QUANTILES
Quartiles, Deciles, and Percentiles — Study Notes
Overview: What Are Quantiles?
Quantiles are statistical measures that describe the position of a score within a distribution.
Unlike the mean or median, they do not always try to find the center — they describe where a
particular value falls relative to the rest of the data.
Type Divides Data Into Number of Points Denominator
Quartile 4 equal parts 3 (Q1, Q2, Q3) 4
Decile 10 equal parts 9 (D1–D9) 10
Percentile 100 equal parts 99 (P1–P99) 100
Part 1: Quartiles
Definition
• Divides data into 4 equal parts.
• Q1 (1st Quartile / P25): 25% of data falls below this value.
• Q2 (2nd Quartile / Median / P50): 50% of data falls below this value.
• Q3 (3rd Quartile / P75): 75% of data falls below this value.
Finding Quartiles for Ungrouped Data
Key Rules:
• Always sort the data in ascending order first.
• Q2 is always the median of the entire dataset.
• Q1 is the median of the lower half (values before Q2).
• Q3 is the median of the upper half (values after Q2).
• If n is even, Q2 = average of the two middle values.
• If n is odd, Q2 = the exact middle value.
Steps:
1. Sort the data in ascending order.
2. Find Q2 (median): position = (n+1)/2 for odd n, or average of n/2 and n/2+1 for even n.
3. Split the data at Q2. Find the median of the lower half = Q1.
4. Find the median of the upper half = Q3.
Finding Quartiles for Grouped Data — Formula
Qk = xlb + c × [ (kn/4) − cfb ] / fqk
Symbol Meaning
k Quartile number (1, 2, or 3)
xlb Lower class boundary of the kth quartile class
c Class width
cfb Cumulative frequency BEFORE the quartile class
n Total number of observations
fqk Frequency of the kth quartile class
Important Notes for Grouped Data:
• Class width c = upper limit − lower limit + 1 (for discrete data).
• Lower class boundary (xlb) = lower limit − 0.5.
• The quartile class is whichever class interval contains the (kn/4)th item.
• Always build a cumulative frequency (<cf) column before computing.
Part 2: Deciles
Definition
• Divides data into 10 equal parts.
• D1 through D9 divide the distribution into ten sections.
• D5 is equivalent to the median (Q2 / P50).
• Each decile point Dk tells you the value below which k × 10% of cases fall.
Formula for Grouped Data
Dk = xlb + c × [ (kn/10) − cfb ] / fdk
The variables have the same meanings as in the quartile formula, except:
• k ranges from 1 to 9.
• fdk = frequency of the kth decile class.
• The decile class is the class interval that contains the (kn/10)th item.
Key Relationships
Decile Equivalent Quartile Equivalent Percentile
D2.5 Q1 P25
D5 Q2 / Median P50
D7.5 Q3 P75
Part 3: Percentiles
Definition
• Divides data into 100 equal parts.
• Percentile rank k means k% of the class scored below that value.
• Pk is the score corresponding to the kth percentile rank.
• P0 = lower boundary of the lowest score; P100 = upper boundary of the highest score.
Interpreting Percentile Rank
If a student has a percentile rank of 75:
• They scored higher than 75% of the group.
• They belong to the top 25% of the group.
• Number of people who scored below them = n × 0.75
Formula for Grouped Data
Pk = xlb + c × [ (kn/100) − cfb ] / fpk
Same variables as quartile and decile formulas, except:
• k ranges from 1 to 99.
• fpk = frequency of the kth percentile class.
• The percentile class contains the (kn/100)th item.
Universal Steps for Grouped Data Computations
These 6 steps apply to quartile, decile, AND percentile computations for grouped data:
5. Build the <cf (less-than cumulative frequency) column.
6. Identify the target class: find kn/divisor and locate which class interval contains
that item.
7. Get cfb: the cumulative frequency of the class just BEFORE the target class.
8. Get xlb: lower class boundary = lower class limit − 0.5.
9. Get fk: frequency of the target class.
10. Apply the formula. Round to the nearest hundredths.
Things to Keep in Mind
• Always sort ungrouped data before finding any quantile.
• Q2 = D5 = P50 = Median — they are all the same value.
• Q1 = P25 and Q3 = P75.
• Class width c = upper class limit − lower class limit + 1.
• Lower boundary xlb = lower class limit − 0.5.
• The 'quartile/decile/percentile class' is found by computing kn/divisor — find the class
whose <cf first exceeds this value.
• cfb is the cumulative frequency of the class immediately BEFORE (not including) the
target class.
• Round all grouped data answers to the nearest hundredths (2 decimal places).
• P100 = upper boundary of the highest score; P0 = lower boundary of the lowest score.
Practice Exercises
Exercise 1: Ungrouped Data — Quartiles
Given the following test scores, find Q1, Q2, and Q3:
45, 50, 62, 71, 38, 55, 60, 48, 73, 65
Guide Questions:
11. What is the sorted order of the data?
12. What is the position of Q2? What is the value?
13. What values make up the lower half? What is Q1?
14. What values make up the upper half? What is Q3?
Quartile Answer
Q1
Q2 (Median)
Q3
Exercise 2: Ungrouped Data — Adding a Value
Using your sorted data from Exercise 1, add the value 58 to the dataset. Recompute Q1, Q2,
and Q3.
Hint: With 11 values (odd), Q2 is the 6th value. Split evenly and find Q1 and Q3 of each half.
Quartile Answer
Q1
Q2 (Median)
Q3
Exercise 3: Grouped Data — Quartiles, Deciles, and Percentiles
The table below shows the scores of 80 students on a midterm exam. Use it to answer the
questions that follow.
Class Interval (CI) Frequency (f) < Cumulative Frequency (<cf)
10 – 19 4
20 – 29 8
30 – 39 15
40 – 49 22
50 – 59 18
60 – 69 9
70 – 79 4
Total 80
Tasks:
15. Complete the <cf column.
16. Compute Q1 (first quartile). Show all steps.
17. Compute Q3 (third quartile). Show all steps.
18. Compute D6 (sixth decile). Show all steps.
19. Compute P35 (35th percentile). Show all steps.
Exercise 4: Percentile Rank Interpretation
In a class of 60 students, Maria scored at the 88th percentile in a science exam.
20. What does Maria's percentile rank of 88 imply?
21. How many students scored below Maria?
22. How many students scored higher than Maria?
Answer Key
Exercise 1
• Sorted: 38, 45, 48, 50, 55, 60, 62, 65, 71, 73 (n = 10, even)
• Q2 = (55 + 60) / 2 = 57.5
• Lower half: 38, 45, 48, 50 → Q1 = (45 + 48) / 2 = 46.5
• Upper half: 62, 65, 71, 73 → Q3 = (65 + 71) / 2 = 68
Exercise 2
• New sorted set: 38, 45, 48, 50, 55, 58, 60, 62, 65, 71, 73 (n = 11, odd)
• Q2 = 6th value = 58
• Lower half (before 58): 38, 45, 48, 50, 55 → Q1 = 48
• Upper half (after 58): 60, 62, 65, 71, 73 → Q3 = 65
Exercise 3 — Completed <cf Column
CI f <cf
10–19 4 4
20–29 8 12
30–39 15 27
40–49 22 49
50–59 18 67
60–69 9 76
70–79 4 80
Q1: kn/4 = 80/4 = 20 → class: 30–39
• xlb = 29.5, cfb = 12, fq1 = 15, c = 10
• Q1 = 29.5 + 10 × (20 − 12) / 15 = 29.5 + 5.33 = 34.83
Q3: 3n/4 = 60 → class: 50–59
• xlb = 49.5, cfb = 49, fq3 = 18, c = 10
• Q3 = 49.5 + 10 × (60 − 49) / 18 = 49.5 + 6.11 = 55.61
D6: 6n/10 = 48 → class: 40–49
• xlb = 39.5, cfb = 27, fd6 = 22, c = 10
• D6 = 39.5 + 10 × (48 − 27) / 22 = 39.5 + 9.55 = 49.05
P35: 35n/100 = 28 → class: 40–49
• xlb = 39.5, cfb = 27, fp35 = 22, c = 10
• P35 = 39.5 + 10 × (28 − 27) / 22 = 39.5 + 0.45 = 39.95
Exercise 4
• Maria scored higher than 88% of the class, placing her in the top 12%.
• Students below Maria: 60 × 0.88 = 52.8 ≈ 52 students
• Students above Maria: 60 − 52 − 1 (Maria) = 7 students