The mode is sim­ply the value that turns up most often, like the shoe size a shop sells the most of. With grouped data, though, the first thing you can see is only which class turns up most often. That class is called the modal class. In this les­son we use a for­mula that pins down the mode inside that class by com­par­ing its fre­quency with the classes on either side. We'll also see what to do when the modal class sits at one end of the table.

The modal class is the class with the high­est fre­quency. Call its lower limit ll, its width hh and its fre­quency f1f_1. The fre­quency of the class just before it is f0f_0, and of the class just after it, f2f_2.

The for­mula

Mode=l+(f1−f02f1−f0−f2)×h\displaystyle \text{Mode} = l + \left(\frac{f_1 - f_0}{2f_1 - f_0 - f_2}\right) \times h

Here's the idea behind it in plain words. The mode lives inside the modal class, but it gets tugged towards whichever neigh­bour is more crowded, and the frac­tion tells you how far along the class that tug takes it.

That frac­tion always lies between 00 and 11, so the mode always lands inside the modal class. If your answer ends up out­side the class, some­thing went wrong when you sub­sti­tuted.

Worked exam­ples

Exam­ple 1. Stu­dents were asked how many hours they study each week:

Hours 00–55 55–1010 1010–1515 1515–2020 2020–2525
Stu­dents 44 99 1717 1111 55

The modal class is 1010–1515, so l=10l = 10, h=5h = 5, f1=17f_1 = 17, f0=9f_0 = 9, f2=11f_2 = 11.

Mode=10+17−934−9−11×5=10+814×5=10+2.86=12.86 hours\displaystyle \text{Mode} = 10 + \frac{17 - 9}{34 - 9 - 11} \times 5 = 10 + \frac{8}{14} \times 5 = 10 + 2.86 = 12.86 \text{ hours}

Histogram of study hours with frequencies 4, 9, 17, 11, 5; the modal class 10 to 15 (f1 = 17) sits between f0 = 9 and f2 = 11, and a dashed line marks the mode 12.86.
Exam­ple 1: the mode inside the modal class 10 to 15

Exam­ple 2. The ages of patients seen at a clinic:

Age 1010–2020 2020–3030 3030–4040 4040–5050 5050–6060 6060–7070
Patients 66 1111 2121 2323 1414 55

The modal class is 4040–5050, so l=40l = 40, h=10h = 10, f1=23f_1 = 23, f0=21f_0 = 21, f2=14f_2 = 14.

Mode=40+23−2146−21−14×10=40+211×10=41.82 years\displaystyle \text{Mode} = 40 + \frac{23 - 21}{46 - 21 - 14} \times 10 = 40 + \frac{2}{11} \times 10 = 41.82 \text{ years}

See how close the mode is to 4040? The class before (2121) is almost as crowded as the modal class (2323), so the mode gets pulled strongly towards the bound­ary between them.

Histogram of patient ages with frequencies 6, 11, 21, 23, 14, 5; modal class 40 to 50 (f1 = 23) has f0 = 21 before it and f2 = 14 after, and the mode 41.82 sits near 40.
Exam­ple 2: a tall class before pulls the mode towards 40

Exam­ple 3. What if the modal class is the very first one?

Class 00–1010 1010–2020 2020–3030 3030–4040
Fre­quency 1616 1212 88 44

Noth­ing comes before 00–1010, so we take f0=0f_0 = 0.

Mode=0+16−032−0−12×10=1620×10=8\displaystyle \text{Mode} = 0 + \frac{16 - 0}{32 - 0 - 12} \times 10 = \frac{16}{20} \times 10 = 8

In the same way, if the modal class is the last one, take f2=0f_2 = 0.

Con­vert­ing inclu­sive classes

Some­times classes are writ­ten like 11–1010, 1111–2020, 2121–3030, with a gap of 11 between them. The for­mula needs con­tin­u­ous classes, so close the gaps first: take 0.50.5 off every lower limit and add 0.50.5 to every upper limit, giv­ing 0.50.5–10.510.5, 10.510.5–20.520.5, 20.520.5–30.530.5. Now ll is the adjusted lower limit, and hh is still 1010.

Exam­ple 4.

Marks 11–1010 1111–2020 2121–3030 3131–4040
Stu­dents 55 1414 99 22

The modal class is 1111–2020, which becomes 10.510.5–20.520.5, so l=10.5l = 10.5.

Mode=10.5+14−528−5−9×10=10.5+914×10=10.5+6.43=16.93\displaystyle \text{Mode} = 10.5 + \frac{14 - 5}{28 - 5 - 9} \times 10 = 10.5 + \frac{9}{14} \times 10 = 10.5 + 6.43 = 16.93

Mean and mode together

The mean takes every value into account, while the mode only tells you the most com­mon one. Go back to the study-hours data in Exam­ple 1. Its mean, by the direct method with class marks 2.5,7.5,12.5,17.5,22.52.5, 7.5, 12.5, 17.5, 22.5, is 10+67.5+212.5+192.5+112.546=59546≈12.93\displaystyle \frac{10 + 67.5 + 212.5 + 192.5 + 112.5}{46} = \frac{595}{46} \approx 12.93. That's very close to the mode of 12.8612.86, because the dis­tri­b­u­tion is nearly sym­met­ric. When the mean and mode are far apart, it's a sign that the data is lop­sided.

Prac­tice

  1. Find the mode.
Class 00–1010 1010–2020 2020–3030 3030–4040 4040–5050
Fre­quency 33 88 1515 1010 44
  1. Find the mode.
Class 100100–120120 120120–140140 140140–160160 160160–180180 180180–200200
Fre­quency 1212 1414 88 66 1010
  1. Find the mode. Notice that the modal class is the last one.
Class 5050–6060 6060–7070 7070–8080 8080–9090
Fre­quency 55 99 1313 1818
  1. Change these to con­tin­u­ous classes, then find the mode.
Class 11–55 66–1010 1111–1515 1616–2020 2121–2525
Fre­quency 33 77 1212 66 22

Answers

Show answers
  1. 20+15−830−8−10×10=20+712×10=25.83\displaystyle 20 + \frac{15 - 8}{30 - 8 - 10} \times 10 = 20 + \frac{7}{12} \times 10 = 25.83.
  2. The modal class is 120120–140140, so the mode is 120+14−1228−12−8×20=120+28×20=125\displaystyle 120 + \frac{14 - 12}{28 - 12 - 8} \times 20 = 120 + \frac{2}{8} \times 20 = 125.
  3. Here f2=0f_2 = 0, so the mode is 80+18−1336−13−0×10=80+523×10=82.17\displaystyle 80 + \frac{18 - 13}{36 - 13 - 0} \times 10 = 80 + \frac{5}{23} \times 10 = 82.17.
  4. The classes become 0.50.5–5.55.5, …, 20.520.5–25.525.5, and the modal class is 10.510.5–15.515.5. So the mode is 10.5+12−724−7−6×5=10.5+511×5=12.77\displaystyle 10.5 + \frac{12 - 7}{24 - 7 - 6} \times 5 = 10.5 + \frac{5}{11} \times 5 = 12.77.