13.2 Mode of Grouped Data
The mean uses every observation, but sometimes you want the value that is most typical: the shoe size a shop should stock most of, or the time most customers wait. For grouped data, find the modal class and then use the mode formula to place the mode inside it.
A shoe shop owner does not care about the average shoe size of her customers. The average might be $7.4$, and no one wears a shoe of size $7.4$. What she needs to know is the size that is bought most often, so that she stocks plenty of it. A bus company asks the same question about waiting times, and a doctor about the age group that falls ill most. The most frequent value is the mode.
For a plain list the mode is easy to spot. For grouped data we can only say which class is the most crowded, but we can do better: the exact position of the mode inside that class depends on how crowded the two neighbouring classes are. A neat formula captures this, and you will see why it works from a picture of the histogram.
Mode and modal class
With grouped data we need a way to pick a single number from the modal class. Think of three neighbouring bars of a histogram. Let the modal class have lower limit $l$ and class size $h$. Write $f_1$ for the frequency of the modal class, $f_0$ for the frequency of the class just before it, and $f_2$ for the frequency of the class just after it. Two ideas decide where the mode sits:
- if the class before is nearly as crowded as the modal class ($f_0$ close to $f_1$), the mode is pulled towards the lower end of the modal class;
- if the class after is nearly as crowded ($f_2$ close to $f_1$), the mode is pulled towards the upper end.
Why it works. Draw the histogram. Join the top-right corner of the modal bar to the top-right corner of the bar before it, and join the top-left corner of the modal bar to the top-left corner of the bar after it. The two lines cross above the modal class (you can see this in Interactive 1), and the mode is the $x$-coordinate of the crossing. Measure $t=\dfrac{x-l}{h}$ along the modal class. The first line rises from height $f_0$ to $f_1$, so its height at $t$ is $f_0+(f_1-f_0)t$. The second falls from $f_1$ to $f_2$, so its height is $f_1+(f_2-f_1)t$. At the crossing the heights are equal: $$f_0+(f_1-f_0)t=f_1+(f_2-f_1)t\ \Longrightarrow\ t\,(2f_1-f_0-f_2)=f_1-f_0\ \Longrightarrow\ t=\frac{f_1-f_0}{2f_1-f_0-f_2}.$$ So the mode is $l+th$, which is the formula. In words, the modal class is split in the ratio $(f_1-f_0):(f_1-f_2)$, and each neighbour pulls in proportion to how crowded it is. The denominator $2f_1-f_0-f_2=(f_1-f_0)+(f_1-f_2)$ is positive because $f_1$ is the largest frequency.
- Make sure the classes touch. If they are written as $10$โ$19$, $20$โ$29$, subtract $0.5$ from each lower limit and add $0.5$ to each upper limit ($9.5$โ$19.5$, $19.5$โ$29.5$).
- Find the highest frequency $f_1$: that row is the modal class.
- Read off $l$ (its lower limit), $h$ (class size), $f_0$ (the class listed just before it) and $f_2$ (class after it).
- Substitute in $l+\dfrac{f_1-f_0}{2f_1-f_0-f_2}\,h$. Check that the answer lies between $l$ and $l+h$.
1. Swapping $f_0$ and $f_2$. $f_0$ is the class before the modal class (smaller values), $f_2$ the class after. A swap gives a wrong answer that still looks reasonable.
2. Reading $l$ from the wrong class. $l$ is the lower limit of the modal class, not of the class before.
3. Using the formula for non-touching classes. Convert $10$โ$19$ type classes first.
4. Forgetting $h$ at the end, or writing $2f_1-f_0+f_2$ in the denominator. The answer must fall inside the modal class; if not, recheck.
Special cases and missing frequencies
Two situations come up often. First, the neighbours are equal: if $f_0=f_2$ the formula gives $t=\dfrac{f_1-f_0}{2(f_1-f_0)}=\dfrac12$, so the mode is exactly the middle of the modal class, its class mark. Second, the modal class is the first or the last class, so one neighbour is missing. A missing neighbour is treated as an empty class, with frequency $0$: use $f_0=0$ for the first class (or $f_2=0$ for the last).
| Class | $10$โ$20$ | $20$โ$30$ | $30$โ$40$ | $40$โ$50$ | $50$โ$60$ |
|---|---|---|---|---|---|
| Frequency | 5 | 10 | 16 | $p$ | 3 |
Exercises
- 1 Find the mode of $3,5,5,7,8,5,9,3$.
- 2 Find the mode of $10,12,12,14,15,15,15,16$.
- 3 Find the modal class: $0$โ$5$: $3$; $5$โ$10$: $9$; $10$โ$15$: $14$; $15$โ$20$: $8$.
- 4 Evaluate $l+\dfrac{f_1-f_0}{2f_1-f_0-f_2}\,h$ for $l=40$, $h=5$, $f_0=6$, $f_1=14$, $f_2=10$ (to $2$ d.p.).
- 5 Why can the denominator $2f_1-f_0-f_2$ never be zero when $f_1$ is the largest frequency?
- 1 $10$โ$20$: $2$; $20$โ$30$: $6$; $30$โ$40$: $11$; $40$โ$50$: $8$; $50$โ$60$: $3$.
- 2 Heights (cm): $140$โ$145$: $4$; $145$โ$150$: $9$; $150$โ$155$: $15$; $155$โ$160$: $10$; $160$โ$165$: $2$.
- 3 Daily wages (โน): $400$โ$500$: $6$; $500$โ$600$: $10$; $600$โ$700$: $14$; $700$โ$800$: $12$; $800$โ$900$: $8$.
- 4 Pocket money (โน): $0$โ$50$: $8$; $50$โ$100$: $14$; $100$โ$150$: $20$; $150$โ$200$: $12$; $200$โ$250$: $6$.
- 5 Pages read: $1$โ$10$: $6$; $11$โ$20$: $9$; $21$โ$30$: $10$; $31$โ$40$: $5$.
- 6 Rainfall (mm): $0$โ$4$: $18$; $4$โ$8$: $10$; $8$โ$12$: $7$; $12$โ$16$: $5$.
- 1 The mode of: $20$โ$30$: $5$; $30$โ$40$: $9$; $40$โ$50$: $15$; $50$โ$60$: $q$; $60$โ$70$: $4$ is $46$. Find $q$.
- 2 A modal class $10$โ$20$ has $f_0=14$, $f_1=20$, $f_2=6$. Find the mode and say whether it is above or below the class mark $15$, and why.
- 3 A modal class $40$โ$50$ has $f_0=f_2=8$ and $f_1=20$. Show that the mode is $45$.
- 4 Ages of $80$ patients in a clinic: $5$โ$15$: $6$; $15$โ$25$: $11$; $25$โ$35$: $21$; $35$โ$45$: $23$; $45$โ$55$: $14$; $55$โ$65$: $5$. Find the modal age (to $2$ d.p.).
- 5 Speeds of $55$ vehicles on a highway (km/h): $40$โ$50$: $6$; $50$โ$60$: $14$; $60$โ$70$: $23$; $70$โ$80$: $9$; $80$โ$90$: $3$. Find the modal speed.
- 1 Show that the mode given by the formula always lies in the modal class, that is, $l\le\text{Mode}\le l+h$.
- 2 A modal class $40$โ$50$ has $f_0=10$ and $f_2=12$ and the mode is $45.5$. Find $f_1$.
- 3 If the first class is the modal class, $f_0=0$ and $f_2=0$ (a single bar), where is the mode? Explain.
How this is examined: Boards ask for the mode of a table in 3 marks (identify the modal class, list $l,h,f_0,f_1,f_2$, substitute), a missing-frequency item in 2โ3 marks, and 1-mark MCQs on the formula. Many marks are lost by swapping $f_0$ and $f_2$ or by forgetting to make classes continuous. JEE Main problems tend to give frequencies as a ratio.
E1. Board ยท 1 mark The modal class of a distribution is $20$โ$30$. The class before it has frequency $8$, the modal class has frequency $15$, and the class after it has frequency $11$. The mode is (A) $26.36$ (B) $25$ (C) $23.64$ (D) $30$
E2. Board ยท 1 mark Assertion (A): If the modal class is $30$โ$40$ and the frequencies of the classes before and after it are equal, the mode is $35$. Reason (R): When $f_0=f_2$, the mode formula gives $l+\dfrac h2$.
(A) Both A and R are true and R is the correct explanation of A. (B) Both A and R are true but R is not the correct explanation of A. (C) A is true but R is false. (D) A is false but R is true.
E3. Board ยท 3 marks Find the mode: $15$โ$20$: $4$; $20$โ$25$: $11$; $25$โ$30$: $19$; $30$โ$35$: $14$; $35$โ$40$: $2$.
E4. Board ยท 2 marks The mode of the following distribution is $56$. Find $p$: $30$โ$40$: $5$; $40$โ$50$: $12$; $50$โ$60$: $18$; $60$โ$70$: $p$; $70$โ$80$: $6$.
E5. Board ยท Case study A bank manager recorded the waiting times (in minutes) of $60$ customers: $0$โ$5$: $5$; $5$โ$10$: $14$; $10$โ$15$: $21$; $15$โ$20$: $12$; $20$โ$25$: $8$. (i) Write the modal class and the values of $l$, $h$, $f_0$, $f_1$, $f_2$. (2) (ii) Find the mode. (1) (iii) How many customers waited $15$ minutes or more, and what fraction of all customers is that? (1)
E6. JEE Main For a grouped distribution with class size $10$, the modal class is $20$โ$30$ and the frequencies of the preceding, modal and succeeding classes are in the ratio $2:5:3$. The mode is (A) $24$ (B) $25$ (C) $26$ (D) $27.5$
Research question. If you measure the same thing for 40 people, how does the mode you find from the raw values compare with the mode from the grouped formula, and how does it change with the class size?
Why it matters. Shops decide what stock to order, cities decide when to run extra buses, and schools decide how many sections to open by looking at the most common value, not the average. When the raw data are unavailable, the grouped mode is the best estimate.
What to do.
- Collect one whole-number measurement from 40 people, such as minutes spent travelling to school.
- Make a tally of the raw values and note the most frequent value (if several values tie, group first).
- Group the values with class size $5$, find the modal class and compute the mode with the formula.
- Repeat with class size $10$.
- Draw both histograms (a spreadsheet chart will do) and mark the mode on each.
Tools: A notebook or a spreadsheet, a calculator
Hand in: The raw data, both frequency tables, the two modes, the histograms, and a short comment on which class size told you more.
Going further. Compare the mode with the mean of the same grouped data (Section 13.1). Which value would the bus company prefer, and which would a student planning her morning prefer?
๐งช Self-check quiz
Q1. The mode of $4,6,6,7,6,9,4$ is
Q2. In the table $0$โ$10$: $3$; $10$โ$20$: $11$; $20$โ$30$: $8$; $30$โ$40$: $5$, the modal class is
Q3. A modal class $40$โ$50$ has $f_0=f_2=7$ and $f_1=15$. The mode is
Q4. For $l=10$, $h=10$, $f_0=4$, $f_1=10$, $f_2=6$ the mode is
Q5. Which statement about the mode of grouped data is correct?