0% found this document useful (0 votes)
5 views13 pages

Topic 5 SSC201

The document discusses measures of partition in statistics, specifically focusing on quartiles and deciles. It explains how to calculate these measures for both ungrouped and grouped data, providing formulas and illustrative examples. The document serves as a guide for understanding how to divide distributions into specified parts for analysis.

Uploaded by

robiatraji3
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
5 views13 pages

Topic 5 SSC201

The document discusses measures of partition in statistics, specifically focusing on quartiles and deciles. It explains how to calculate these measures for both ungrouped and grouped data, providing formulas and illustrative examples. The document serves as a guide for understanding how to divide distributions into specified parts for analysis.

Uploaded by

robiatraji3
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

SSC201: STATISTICAL METHODS AND SOURCE 1

Department of Economics
Faculty of Social Sciences
Obafemi Awolowo University
Ile-Ife, Nigeria

Topic 5: Measures of Partition

These are measures that divide a distribution into a specified number of parts. They are also called
quantiles/fractiles. Examples include quartiles, deciles and percentiles.

1. Quartiles

As the name suggest, this deals with the division of a distribution into 4 equal parts. The first quartile is
known as the lower quartile, Q1, and it is ¼ of a distribution. The second quartile, Q2, is the division of a
distribution into two equal parts usually referred to as the median and it is ½ (2/4) of a distribution. The
third quartile, Q3, is known as the upper quartile and it is ¾ of a distribution.

Case A: Ungrouped data (individual observations and Discrete Series)

For an ungrouped data, the upper (Q3) and the lower (Q1) quartiles can be calculated as follows:

𝑸𝟏 = 𝟏⁄𝟒 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

𝑸𝟑 = 𝟑⁄𝟒 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

Note: 𝑛 = ∑ 𝑓 for a discrete series.

Illustration

1. Given the series, 3, 4, 1, 2, 7, 12 and 5. Find the lower and upper quartiles.

Solution

Re-arranging the data: 1, 2, 3, 4, 5, 7, 12

1
a. 𝑄1 = 1⁄4 (𝑛 + 1)𝑡ℎ term of the distribution

n = 7; 𝑄1 = 1⁄4 (7 + 1)𝑡ℎ term; 𝑄1 = 1⁄4 (8)𝑡ℎ term = 2nd item of the distribution
𝑄1 = 2
b. 𝑄3 = 3⁄4 (𝑛 + 1)𝑡ℎ term of the distribution

n = 7; 𝑄3 = 3⁄4 (7 + 1)𝑡ℎ term; 𝑄1 = 3⁄4 (8)𝑡ℎ term = 6th item of the distribution
𝑄3 = 7

2. Find Q1 and Q3 from the following distribution; 1, 5, 3, 6, 9, and 8.

Solution

Re-arranging the data: 1, 3, 5, 6, 8, 9

a. 𝑄1 = 1⁄4 (𝑛 + 1)𝑡ℎ term of the distribution

n = 6; 𝑄1 = 1⁄4 (6 + 1)𝑡ℎ term; 𝑄1 = 1⁄4 (7)𝑡ℎ term = 1.75th item of the distribution.
The first (lower) quartile falls between the 1st and 2nd items of the distribution, that is, 1 and 3.
𝑄1 = 1st value + 0.75 of the distance between the 1st and 2nd values.
𝑄1 = 1 + 0.75(3 − 1) = 1 + 1.5 = 2.5
b. 𝑄3 = 3⁄4 (𝑛 + 1)𝑡ℎ term of the distribution

n = 6; 𝑄3 = 3⁄4 (6 + 1)𝑡ℎ term; 𝑄1 = 3⁄4 (7)𝑡ℎ term = 5.25th item of the distribution
The third (upper) quartile falls between the 5th and 6th items of the distribution, that is, 8 and 9.
𝑄3 = 8th value + 0.25 of the distance between the 5th and 6th values.
𝑄3 = 8 + 0.25(9 − 8) = 8 + 0.25 = 8.25
3. Obtain the lower and upper quartiles from the distribution below.

x 10 15 20 30 40 50
f 3 2 5 1 4 6

2
Solution

To obtain the quartiles, we need the cumulative frequency.

x f CF
10 3 3
15 2 5
20 5 10
30 1 11
40 4 15
50 6 21
Total ∑ 𝒇 = 𝟐𝟏

a. 𝑄1 = 1⁄4 (𝑛 + 1)𝑡ℎ term of the distribution

𝑛 = ∑ 𝑓 = 21; 𝑄1 = 1⁄4 (21 + 1)𝑡ℎ term; 𝑄1 = 1⁄4 (22)𝑡ℎ term = 5.5th item of the
distribution.
𝑄1 = 20 (it falls within the third CF)
b. 𝑄3 = 3⁄4 (𝑛 + 1)𝑡ℎ term of the distribution

𝑛 = ∑ 𝑓 = 21; 𝑄3 = 3⁄4 (21 + 1)𝑡ℎ term; 𝑄1 = 3⁄4 (22)𝑡ℎ term = 16.5th item of the
distribution
𝑄3 = 50 (it falls within the last CF)

Case B: Grouped data (Continuous Series)

To compute the lower and upper quartiles for a continuous series, the procedure is similar to the one used
in calculating the media. The lower and upper quartiles can be calculated as follows:

𝑵⁄ − 𝑪𝑭
𝑸𝟏 = 𝑳𝒄 + [ 𝟒
𝑸𝟏
]𝒄
𝑭 𝑸𝟏

where Lc is the lower-class boundary of the first quartile class

N is the sum of the frequency, that is, ∑ 𝑓

𝐶𝐹𝑄1 is the cumulative frequency of the class before the first quartile class

3
𝐹𝑄1 is the frequency of the first quartile class

c is the class size of the quartile class

𝑁⁄ is used to determine the first quartile class


4

𝟑𝑵⁄ − 𝑪𝑭
𝑸𝟑 = 𝑳𝒄 + [ 𝟒 𝑸𝟑
]𝒄
𝑭 𝑸𝟑

where Lc is the lower-class boundary of the third quartile class

N is the sum of the frequency, that is, ∑ 𝑓

𝐶𝐹𝑄3 is the cumulative frequency of the class before the third quartile class

𝐹𝑄3 is the frequency of the third quartile class

3𝑁⁄ is used to determine the third quartile class


4

Illustration

Given the table below, compute Q1 and Q3.

Class Interval 1-10 11-20 21-30 31-40 41-50


Frequency 3 8 15 9 5

Solution

Class Interval f CF Class Boundary


1-10 3 3 0.5-10.5
11-20 8 11 10.5-20.5
21-30 15 26 20.5-30.5
31-40 9 35 30.5-40.5
41-50 5 5 40.5-50.5
Total ∑ 𝒇 = 𝟒𝟎

𝑵⁄ −𝑪𝑭
𝟒 𝑸𝟏
a. 𝑸𝟏 = 𝑳𝒄 + [ ]𝒄
𝑭𝑸𝟏

4
∑ 𝑓⁄ 40
We first obtain the quartile class, 𝑁⁄4 = 4 = ⁄4 = 10𝑡ℎ item of the distribution.

The quartile class, 𝑁⁄4, is 11-20 (it falls within the second CF)

Lc = 10.5; 𝐶𝐹𝑄1 = 3; 𝐹𝑄1 = 8; 𝑐 = 20.5 − 10.5 = 10

40⁄ −3 10−3 7
4
𝑄1 = 10.5 + [ ] 10; 𝑄1 = 10.5 + [ ] 10; 𝑄1 = 10.5 + [8] 10
8 8

𝑄1 = 10.5 + 8.75 = 19.25

𝟑𝑵⁄ −𝑪𝑭
𝟒 𝑸𝟑
b. 𝑸𝟑 = 𝑳𝒄 + [ ]𝒄
𝑭𝑸𝟑

3 ∑ 𝑓⁄ 3(40)⁄
We first obtain the quartile class, 3𝑁⁄4 = 4= 4 = 30𝑡ℎ item of the distribution.

The quartile class, 3 𝑁⁄4, is 31-40 (it falls within the second to the last CF)

Lc = 30.5; 𝐶𝐹𝑄3 = 26; 𝐹𝑄3 = 9; 𝑐 = 40.5 − 30.5 = 10

3(40)⁄
4−26 30−26 4
𝑄3 = 30.5 + [ ] 10; 𝑄3 = 30.5 + [ ] 10; 𝑄3 = 30.5 + [9] 10
9 9

𝑄3 = 30.5 + 4.44 = 34.94

2. Deciles

Deciles are measures of partition that divide a distribution into 10 equal parts. A decile is any of the nine
values that divide the sorted data into ten equal parts, so that each part represents 1/10 of the distribution.
D1, D2, D3,… D9 represent the first, second, third …and nineth deciles. Note: the 5th decile represents the
median of a distribution.

Case A: Ungrouped data (individual observations and Discrete Series)

For an ungrouped data, the first (D1), second (D2), third(D3) …. and nineth (D9) deciles can be calculated
as follows:

5
𝑫𝟏 = 𝟏⁄𝟏𝟎 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

𝑫𝟐 = 𝟐⁄𝟏𝟎 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

𝑫𝟑 = 𝟑⁄𝟏𝟎 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

𝑫𝟗 = 𝟗⁄𝟏𝟎 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

The ith decile is given as:

𝑫𝒊 = 𝒊⁄𝟏𝟎 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

where i=1,2,3,…9

Note: 𝑛 = ∑ 𝑓 for a discrete series.

Illustration

1. Given the series, 6, 9, 3, 8, 7, 10 and 5. Find D4 and D6.

Solution

Re-arranging the data: 3, 5, 6, 7, 8, 9, 10

a. 𝐷4 = 4⁄10 (𝑛 + 1)𝑡ℎ term of the distribution

n = 7; 𝐷4 = 4⁄10 (7 + 1)𝑡ℎ term; 𝐷4 = 4⁄10 (8)𝑡ℎ term = 3.2nd item of the distribution
The fourth (D4) decile falls between the 3rd and 4th items of the distribution, that is, 6 and 7.
𝐷4 = 3rd value + 0.2 of the distance between the 3rd and 4th values.
𝐷4 = 6 + 0.2(6 − 7) = 6 + 0.2 = 6.2
b. 𝐷6 = 6⁄10 (𝑛 + 1)𝑡ℎ term of the distribution

n = 7; 𝐷6 = 6⁄10 (7 + 1)𝑡ℎ term; 𝐷6 = 6⁄10 (8)𝑡ℎ term = 4.8th item of the distribution

6
The sixth (D6) decile falls between the 4th and 5th items of the distribution, that is, 7 and 8.
𝐷6 = 4th value + 0.8 of the distance between the 7th and 8th values.
𝐷6 = 7 + 0.8(8 − 7) = 7 + 0.8 = 7.8
2. Obtain the third and the nineth deciles from the distribution below.

x 10 15 20 30 40 50
f 3 2 5 1 4 6

Solution

To obtain the deciles, we need the cumulative frequency.

X f CF
10 3 3
15 2 5
20 5 10
30 1 11
40 4 15
50 6 21
Total ∑ 𝒇 = 𝟐𝟏

a. 𝐷3 = 3⁄10 (𝑛 + 1)𝑡ℎ term of the distribution

𝑛 = ∑ 𝑓 = 21; 𝐷3 = 3⁄10 (21 + 1)𝑡ℎ term; 𝐷3 = 3⁄10 (22)𝑡ℎ term = 6.6th item of the
distribution.
𝐷3 = 20 (it falls within the third CF)
b. 𝐷9 = 9⁄10 (𝑛 + 1)𝑡ℎ term of the distribution

𝑛 = ∑ 𝑓 = 21; 𝐷9 = 9⁄10 (21 + 1)𝑡ℎ term; 𝐷9 = 9⁄10 (22)𝑡ℎ term = 19.8th item of the
distribution.
𝐷9 = 50 (it falls within the last CF)

Case B: Grouped data (Continuous Series)

To compute the deciles for a continuous series, the procedure is similar to the one used in calculating the
media. Generally, the deciles can be calculated as follows:

7
𝒊𝑵⁄ − 𝑪𝑭
𝑫𝒊 = 𝑳𝒄 + [ 𝟏𝟎 𝑫𝒊
]𝒄
𝑭𝑫𝒊

where “i” represents the deciles to be calculated, that is, i=1,2,3, …, 9

Lc is the lower-class boundary of the decile class

N is the sum of the frequency, that is, ∑ 𝑓

𝐶𝐹𝐷𝑖 is the cumulative frequency of the class before the decile class

𝐹𝐷𝑖 is the frequency of the decile class

c is the class size of the decile class

𝑖𝑁⁄ is used to determine the decile class


10

Illustration

Given the table below, compute D4 and D7.

Class Interval 1-10 11-20 21-30 31-40 41-50


Frequency 3 8 15 9 5

Solution

Class Interval f CF Class Boundary


1-10 3 3 0.5-10.5
11-20 8 11 10.5-20.5
21-30 15 26 20.5-30.5
31-40 9 35 30.5-40.5
41-50 5 5 40.5-50.5
Total ∑ 𝒇 = 𝟒𝟎

𝟒𝑵⁄ −𝑪𝑭
𝟏𝟎 𝑫𝟒
a. 𝑫𝟒 = 𝑳𝒄 + [ ]𝒄
𝑭𝑫𝟒

4 ∑ 𝑓⁄ 40
We first obtain the decile class, 4𝑁⁄10 = 10 = 4( ⁄10) = 16𝑡ℎ item of the distribution.

8
The fourth decile class, 4𝑁⁄10, is 21-30 (it falls within the third CF)

Lc = 20.5; 𝐶𝐹𝐷4 = 11; 𝐹𝐷4 = 15; 𝑐 = 30.5 − 20.5 = 10

16−11 5
𝐷4 = 20.5 + [ ] 10; 𝐷4 = 20.5 + [15] 10
15

𝐷4 = 20.5 + 3.33 = 23.83

𝟕𝑵⁄ −𝑪𝑭
𝟏𝟎 𝑫𝟕
b. 𝑫𝟕 = 𝑳𝒄 + [ ]𝒄
𝑭𝑫𝟕

7 ∑ 𝑓⁄ 40
We first obtain the decile class, 7𝑁⁄10 = 10 = 7( ⁄10) = 28𝑡ℎ item of the distribution.

The seventh decile class, 7𝑁⁄10, is 31-40 (it falls within the fourth CF)

Lc = 30.5; 𝐶𝐹𝐷7 = 26; 𝐹𝐷7 = 9; 𝑐 = 40.5 − 30.5 = 10

28−26 2
𝐷7 = 30.5 + [ ] 10; 𝐷4 = 30.5 + [9] 10
9

𝐷4 = 30.5 + 2.22 = 32.72

3. Percentiles

Percentiles are measures of partition that divide a distribution into 100 equal parts. A percentile is any of
the 99 values that divide the sorted data into 100 equal parts, so that each part represents 1/100 of the
distribution. P1, P2, P3,… P99 represent the first, second, third …and ninety nine percentiles. Note: the
lower quartile (Q1) which is ¼ is the 25th percentile (1/4 x 100), while the upper quartile (Q3) which is ¾
is the 75th percentile (3/4 x 100). The 50th percentile (50/100 = ½) represents the median of a distribution.

Case A: Ungrouped data (individual observations and Discrete Series)

For an ungrouped data, the first (P1), second (P2), third (P3) …. and ninety-nine (P99) percentiles can be
calculated as follows using a general formula:

The ith percentile is given as:

9
𝑷𝒊 = 𝒊⁄𝟏𝟎𝟎 (𝒏 + 𝟏)𝒕𝒉 term of the distribution

where i=1,2,3,…99

Note: 𝑛 = ∑ 𝑓 for a discrete series.

Illustration

1. Given the series, 6, 9, 3, 1, 11, 15, 21, 13, 14, 8, 7, 10 and 5. Find P30 and P80.

Solution

Re-arranging the data: 1, 3, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 21

a. 𝑃30 = 30⁄100 (𝑛 + 1)𝑡ℎ term of the distribution

n = 13; 𝑃30 = 30⁄100 (13 + 1)𝑡ℎ term; 𝑃30 = 30⁄100 (14)𝑡ℎ term = 4.2nd item of the
distribution
𝑃30 falls between the 4th and 5th items of the distribution, that is, 6 and 7.
𝑃30 = 4th value + 0.2 of the distance between the 4th and 5th values.
𝑃30 = 6 + 0.2(6 − 7) = 6 + 0.2 = 6.2
b. 𝑃80 = 80⁄100 (𝑛 + 1)𝑡ℎ term of the distribution

n = 13; 𝑃80 = 80⁄100 (13 + 1)𝑡ℎ term; 𝑃80 = 80⁄100 (14)𝑡ℎ term = 11.2nd item of the
distribution
𝑃80 falls between the 11th and 12th items of the distribution, that is, 14 and 15.
𝑃80 = 11th value + 0.2 of the distance between the 11th and 12th values.
𝑃80 = 14 + 0.2(15 − 14) = 14 + 0.2 = 14.2
2. Obtain P28 and P74 from the distribution below.

x 10 15 20 30 40 50
f 3 2 5 1 4 6

Solution

To obtain the deciles, we need the cumulative frequency.

10
X f CF
10 3 3
15 2 5
20 5 10
30 1 11
40 4 15
50 6 21
Total ∑ 𝒇 = 𝟐𝟏

a. 𝑃28 = 28⁄100 (𝑛 + 1)𝑡ℎ term of the distribution

𝑛 = ∑ 𝑓 = 21; 𝑃28 = 28⁄100 (21 + 1)𝑡ℎ term; 𝑃28 = 28⁄100 (22)𝑡ℎ term = 6.16th item of the
distribution.
𝑃28 = 20 (it falls within the third CF)
b. 𝑃74 = 74⁄100 (𝑛 + 1)𝑡ℎ term of the distribution

𝑛 = ∑ 𝑓 = 21; 𝑃74 = 74⁄100 (21 + 1)𝑡ℎ term; 𝑃74 = 74⁄100 (22)𝑡ℎ term = 16.28th item of the
distribution.
𝑃74 = 50 (it falls within the last CF)

Case B: Grouped data (Continuous Series)

To compute the percentiles for a continuous series, the procedure is similar to the one used in calculating
the media. Generally, the percentiles can be calculated as follows:

𝒊𝑵⁄
𝑷𝒊 = 𝑳 𝒄 + [ 𝟏𝟎𝟎 − 𝑪𝑭𝑷𝒊 ] 𝒄
𝑭𝑷 𝒊

where “i” represents the percentile to be calculated, that is, i=1,2,3, …, 99

Lc is the lower-class boundary of the percentile class

N is the sum of the frequency, that is, ∑ 𝑓

𝐶𝐹𝑃𝑖 is the cumulative frequency of the class before the percentile class

𝐹𝑃𝑖 is the frequency of the percentile class

11
c is the class size of the percentile class

𝑖𝑁⁄
100 is used to determine the percentile class

Illustration

Given the table below, compute P45 and P65.

Class Interval 1-10 11-20 21-30 31-40 41-50


Frequency 3 8 15 9 5

Solution

Class Interval f CF Class Boundary


1-10 3 3 0.5-10.5
11-20 8 11 10.5-20.5
21-30 15 26 20.5-30.5
31-40 9 35 30.5-40.5
41-50 5 5 40.5-50.5
Total ∑ 𝒇 = 𝟒𝟎

𝟒𝟓𝑵⁄
𝟏𝟎𝟎−𝑪𝑭𝑷𝟒𝟓
a. 𝑷𝟒𝟓 = 𝑳𝒄 + [ ]𝒄
𝑭𝑷𝟒𝟓

45 ∑ 𝑓⁄
We first obtain the percentile class, 45𝑁⁄100 = 40
100 = 45( ⁄100) = 18𝑡ℎ item of the
distribution.

The P45 class, 45𝑁⁄100, is 21-30 (it falls within the third CF)

Lc = 20.5; 𝐶𝐹𝑃45 = 11; 𝐹𝑃45 = 15; 𝑐 = 30.5 − 20.5 = 10

18−11 7
𝑃45 = 20.5 + [ ] 10; 𝑃45 = 20.5 + [15] 10
15

𝑃45 = 20.5 + 4.67 = 25.17

12
𝟔𝟓𝑵⁄
𝟏𝟎𝟎−𝑪𝑭𝑷𝟔𝟓
b. 𝑷𝟔𝟓 = 𝑳𝒄 + [ ]𝒄
𝑭𝑷𝟔𝟓

65 ∑ 𝑓⁄
We first obtain the percentile class, 65𝑁⁄100 = 40
100 = 65( ⁄100) = 26𝑡ℎ item of the
distribution.

The P65 class, 65𝑁⁄100, is 21-30 (it falls within the third CF)

Lc = 20.5; 𝐶𝐹𝑃65 = 11; 𝐹𝑃65 = 15; 𝑐 = 30.5 − 20.5 = 10

26−11 15
𝑃65 = 20.5 + [ ] 10; 𝑃65 = 21 + [15] 10
15

𝑃45 = 20.5 + 10 = 30.5

Activity

The set of observations below gives the grade on a subject for a class of 40 students.

Find: (i) mean, median and mode (ii) lower and upper quartile (ii) D3 and D8 (iii) P35, P65, and P85

7 5 6 2 8 7 6 7 3 9 10 4 5 5 4 6 7 4 8 2

3 5 6 7 9 8 2 4 7 9 4 6 7 8 3 6 7 9 10 5

13

You might also like