How to Learn Cumulative Frequency and Box Plots Quickly and Correctly | Quick Notes

To learn Cumulative Frequency and Box Plots quickly and correctly, begin with the idea of counting how many observations lie at or below a point. Ordinary frequency tells you how many values fall into one category or interval; cumulative frequency adds the counts as you move along the scale. A box plot then uses five key statistics to summarise the position and spread of numerical data. These eduKateSG Mathematics Quick Notes explain how to make a cumulative frequency table, read a graph, calculate quartiles from raw data and compare distributions without guessing.

• Frequency: the number of observations in a particular value or interval.
• Cumulative frequency: a running total as values or intervals increase.
• Total observations: the final cumulative frequency.
• Median: the middle position after ordering the values.
• Quartiles: lower quartile Q1 and upper quartile Q3 describe the spread of the middle half.
• Interquartile range: IQR = Q3 − Q1.
• Box plot: shows a five-number summary, using the conventions specified for the question.

First, practise cumulative frequency as a running addition. Suppose 20 students have waiting times grouped into four intervals: 0–10 minutes: 4 students; 10–20: 6; 20–30: 8; and 30–40: 2. To avoid counting a boundary twice, understand the classes as 0 ≤ t < 10, 10 ≤ t < 20, 20 ≤ t < 30 and 30 ≤ t < 40. The cumulative frequencies at the upper boundaries are 4, 10, 18 and 20. Each entry includes the students in all earlier intervals. The final number 20 is the total sample size, which also equals 4 + 6 + 8 + 2.

Second, learn to draw the graph from upper class boundaries. Put the measured variable, waiting time, on the horizontal axis and cumulative frequency on the vertical axis. Begin at the lower boundary with (0, 0). Then plot (10, 4), (20, 10), (30, 18) and (40, 20). Join them as directed by your syllabus, commonly with a smooth cumulative frequency curve. The graph must not fall as x increases because a running count cannot decrease. You can use it to estimate how many observations lie below a chosen time and to locate the median or quartiles.

Third, understand the five-number summary before drawing a box plot. Order the observations from smallest to largest and find the minimum, lower quartile Q1, median Q2, upper quartile Q3 and maximum. The box runs from Q1 to Q3, with a line at the median; its width represents the interquartile range. For a basic box-and-whisker plot, whiskers extend towards the minimum and maximum. Some statistical box-plot conventions instead mark outliers separately and end the whiskers at the most extreme non-outlier values. Follow the convention stated in your school question.

Worked example: make the cumulative frequency table

For the 20 waiting times grouped above, write each interval beside its frequency and running total. At 10 minutes, the cumulative count is 4. At 20 minutes, it is 4 + 6 = 10. At 30 minutes, it is 10 + 8 = 18. At 40 minutes, it is 18 + 2 = 20. Check that the final cumulative total equals the sum of the four ordinary frequencies. The differences between successive cumulative counts, 4, 6, 8 and 2, should reproduce the original interval frequencies.

Finding the median from a cumulative frequency graph

With 20 observations, the cumulative frequency halfway through the sample is 20/2 = 10. On the graph, move horizontally from cumulative frequency 10 to the curve and then down to the time axis. In our illustrative grouped table, the cumulative count reaches 10 at the upper boundary of 20 minutes, so a graph using the class boundaries estimates the median at about 20 minutes. This is a grouped-data estimate, not proof that an actual observed waiting time was exactly 20 minutes.

Finding quartiles from a grouped-data curve

In the common cumulative-frequency graph method, find the lower quartile near cumulative frequency N/4 and the upper quartile near 3N/4. With N = 20, these positions are 5 and 15. Read across from 5 and 15 on the cumulative frequency axis to estimate Q1 and Q3. If you join the interval-boundary points linearly, Q1 lies in the 10–20 interval at approximately 11.7 minutes, and Q3 lies in the 20–30 interval at approximately 26.3 minutes. The estimated IQR is about 14.6 minutes. These values use interpolation within grouped intervals; the exact individual waiting times are unknown.

Why the graph estimates rather than recovers exact data

Grouped intervals tell us how many values fall into each band, but not exactly where those values lie inside it. Six students in a 10–20-minute interval might be clustered near 11 minutes, near 19 minutes, or spread across the band. A curve drawn from class boundaries cannot reveal the missing details. Linear interpolation makes a reasonable mathematical estimate under an assumed approximately even distribution within each interval. Be careful not to present estimated grouped-data quartiles as exact observed values. The exact method and position convention used in examination questions should follow the stated syllabus.

Worked example: exact box-plot values from raw data

Suppose the ten values, already in increasing order, are 2, 3, 4, 5, 7, 8, 10, 12, 14, 18. The minimum is 2 and the maximum is 18. The median is the average of the fifth and sixth values: (7 + 8)/2 = 7.5. Using the school-style median-of-halves method, the lower half is 2, 3, 4, 5, 7, whose middle is Q1 = 4. The upper half is 8, 10, 12, 14, 18, whose middle is Q3 = 12. Draw the box from 4 to 12 with its median line at 7.5, and basic whiskers reaching 2 and 18.

Interquartile range versus full range

For the same ten values, the full range is maximum minus minimum: 18 − 2 = 16. The interquartile range is Q3 − Q1 = 12 − 4 = 8. The full range is sensitive to the two extreme observations, while the IQR describes the width of the middle half of the data under the chosen quartile convention. The IQR is a measure of spread, not an average. Always keep the two calculations separate.

How to read a basic box plot

• Left whisker endpoint: minimum under the basic min-to-max convention.
• Left edge of box: lower quartile Q1.
• Line inside box: median.
• Right edge of box: upper quartile Q3.
• Right whisker endpoint: maximum under the basic convention.
• Box width: IQR = Q3 − Q1.
• Whole whisker-to-whisker span: range when endpoints represent min and max.

Quartile conventions are not universal

For ungrouped data, textbooks and statistical software sometimes use different rules for calculating Q1 and Q3, especially with an even sample size or when a quartile position falls between ranks. This lesson used the median-of-halves convention for the exact ten-number example, and the N/4 and 3N/4 graph method for grouped cumulative frequencies. Both are common educational approaches, but do not mix their numerical conventions midway through one question. Use the explicit method given by your teacher, syllabus or examination.

Comparing two groups with box plots

Suppose Group A has median 12 and IQR 4, while Group B has median 15 and IQR 9, measured in the same units. Group B has a higher median, so its typical central value is higher by that measure. Group A has the smaller IQR, so its middle half is more tightly clustered. A box plot does not show every individual data value, and two groups may overlap considerably even when their medians differ. Compare plots drawn on the same scale and state exactly what the statistics support.

How a cumulative frequency curve differs from a histogram

A histogram describes how observations are distributed across numerical intervals. A cumulative frequency curve describes how many observations have accumulated up to each boundary. The histogram’s interval frequencies may rise and fall, but cumulative frequencies cannot decrease as x increases. If a grouped table uses unequal interval widths, histogram heights need frequency density, while the cumulative frequencies are still formed by adding the actual class counts. These are two different views of the same observations.

A six-step method for a grouped-data question

1. Write all class intervals with clear lower and upper boundaries.
2. Add successive frequencies and check that the final cumulative total is N.
3. Plot cumulative frequency against the upper class boundary, starting from zero at the first lower boundary.
4. Read estimated median and quartile positions using the graph convention required.
5. Calculate the estimated IQR as Q3 minus Q1.
6. Label every estimated result clearly and avoid claiming exact positions that grouped data do not reveal.

Common mistakes and how to repair them

• Starting each cumulative count from zero: keep adding the earlier frequencies.
• Plotting frequencies rather than cumulative totals: a cumulative curve uses running counts.
• Allowing the curve to fall: cumulative frequencies cannot decrease.
• Using class midpoints instead of upper boundaries: cumulative frequency points normally use the upper class boundary.
• Confusing IQR with range: Q3 − Q1 is not maximum minus minimum.
• Assuming a grouped-data quartile is exact: it is an estimate without individual observations.
• Using mismatched scales to compare two box plots: compare on a consistent number line.
• Ignoring the specified quartile or whisker convention: read the question’s method carefully.

Mini-test: check the concepts

1. Four consecutive class frequencies are 3, 7, 6 and 4. Write the cumulative frequencies.
2. How many observations are represented by those four frequencies?
3. An ordered set is 2, 3, 4, 5, 7, 8, 10, 12, 14, 18. Find its median.
4. Using the median-of-halves method, find Q1 and Q3 for that set.
5. Find its interquartile range and full range.
6. Can a valid cumulative frequency curve decrease as the measured value increases?

Answers and explanations

1. The cumulative totals are 3, 10, 16, 20. 2. There are 20 observations. 3. Median = (7 + 8)/2 = 7.5. 4. Lower-half median is 4; upper-half median is 12. 5. IQR = 12 − 4 = 8, and range = 18 − 2 = 16. 6. No; a running count stays level or increases, never falls.

Your twelve-minute data interpretation routine

1. Minutes 1–3: turn a short grouped frequency table into cumulative totals.
2. Minutes 4–6: label axes and plot a cumulative frequency curve using class boundaries.
3. Minutes 7–8: locate the approximate median and quartile positions.
4. Minutes 9–10: find the five-number summary for a short ordered raw-data set.
5. Minutes 11–12: draw a box plot and explain the difference between median, IQR and range.

Continue with Mean, Median and Mode, Statistical Charts and Data Interpretation, and Probability. For the broader learning approach, see Secondary 1 Mathematics Tutor Clementi | Small Groups Tutorials, which remains unchanged.