watch out. Read across from the cumulative frequency axis. The frequency axis is the vertical one, the data axis is the horizontal one.

Watch out

Plot at the upper class boundary (not the midpoint), and start the curve at (lowest boundary, 0). When reading, use n ÷ 2, n ÷ 4 and 3n ÷ 4: do not add 1.

Exercise 24M  Cumulative frequency graphs

Non-calculator
1
FThe table shows the heights of 60 plants.
HeightFrequency
10 < h ≤ 20 cm5
20 < h ≤ 30 cm12
30 < h ≤ 40 cm24
40 < h ≤ 50 cm14
50 < h ≤ 60 cm5
Work out the cumulative frequency for each class.
2
FUse the table in question 1 to draw a cumulative frequency graph.
3
FThe graph shows the times, in minutes, taken by 80 workers to complete a task. Use the graph to estimate (a) the median, (b) the lower quartile, (c) the upper quartile, (d) the interquartile range.
0204060800204060Time taken (minutes)Cumulative frequency
(a)Median
(b)Lower quartile
(c)Upper quartile
(d)Interquartile range
4
PSUse the same graph to estimate (a) the number of workers who took more than 45 minutes, (b) the number of workers who took less than 25 minutes, (c) the percentage of workers who took less than 20 minutes.
0204060800204060Time taken (minutes)Cumulative frequency
(a)More than 45 minutes
(b)Less than 25 minutes
(c)Percentage less than 20 minutes
5
FUse the graph in question 3 to estimate the 90th percentile of the times.
6
FThe table shows the marks of 100 students in a test.
MarkFrequency
0 < m ≤ 204
20 < m ≤ 3010
30 < m ≤ 4018
40 < m ≤ 5024
50 < m ≤ 6022
60 < m ≤ 8016
80 < m ≤ 1006
Draw a cumulative frequency graph.
7
PSThe cumulative frequency graphs show the marks of 40 students in class A (blue) and class B (orange). (a) Estimate the median of each class. (b) Estimate the interquartile range of each class. (c) Compare the two classes.
05101520253035400102030405060708090100MarkCumulative frequency
(a)Medians
(b)Interquartile ranges
(c)Comparison
8
PSThe cumulative frequency graph in question 3 shows the times of 80 workers. The company gives a bonus to the fastest 15% of workers. Estimate the longest time that a worker can take and still get the bonus.
9
PSUse the graph in question 3 to estimate the number of workers who took between 25 and 45 minutes.
10
RExplain why cumulative frequency is plotted at the upper boundary of each class, not at the midpoint.
11
PSA cumulative frequency curve passes through (0, 0), (10, 12), (20, 40), (30, 70) and (40, 80). (a) How many values are in the class 20 < x ≤ 30? (b) What percentage of all the values are 20 or less?
(a)Values in 20 < x ≤ 30
(b)Percentage 20 or less
12
PSThe table is partly completed.
Time, t (minutes)FrequencyCumulative frequency
0 < t ≤ 1066
10 < t ≤ 2014?
20 < t ≤ 30?45
30 < t ≤ 409?
40 < t ≤ 50?60
Complete the table.
13
RA cumulative frequency graph for 120 values has median 38, lower quartile 30 and upper quartile 50. Estimate how many values are between 30 and 50. Explain your answer.

Websink Prep UK is an independent site. It is not affiliated with, endorsed by or sponsored by AQA, Pearson Edexcel, OCR, WJEC, Eduqas, CCEA, SQA, GL Assessment, CEM or the Department for Education. All exam board names and trademarks belong to their owners. Past papers are linked from the exam boards' own websites and are not hosted here.

About Websink Prep UK · Find a tutor · Privacy · Terms · Cookie settings