Secondary 3 G2 Mathematics Tutor Bukit Timah | Cumulative Frequency, Quartiles and Box Plots. Did you know? The last cumulative frequency is the total number of observations, while a box plot summarises the middle half using quartiles. Both can describe the same dataset from different angles.
Cumulative frequency, quartiles and box-and-whisker plots can be intimidating because they ask students to move between tables, curves, ordered data and concise summaries. For Secondary 3 G2 Mathematics learners in Bukit Timah, it helps to understand what each representation tells us before learning the drawing routine.
The official 2027 SEC G2 Mathematics K210 syllabus contains these skills under Statistics and Probability S1. Some are underlined for the specific Paper 2 Section B choice, so this article explains the exam context accurately while keeping the learning steps accessible.
Quick answer: what the diagrams mean
Cumulative frequency is a running total. To estimate a median or quartile from a cumulative curve, use the appropriate fraction of the total count and read across to the curve. A box plot is built from a five-number summary; its box extends from Q1 to Q3, and the distance between those values is the interquartile range.
Choose your Statistics question
- Why cumulative frequency answers a different question
- Start by checking the data total
- Build the cumulative frequency table step by step
- How to sketch a cumulative frequency diagram
- Cumulative frequency never means ‘frequency in that band’
- Median: the halfway position in an ordered set
- Quartiles divide an ordered distribution into four parts
- A complete raw-data quartile example
- Interquartile range measures the middle-half spread
- How to construct a box-and-whisker plot
- Why a box plot is not a histogram
- Compare two groups fairly using median and IQR
- Quartiles and percentiles: related but not identical words
- Reading percentiles from a cumulative frequency curve
- Why a cumulative curve should rise but need not rise evenly
- Grouped data creates estimation, not exact individual values
- From a cumulative table back to class frequency
- Common errors and the first-line correction
- The 2027 SEC K210 Paper 2 Section B choice
- What a tutor should assess before teaching graph construction
- A two-week cumulative frequency and box-plot practice plan
- Try three short questions without the model beside you
- How Bukit Timah parents can choose useful Mathematics support
Why cumulative frequency answers a different question
Did you know? An ordinary frequency table tells you how many observations fall in a group. A cumulative frequency table tells you how many have been counted up to that group. Those two counts can be made from the same data, yet answer different questions.
Suppose a fictional quiz has three scores from 0–9, five from 10–19, eight from 20–29 and four from 30–39. The frequencies are 3, 5, 8 and 4. The cumulative frequencies are 3, 8, 16 and 20, because each running total includes all preceding groups.
A useful Secondary 3 G2 Mathematics tutor asks students to explain the word ‘cumulative’ in a sentence: ‘Sixteen pupils scored in the first three score bands combined.’ That is a much stronger sign of understanding than correctly copying 16 into a table.
Start by checking the data total
For the four groups above, total frequency is 3 + 5 + 8 + 4 = 20. The final cumulative frequency must also be 20. If a student’s running total ends at 24, something has been added incorrectly or counted more than once.
Each cumulative frequency should be at least as large as the previous one because counts accumulate. A declining cumulative frequency line is a clear warning unless the chart is describing a different kind of running measure.
Encourage pupils to write a total row in a grouped table. An honest ten-second total check often catches an error that would otherwise affect the graph, median estimate and quartile readings together.
Build the cumulative frequency table step by step
Using the quiz example, begin with 0–9: frequency 3, cumulative 3. Add the 10–19 frequency of 5 to get 8. Add the 20–29 frequency of 8 to get 16. Finally add the 30–39 frequency of 4 to get 20.
The second cumulative entry means eight pupils scored below 20 in this grouping. The third says sixteen scored below 30. Neither value represents the frequency of its own class alone.
Be careful about the definition of the class boundaries, particularly for continuous measurements. Where scores are whole-number data, boundaries such as 9.5, 19.5, 29.5 and 39.5 can be used for a continuous graph representation. Follow the question’s specified grouping.
How to sketch a cumulative frequency diagram
Place the measured score on the horizontal axis and cumulative frequency on the vertical axis. For whole-number score bands 0–9, 10–19, 20–29 and 30–39, one conventional grouped representation plots cumulative points at the upper class boundaries: (9.5, 3), (19.5, 8), (29.5, 16) and (39.5, 20).
Include a sensible starting point at the lower boundary with cumulative frequency zero when appropriate to the data, here (−0.5, 0). Join the plotted points with a smooth non-decreasing curve or suitable line segments according to the school question.
The resulting plot is an approximation to the distribution within the classes. A student should not pretend that the graph reveals the precise individual marks of all twenty pupils; grouping has hidden those details.
Cumulative frequency never means ‘frequency in that band’
If a graph reaches 16 at the upper boundary 29.5, the interpretation is ‘about sixteen observations up to this score’, not ‘sixteen pupils scored between 20 and 29’. The actual frequency of that particular class is eight.
To recover a class frequency from a cumulative table, subtract consecutive totals. Here the 20–29 group’s frequency is 16 − 8 = 8. This is a powerful checking step and helps connect the table to the graph.
A tutor can deliberately place a class-frequency question beside a cumulative-count question using the same figures. Learners who answer both with sixteen have revealed a misunderstanding of the y-axis.
Median: the halfway position in an ordered set
The median divides ordered observations into two halves. For twenty observations, the central values are the tenth and eleventh. In a grouped cumulative frequency diagram, a typical graphical estimate is read at a cumulative position close to half the total, here 10.
Draw a horizontal line from cumulative frequency 10 to the curve, then move vertically down to the score axis. The score is an estimate because the observations have been grouped rather than listed individually.
Check the method specified by the examination or school, including plotting conventions and how to handle data positions. The key idea is to locate the middle of the accumulated count, not to divide the highest score by two.
Quartiles divide an ordered distribution into four parts
The first quartile Q1 is associated with about one-quarter of the observations; the second quartile is the median; the third quartile Q3 is associated with about three-quarters. For a graph representing eighty observations, useful cumulative positions are about 20, 40 and 60.
From those vertical-axis positions, read horizontally to the curve and down to the score axis to estimate the corresponding values. The graph reveals the approximate spread of the middle half of the observations.
Schools and examination questions may specify quartile positions or use a particular convention for raw data. Follow that convention rather than assuming that one interpolation rule must apply to every context.
A complete raw-data quartile example
Consider twelve already ordered scores: 2, 3, 4, 5, 7, 8, 9, 11, 12, 14, 16 and 19. For this worked example, use the median-of-halves convention. The median is the average of the sixth and seventh entries: (8 + 9)/2 = 8.5.
The lower six scores are 2, 3, 4, 5, 7 and 8. Their median is (4 + 5)/2 = 4.5, so Q1 = 4.5. The upper six are 9, 11, 12, 14, 16 and 19. Their median is (12 + 14)/2 = 13, so Q3 = 13.
Thus the five-number summary is minimum 2, Q1 4.5, median 8.5, Q3 13 and maximum 19. If the school supplies a different quartile convention, use the required procedure; the purpose here is to show one consistent method transparently.
Interquartile range measures the middle-half spread
Interquartile range, or IQR, is Q3 − Q1. In the twelve-score example, it is 13 − 4.5 = 8.5. It describes the width of the central half of the ordered observations according to the chosen quartile convention.
The ordinary range is maximum minus minimum: 19 − 2 = 17. The two measures tell different stories. Range uses the extremes, while IQR focuses on the middle portion of the distribution.
If a student simply reports the larger of Q1 and Q3 or subtracts median from maximum, they may remember the term ‘spread’ but not the exact quantity being requested. Ask which two boundaries enclose the middle half.
How to construct a box-and-whisker plot
Use the five-number summary 2, 4.5, 8.5, 13 and 19. Draw a number line with a consistent scale. Draw a box from 4.5 to 13, a line inside it at the median 8.5, and whiskers extending toward the minimum 2 and maximum 19 under the simple min-to-max convention.
The box’s width in a horizontal plot represents the interquartile range, 8.5 in this example. The median line can sit off-centre if the middle-half distribution is asymmetric.
Some statistical software uses special rules for whiskers and marks outliers separately. For school work, follow the plot definition given in the question or current syllabus materials rather than assuming all box plots use identical outlier conventions.
Why a box plot is not a histogram
A box plot summarises position and spread using a handful of statistics. It does not tell you exactly how many observations fall into every equal-width score band. A histogram shows class frequencies and can reveal groups, gaps or peaks in a way a five-number summary cannot.
Two very different data distributions may share a similar median and interquartile range. A box plot can make comparisons quick, but it cannot recover every shape detail that was present in the raw data.
A useful tutor asks pupils what information is missing from the diagram as well as what it displays. That prevents overconfident claims about a dataset based on one graph.
Compare two groups fairly using median and IQR
Suppose Group A has a median of 8.5 and IQR of 8.5, while Group B has a median of 8 and IQR of 4 using the same score scale and quartile convention. The typical central score is slightly higher in A, but the central half is more tightly clustered in B.
It is too strong to conclude from these two numbers alone that every learner in B performed better. Their extremes, distribution shapes and sample sizes may differ, and median does not rank every individual observation.
Mathematical communication matters here. Say precisely which measure is higher or lower and what that measure describes. A fair comparison names the statistic before interpreting it.
Quartiles and percentiles: related but not identical words
A percentile indicates a position within an ordered distribution, usually expressed as a percentage of observations at or below a value. The first quartile corresponds approximately to the 25th percentile, the median to the 50th and the third quartile to the 75th.
That does not mean that a Q3 score of 60 implies ‘75% marks’. It means a position in the distribution: about three-quarters of observations lie at or below that score under the relevant convention.
Ask the student to finish this sentence: ‘This score is at the 75th percentile because …’ A correct answer refers to the proportion of the group below or at the score, not to earning 75 out of 100 marks.
Reading percentiles from a cumulative frequency curve
Imagine a cumulative frequency graph covering forty observations. To estimate the 75th percentile, begin with a cumulative count around 30, locate that height on the vertical axis, follow across to the curve and read the value on the horizontal axis.
Now suppose the curve shows a score of approximately 72 at that position. The interpretation is not that the student achieved 72% of available marks unless that is how scores are defined. It is an approximate score threshold below which about 75% of observations fall.
Be explicit when the axis represents measured time, height or another quantity rather than test marks. A percentile can apply to many ordered numerical variables.
Why a cumulative curve should rise but need not rise evenly
Where many observations fall in a small score interval, cumulative frequency rises more steeply across that interval. Where few observations occur, the curve rises slowly. Flat portions may indicate no observations in the corresponding range, depending on plotting conventions.
The changing steepness is information about where data are concentrated. But a cumulative frequency graph is not itself a speed-time graph; its slope is not necessarily a physical rate. The axes must decide the interpretation.
A child who draws a steep line and says ‘the scores are increasing faster over time’ may be importing language from the wrong topic. Ask which variable lies on the horizontal axis and what is being accumulated.
Grouped data creates estimation, not exact individual values
Suppose eight pupils have scores somewhere between 20 and 29. A grouped table does not say which of those eight scored 20, 25 or 29. When a cumulative frequency curve is used to estimate quartiles inside that band, it makes an assumption about the distribution within the class.
For a more precise median, a complete list of raw scores could be used instead if available. Students should not present a grouped-data estimate as an exact result from a list that was never supplied.
This distinction helps with the confidence of an answer. Use words such as ‘approximately’ when reading a curved graph, and follow the examination’s rounding instructions.
From a cumulative table back to class frequency
A useful reverse exercise begins with cumulative totals 4, 11, 19 and 25. The class frequencies are 4, then 11 − 4 = 7, then 19 − 11 = 8, and finally 25 − 19 = 6. Adding 4 + 7 + 8 + 6 returns the total 25.
This is a good diagnostic for pupils who can construct cumulative totals but cannot interpret them. It also catches a common mistake in which each cumulative total is copied as the frequency of its respective band.
Keep the first row special: it already equals the first frequency when the running total begins at zero. Every later class frequency is the difference between consecutive cumulative totals.
Common errors and the first-line correction
Error one: adding frequencies incorrectly, so the final total is wrong. Error two: plotting a class midpoint against the cumulative count when the task requires upper class boundaries. Error three: reading the median from half the horizontal-axis maximum rather than half the observation count.
Error four: using maximum minus minimum when asked for IQR. Error five: confusing a box’s width with its entire range. Error six: claiming that quartile position automatically tells the percentage mark. Each problem needs a different correction.
After marking, have the pupil identify the first incorrect assumption and answer a parallel question without the worked solution in view. Repeating a correctly copied graph is not a strong test of understanding.
The 2027 SEC K210 Paper 2 Section B choice
The official 2027 K210 assessment has two papers of two hours and seventy marks each, weighted 50% apiece. Paper 2 includes a Section B choice: students answer one of two questions, one from underlined Geometry and Measurement content and the other from underlined Statistics and Probability content.
In the S1 syllabus list, cumulative frequency diagrams, box-and-whisker plots, quartiles and percentiles, interquartile range and standard deviation are among the underlined content. Those parts are designated for Paper 2 Section B under the published syllabus. Each optional question carries seven or eight marks.
That choice is not a reason to ignore Statistics throughout Secondary 3. Data interpretation and measures of central tendency also connect to broader mathematical reasoning and other tasks. Understand the actual format and follow the school’s current examination preparation.
What a tutor should assess before teaching graph construction
Give the student a simple frequency table and ask them to produce running totals. Next, provide a cumulative table and ask for the class frequencies. Then ask for the median estimate on a pre-drawn diagram. These three tasks separate arithmetic, meaning and graphical reading.
One learner may understand quartiles but misplace plotted points because the class boundary is wrong. Another may draw a neat graph but mistake Q3 for 75% of the highest score. Three-pupil attention is useful only if these different failures receive different corrections.
Ask for a delayed independent question. The student should be able to state which axis holds cumulative count, where the total number of observations appears and what the median reading means.
A two-week cumulative frequency and box-plot practice plan
Days 1–2: move between ordinary frequency and cumulative frequency tables. Days 3–4: mark upper class boundaries and plot a sensible running-total graph. Day 5: practise reading counts below a threshold. Day 6: locate an approximate median. Day 7: no-notes mixed review.
Days 8–9: order raw observations and calculate quartiles under the school’s chosen convention. Day 10: calculate range and IQR and explain their difference. Day 11: make a five-number summary and a box plot. Day 12: compare two distributions using median and IQR. Days 13–14: interpret percentiles and complete an unseen mixed problem.
The plan is illustrative and should not crowd out school assignments, rest or prerequisite concepts. A careful correction and a later independent retest matter more than finishing a worksheet every day.
Try three short questions without the model beside you
Question A: class frequencies are 2, 6, 5 and 7. Find cumulative frequencies. Answer: 2, 8, 13 and 20. Question B: cumulative totals are 5, 11, 18 and 24. Find class frequencies. Answer: 5, 6, 7 and 6.
Question C: under the median-of-halves convention, use the sorted dataset 2, 3, 4, 5, 7, 8, 9, 11, 12, 14, 16, 19. Find the median, Q1, Q3 and IQR. Answers: 8.5, 4.5, 13 and 8.5 respectively.
The value of this mini-quiz is not memorising those numbers. Ask whether the learner can explain each difference between a running count, a central position and a spread measure.
How Bukit Timah parents can choose useful Mathematics support
Bring a marked Statistics question and the school’s current topic list. Ask whether the child can explain the difference between frequency and cumulative frequency and whether the tutor teaches accurate graph reading before introducing timed practice.
eduKateSG’s existing reference describes a three-pupil tutorial model near Sixth Avenue. Small groups can allow a tutor to see where one child has misunderstood quartile position while another has a plotting issue, but group size alone is not a learning guarantee.
Confirm the availability of a suitable Secondary 3 G2 group, teacher, fees, timing and exact location directly. The most valuable result is an unfamiliar statistical display interpreted independently with accurate mathematical language.
Frequently asked questions
Are cumulative frequency diagrams examined at G2 in 2027?
Yes. The published SEC K210 S1 list includes cumulative frequency diagrams, box plots, quartiles and related spread measures as underlined content assessed in Paper 2 Section B.
What does cumulative frequency tell me?
It tells how many observations have accumulated up to a value or class boundary, rather than how many appear only in the last class.
How do I estimate Q1, the median and Q3 from a cumulative curve?
Use approximately one-quarter, one-half and three-quarters of the total cumulative count as the starting heights, then read across to the graph and down to the measured quantity. Follow the school’s stated quartile conventions.
Does a box plot reveal every individual score?
No. It summarises the distribution using a minimum, lower quartile, median, upper quartile and maximum under a specified plotting convention.
Is IQR the same as range?
No. Range equals maximum minus minimum; interquartile range equals Q3 minus Q1 and represents the width of the middle half of the ordered observations.
Can candidates skip Statistics because Paper 2 Section B has a choice?
Do not assume that. Only specified underlined material is tied to the Section B choice. Other Statistics skills and broader data interpretation remain useful and may be assessed according to the full syllabus.
Syllabus evidence and related guides
The official SEAB 2027 K210 Mathematics syllabus explains Statistics and Probability S1 and the Paper 2 Section B choice. SEAB’s G2 school-candidate list confirms the K210 2027 code. Use the school’s topic sequence and requested quartile method when practising.
The unchanged eduKateSG small-group Mathematics reference describes three-pupil tutorials near Sixth Avenue. For broad context, read the G2 Statistics, Probability and Data Interpretation guide and the G2 revision and error-analysis guide.
Continue the Secondary 3 G2 Mathematics Bukit Timah series
- Secondary 3 G2 Mathematics Tutor Bukit Timah | Simultaneous Equations: Elimination, Substitution and Graphs
- Secondary 3 G2 Mathematics Tutor Bukit Timah | Quadratic Equations: Formula, Factorisation and Completing the Square
- Secondary 3 G2 Mathematics Tutor Bukit Timah | Coordinate Geometry: Distance, Gradient and Line Equations
Bring one marked graph question to ask what the student needs to repair first. Contact eduKateSG about Secondary 3 G2 Mathematics tutorials. Confirm suitable group availability, fees, teacher, venue and timetable directly.