• Foundation
  • About 18 min
  • 5 practice questions

Quartiles, percentiles and the mode of a discrete series

Calculate quartiles of ordered data, interpret percentiles, and find the mode of discrete data, including bimodal and no-mode cases.

If something looks wrong, please tell us.

By the end of this lesson you can

  • Define quartiles and percentiles and interpret a percentile rank.
  • Calculate the lower and upper quartiles of ordered data, interpolating where needed.
  • Find the mode of a discrete series and recognise unimodal, bimodal and no-mode data.

The median splits data into two halves. Other positional measures split it into more parts, and the mode looks for the most common value.

Quartiles

Quartiles divide ordered data into four equal parts. There are three: the first (lower) quartile Q1, with 25 per cent of items below it and 75 per cent above; the second quartile Q2, which is the median (50 per cent each side); and the third (upper) quartile Q3, with 75 per cent below and 25 per cent above. Q1 and Q3 are the limits within which the central 50 per cent of the data lies.

For ordered ungrouped data: Q1 = size of the (N + 1) ÷ 4 th item; Q3 = size of the 3(N + 1) ÷ 4 th item. A fractional position is found by interpolating between neighbouring items.

Worked example 1: quartiles at whole-number positions

Illustrative ordered data (N = 11): 5, 8, 12, 15, 18, 21, 24, 27, 30, 34, 40. Q1: (11 + 1) ÷ 4 = 3rd item = 12. Median: (11 + 1) ÷ 2 = 6th item = 21. Q3: 3 × 12 ÷ 4 = 9th item = 30. The central half of the data lies between 12 and 30.

Worked example 2: quartiles by interpolation

Illustrative ordered data (N = 10): 14, 17, 20, 23, 26, 29, 33, 36, 40, 45. Q1 position: (10 + 1) ÷ 4 = 2.75th item, that is, the 2nd item plus 0.75 of the gap to the 3rd: Q1 = 17 + 0.75 × (20 − 17) = 17 + 2.25 = 19.25. Q3 position: 3 × 11 ÷ 4 = 8.25th item: Q3 = 36 + 0.25 × (40 − 36) = 36 + 1 = 37.

For a continuous series, quartiles are found like the median: the Q1 class contains the N ÷ 4 th item and Q1 = L + ((N ÷ 4 − c.f.) ÷ f) × h; for Q3 use 3N ÷ 4. For the marks of 50 students used in lessons 2 and 3, N ÷ 4 = 12.5 falls in class 20-30 (cf before it 12, f = 14), so Q1 = 20 + (0.5 ÷ 14) × 10 ≈ 20.36; 3N ÷ 4 = 37.5 falls in class 30-40 (cf before 26, f = 12), so Q3 = 30 + (11.5 ÷ 12) × 10 ≈ 39.58.

Percentiles

Percentiles divide a distribution into 100 equal parts, giving 99 dividing points P1, P2, ..., P99. P50 is the median (and P25, P75 are Q1 and Q3). A score at the 82nd percentile means 82 per cent of candidates are below that position and 18 per cent are above it.

Worked example 3: interpreting a percentile

Illustrative: 2,00,000 candidates take a made-up entrance test and Meera (made up) is at the 90th percentile. Share above her: 100 − 90 = 10 per cent. Number above her: 10 per cent of 2,00,000 = 20,000. So about 20,000 candidates did better, and about 1,80,000 were at or below her position.

The mode

A shoe manufacturer wants the size in greatest demand; a garment maker wants the most popular shirt style. For such questions the mode is the right average. The word comes from the French 'la mode', meaning the fashion. The mode (Mo) is the most frequently observed value: the value around which the maximum concentration of items occurs.

Worked example 4: mode of a discrete series

Illustrative data: a made-up shop's sales in a month
Shoe sizePairs sold
615
732
841
927
109

The highest frequency is 41, against size 8, so the mode is size 8. Note that the mode is the value (8), not the frequency (41). This series has one mode, so it is unimodal.

How many modes?

  • Unimodal One value occurs most often
    • Example: 2, 3, 3, 4, 5: mode 3
  • Bimodal Two values tie for most frequent
    • Example: 2, 3, 3, 5, 7, 7, 9: modes 3 and 7
  • Multimodal More than two modes
  • No mode No value occurs more often than any other
    • Example: 1, 1, 2, 2, 3, 3: every value twice

Unlike the mean and the median, the mode is not necessarily unique, and a series may have no mode at all.

Key takeaways: quartiles split ordered data into four parts (Q1 at (N + 1) ÷ 4, Q3 at 3(N + 1) ÷ 4, Q2 = median); Q1 to Q3 holds the central 50 per cent. Percentiles split it into 100 parts; the 82nd percentile has 18 per cent above it. The mode is the most frequent value; data can be unimodal, bimodal, multimodal or have no mode.

Check your understanding

  1. 1. Illustrative ordered data: 3, 6, 9, 12, 15, 18, 21. The lower quartile Q1 is:

    1. 9
    2. 6
    3. 12
    4. 18
    Show answer

    Answer: B. N = 7; position of Q1 = (7 + 1) ÷ 4 = 2nd item = 6. (Q3 is the 6th item, 18.)

  2. 2. Illustrative ordered data: 10, 14, 18, 22, 26, 30, 34. The upper quartile Q3 is:

    1. 22
    2. 26
    3. 34
    4. 30
    Show answer

    Answer: D. N = 7; position of Q3 = 3 × (7 + 1) ÷ 4 = 6th item = 30.

  3. 3. A candidate at the 75th percentile of a test taken by 40,000 people has about how many candidates above her?

    1. 10,000
    2. 30,000
    3. 7,500
    4. 25,000
    Show answer

    Answer: A. Share above = 100 − 75 = 25 per cent. 25 per cent of 40,000 = 10,000.

  4. 4. The mode of the series 4, 6, 6, 8, 9, 9, 9, 11 is:

    1. 6
    2. 8
    3. 9
    4. 3
    Show answer

    Answer: C. 9 occurs three times, more than any other value (6 occurs twice), so the mode is 9. 3 is the frequency of the mode, not the mode.

  5. 5. Assertion (A): A series can have more than one mode. Reason (R): Unlike the mean and median, the mode is not necessarily unique.

    1. Both A and R are true, and R is the correct explanation of A
    2. Both A and R are true, but R is not the correct explanation of A
    3. A is true, but R is false
    4. A is false, but R is true
    Show answer

    Answer: A. Both are true, and R explains A: since two or more values can tie for the highest frequency, a series can be bimodal or multimodal.

Privacy · Terms · Refunds · Delivery