Octiles Overview
Octiles divide a dataset into 8 equal parts, with 7 dividing points: O₁, O₂, …, O₇. Each octile represents 12.5% of the data. Octiles are less commonly used than quartiles or deciles but are useful for more granular data division.
Octiles for Ungrouped Data
Formula
The formula for the i-th octile in ungrouped data is:
$$O_i = \text{Value of } \left(\frac{i(n+1)}{8}\right)^{th} \text{ observation, } i=1,2,…,7$$
where n is the total number of observations.
Example: Ungrouped Data
Test scores of 32 students (arranged in ascending order):
25, 28, 30, 32, 35, 38, 40, 42, 45, 48, 50, 52, 55, 58, 60, 62, 65, 68, 70, 72, 75, 78, 80, 82, 85, 88, 90, 92, 95, 98, 100, 102
Find O₁, O₂, O₄, and O₇.
Solution:
n = 32
First Octile (O₁):
$$O_1 = \text{Value of } \left(\frac{1(32+1)}{8}\right)^{th} \text{ obs.} = \text{Value of } (4.125)^{th} \text{ obs.}$$
$$= 4^{th} + 0.125(5^{th} - 4^{th}) = 32 + 0.125(35 - 32) = 32.375$$
Second Octile (O₂):
$$O_2 = \text{Value of } \left(\frac{2(32+1)}{8}\right)^{th} \text{ obs.} = \text{Value of } (8.25)^{th} \text{ obs.}$$
$$= 8^{th} + 0.25(9^{th} - 8^{th}) = 42 + 0.25(45 - 42) = 42.75$$
Fourth Octile (O₄ - Median):
$$O_4 = \text{Value of } \left(\frac{4(32+1)}{8}\right)^{th} \text{ obs.} = \text{Value of } (16.5)^{th} \text{ obs.}$$
$$= 16^{th} + 0.5(17^{th} - 16^{th}) = 62 + 0.5(65 - 62) = 63.5$$
Seventh Octile (O₇):
$$O_7 = \text{Value of } \left(\frac{7(32+1)}{8}\right)^{th} \text{ obs.} = \text{Value of } (28.875)^{th} \text{ obs.}$$
$$= 28^{th} + 0.875(29^{th} - 28^{th}) = 92 + 0.875(95 - 92) = 94.625$$
Octiles for Grouped Data
Formula
For grouped data:
$$O_i = L + \left(\frac{\frac{i \cdot N}{8} - CF}{f}\right) \times h$$
where:
- L = Lower class boundary of the octile class
- N = Total frequency
- CF = Cumulative frequency before the octile class
- f = Frequency of the octile class
- h = Class width
- i = 1, 2, 3, …, 7
Example: Grouped Data
| Class Interval | Frequency |
|---|---|
| 0-10 | 5 |
| 10-20 | 8 |
| 20-30 | 12 |
| 30-40 | 14 |
| 40-50 | 10 |
| 50-60 | 6 |
Find O₁, O₄, and O₇.
Solution:
Cumulative Frequencies:
| Class | f | CF |
|---|---|---|
| 0-10 | 5 | 5 |
| 10-20 | 8 | 13 |
| 20-30 | 12 | 25 |
| 30-40 | 14 | 39 |
| 40-50 | 10 | 49 |
| 50-60 | 6 | 55 |
N = 55
O₁ (First Octile):
Position = (1 × 55)/8 = 6.875
Octile class: 10-20
$$O_1 = 10 + \left(\frac{6.875 - 5}{8}\right) \times 10 = 12.34$$
O₄ (Fourth Octile - Median):
Position = (4 × 55)/8 = 27.5
Octile class: 30-40
$$O_4 = 30 + \left(\frac{27.5 - 25}{14}\right) \times 10 = 31.79$$
O₇ (Seventh Octile):
Position = (7 × 55)/8 = 48.125
Octile class: 40-50
$$O_7 = 40 + \left(\frac{48.125 - 39}{10}\right) \times 10 = 49.125$$
Octile Interpretation
| Octile | Percentage Below |
|---|---|
| O₁ | 12.5% |
| O₂ | 25.0% |
| O₃ | 37.5% |
| O₄ | 50.0% (median) |
| O₅ | 62.5% |
| O₆ | 75.0% |
| O₇ | 87.5% |
Octiles vs Other Position Measures
| Measure | Parts | Dividing Points |
|---|---|---|
| Quartiles | 4 | 3 (Q₁, Q₂, Q₃) |
| Quintiles | 5 | 4 |
| Octiles | 8 | 7 |
| Deciles | 10 | 9 |
| Percentiles | 100 | 99 |
When to Use Octiles
- Detailed data distribution analysis
- When you need 8-part segmentation (e.g., weekly data analysis)
- Academic or research contexts requiring finer granularity
- Performance evaluation with 8 performance tiers
- Quality control with 8-level acceptance criteria
Relationship to Other Measures
- O₁ = P₁₂.₅ (First octile = 12.5th percentile)
- O₂ = P₂₅ = Q₁ (Second octile = 25th percentile = First quartile)
- O₄ = P₅₀ = Q₂ (Fourth octile = Median)
- O₆ = P₇₅ = Q₃ (Sixth octile = 75th percentile = Third quartile)