MODULE 5

STATISTICAL METHODS
AND PROBABILITY
 ISC XI- Applied Mathematics

5
(i) Measures of Dispersion

SCOPE

Measures of dispersion: range, Quartile Deviation, mean deviation, variance and standard deviation of
ungrouped/grouped data.
• Range
• Quartiles, Interquartile range, Quartile deviation, Coefficient of Quartile deviation
• Mean deviation about mean and median, coefficient of mean deviation
• Mean deviation about median.
• Standard deviation - by direct method, short cut method and step deviation method, Variance and
coefficient of variance
• Combined mean and standard deviation
• Mode of grouped and ungrouped data.
• Differentiate between range, quartile deviation, mean deviation and standard deviation.
• Choose appropriate measure of dispersion to calculate spread of data.

LEARNING OUTCOMES

Students will be able to:

• describe dispersion in data.
• define and calculate different measures of dispersion like range, mean deviation from a central value
and standard deviation.

PRE-REQUISITES

• Knowledge of collection, classification and tabulation of data.

• Concept of frequency distributions
• Measures of Central Tendency (Mean, median, mode)

1
 ISC XI- Applied Mathematics

Introduction
Statistics is about collecting, presenting, analyzing and interpreting, and data such that we can draw
meaningful conclusions from datasets.
The study of statistics can be categorized into two main types:
• Descriptive Statistics – Describing the main features of the dataset.
• Inferential Statistics – Making inferences from a data.

Descriptive Statistics
Descriptive statistics involves methods for summarizing and describing the main features of a dataset. These
methods include calculating measures of central tendency (mean, median, mode) and measures of variability
or dispersion (standard deviation, variance, range). Descriptive statistics are used to present data in a clear and
understandable way.

Dispersion
Suppose there are three series of 5 items each as follows:
Series A 40 40 40 40 40

Series B 38 39 40 41 42

Series C 6 21 40 57 76

2
 ISC XI- Applied Mathematics

In all the three series, the mean is 40. But in Series A all the items are identical, in series B the items are not
identical but are not scattered a lot. In contrast, in Series C the non-identical items are widely scattered.

The term “dispersion” refers to how scattered or dispersed a set of data is. The measure of dispersion is always
a non-negative real number that starts at zero when all the items in the data are the same and rise as the data
get more varied. These measures help determine the homogeneity or heterogeneity of the data distribution,
providing insight into how much the values differ from each other.

Measures of Dispersion
Measures of dispersion describe how spread out or varied the data in a dataset is.

Characteristics of a good Measure of Dispersion

A measure of dispersion should be rigidly defined.
• It should be based on all the observations of the series.
• It must be easy to calculate and understand.
• It should be amenable to further algebraic treatment and
inferences.
• It should not be affected much by the fluctuations of observations.
Types of Measures of Dispersion
Measures of dispersion can be categorized into:
• Absolute Measure of Dispersion: It gives the dispersion in the same units as the data (e.g., range, variance,
standard deviation).
• Relative Measure of Dispersion: It expresses dispersion as a ratio or percentage of the average, making
it unit-free and comparable across datasets.

3
 ISC XI- Applied Mathematics

Range
Range is the simplest possible measure of dispersion or the extent of scatter. It is mathematically expressed as
‘The difference between the largest and the smallest values in a data set’.

Range = Largest Value – Smallest Value

When we buy things, they are always sold within a price range. Take the example of your favourite pair
of jeans.
The store from where you made the purchase probably had a range of colours, a range of fits, a range of
sizes, and a range of prices.

A wide range indicates high variability meaning the data is spread out, and a small range specifies low
variability in the distribution.

High Range = High Variability

Low Range = Low Variability

4
 ISC XI- Applied Mathematics

Example 1.
(i) The profits of a company for the last 6 years are given below. Calculate the range.

Year 2019 2020 2021 2022 2023 2024

Profits (in lakhs) 40 30 80 100 120 90

Solution:
Range = largest value – smallest value = 120 – 30 = 90

(ii) Find the range of the given data distribution.:

Classes 115 - 125 125 - 135 135 - 145 145 - 155 155 - 165 165 - 175
Frequency 1 4 6 1 3 5

Solution:
Range = upper limit of highest class – lower limit of the lowest class = 175 – 115 = 60

Limitations of Range
• Range is greatly affected by fluctuations of sampling.
• It is not based on all the observations of the series.
• Only the values of the variables are taken into account and the frequencies are completely ignored.

Consider two sets of marks obtained by students out of 100:

• Set A: 10, 20, 30, 40, 100
• Set B: 40, 41, 42, 43, 100
Range of Set A = 100 – 10 = 90
Range of Set B = 100 – 40 = 60
Illustration of Limitations:
1. Affected by fluctuations of sampling: If the maximum value in Set A (100) were wrongly recorded
as 95, the range would change drastically from 90 to 85.
2. Not based on all observations: In both sets, the range depends only on the minimum and
maximum values; the middle values (20, 30, 40 in Set A or 41, 42, 43 in Set B) have no effect on
the result.
3. Frequencies ignored: Suppose in Set A, the value 40 occurred 20 times, the range would still
remain 90, ignoring how frequently a number appears.

5
 ISC XI- Applied Mathematics

Quartiles
Just as the median is a positional average that divides a data set into two equal parts, there are other positional
values that split a series into multiple sections. One of the most commonly used among them is quartiles.
Quartiles divide a data set into four equal parts, helping to analyze the spread and distribution of the data.
The three quartiles are:
1. First Quartile (Q₁) or Lower Quartile – This is the value that separates the lowest one-fourth (25%) of
the data from the rest.
2. Second Quartile (Q₂) or Median – This is the middle value, dividing the data set into two equal halves.
3. Third Quartile (Q₃) or Upper Quartile – This is the value that separates the lowest three-fourths (75%)
of the data from the highest one-fourth.
𝑁𝑁+1
For ungrouped data: 𝑄𝑄𝑖𝑖 = the value of the 𝑖𝑖 � � th item
4

For grouped data: Identify the 𝑄𝑄𝑖𝑖 class that has the
𝑁𝑁
cumulative frequency as 𝑖𝑖 � �. Name it as 𝑄𝑄𝑖𝑖 class
4

Then,
𝑁𝑁
𝑁𝑁 𝑖𝑖� �−𝑐𝑐𝑐𝑐
𝑄𝑄𝑖𝑖 = the value of the 𝑖𝑖 � � th item = 𝐿𝐿 + 4
× 𝐼𝐼
4 𝑓𝑓

where i = 1,2,3.
L = lower boundary of the 𝑄𝑄𝑖𝑖 class
cf = cumulative frequency of the class preceding the 𝑄𝑄𝑖𝑖
class
f = frequency of the 𝑄𝑄𝑖𝑖 class
I = size of class (assuming all classes have equal size)

Example 2. Calculate the lower and upper quartiles from the following data.

Weight (in lbs.) 4 5 6 8 11 13 14

No of babies 2 4 5 7 3 2 1

Solutions:
N = 24

Q1 = �
N+1 th
� item = �
24+1 th
� item x f c.f. less than
4 4
1th 4 2 2
=6 item
4
5 4 6
1
= 6th item +
4
�7𝑡𝑡ℎ − 6𝑡𝑡ℎ � item
6 5 11
1
= 5 + (6 − 5) = 5 ⋅ 25 𝑙𝑙𝑙𝑙𝑙𝑙 8 7 18
4
Q3 = 3 �
N+1 th
� item = 3 �
24+1 th
� item 11 3 21
4 4
3th 13 2 23
= 18 item
4
3
14 1 24
= 18th item +
4
�19𝑡𝑡ℎ − 18𝑡𝑡ℎ � item

6
 ISC XI- Applied Mathematics

3
= 8 + (11 − 8) = 10 ⋅ 25 𝑙𝑙𝑙𝑙𝑙𝑙
4
Example 3. Calculate the lower and upper quartiles from the following data.
CI 10 - 19 20 - 29 30 - 39 40 - 49 50 - 59 60 - 69 70 - 79 80 - 89
Frequency 5 9 14 20 25 15 8 4

Solution:

Cumulative
Class interval Exclusive CI Frequency (f) frequency
(c.f less than)
10 – 19 9.5 – 19.5 5 5

20 – 29 19.5 – 29.5 9 14

30 – 39 29.5 – 39.5 14 28

40 – 49 39.5 – 49.5 20 48

50 – 59 49.5 – 59.5 25 73

60 - 69 59.5 – 69.5 15 88

70 - 79 69.5 – 79.5 8 96

80 - 89 79.5 – 89.5 4 100

N = 100
𝑁𝑁 100 th
𝑸𝑸𝟏𝟏 =� �th item =� � item = 25th item
4 4

∴ 𝑄𝑄1 lies in the class = 29.5 – 39.5

𝑁𝑁
� �−𝑐𝑐𝑐𝑐 25−14
𝑄𝑄1 = 𝐿𝐿 + 4
𝑓𝑓
× 𝐼𝐼 = 29.5 + × 10 = 37.36
14

𝑁𝑁 100 th
𝑸𝑸𝟐𝟐 =2 � �th item =2 � � item = 50th item
4 4

∴ 𝑄𝑄2 lies in the class = 49.5 – 59.5

𝑁𝑁
2 � �−𝑐𝑐𝑐𝑐 50−48
𝑄𝑄2 = 𝐿𝐿 + 4
𝑓𝑓
× 𝐼𝐼 = 49.5 + × 10 = 50.3
25

𝑁𝑁 100 th
𝑸𝑸𝟑𝟑 =3 � �th item =3 � � item =75th item
4 4

∴ 𝑄𝑄3 lies in the class = 59.5 – 69.5

𝑁𝑁
3 � �−𝑐𝑐𝑐𝑐 75−73
𝑄𝑄3 = 𝐿𝐿 + 4
𝑓𝑓
× 𝐼𝐼 = 59.5 + × 10 = 60.83
15

7
 ISC XI- Applied Mathematics

Inter Quartile Range

The inter quartile range (IQR) is a measure of statistical dispersion, representing the spread of the middle 50%
of a dataset, calculated as the difference between the third quartile (Q3) and the first quartile (Q1).
Inter Quartile Range = 𝑸𝑸𝟑𝟑 − 𝑸𝑸𝟏𝟏
The IQR provides a robust measure of variability, less susceptible to outliers compared to the range.

Quartile Deviation
Quartile deviation is also called the Semi-inter quartile range and is calculated as
𝑸𝑸𝟑𝟑 −𝑸𝑸𝟏𝟏
Quartile deviation = 𝟐𝟐

The coefficient of quartile deviation, also known as the quartile coefficient of dispersion, is a statistical
measure that quantifies the spread or dispersion of a dataset, specifically focusing on the central 50% of the
data.
𝑸𝑸 −𝑸𝑸
Coefficient of Quartile Deviation = 𝑸𝑸𝟑𝟑+𝑸𝑸𝟏𝟏
𝟑𝟑 𝟏𝟏

It's a relative measure, making it suitable for comparing data sets with different scales or units. Since it relies
on quartiles (Q1 and Q3) rather than the mean and standard deviation, it's less sensitive to the influence of
extreme values (outliers) in the data.

Example 4. Find the quartile deviation for the following given data.
23, 8, 5, 16, 33, 7, 24, 5, 30, 33, 37, 30, 9, 11, 26, 32
Solution:
Let us arrange this data in the following ascending order.
5, 5, 7, 8, 9, 11, 16, 23, 24, 26, 30, 30, 32, 33, 33, 37
N =16
N+1 th 16+1 th 1th 1
Q1 = �
4
� item = �
4
� item = 4
4
item =4th item +
4
�5𝑡𝑡ℎ − 4𝑡𝑡ℎ � item

1
= 8 + (9 − 8) = 8 ⋅ 25
4
N+1 th 16+1 th 3th 3
Q3 = 3 �
4
� item = 3 �
4
� item = 12
4
item =12th item +
4
�13𝑡𝑡ℎ − 12𝑡𝑡ℎ � item

3
= 30 + (32 − 30) = 31.5
4
𝑄𝑄3 −𝑄𝑄1 31⋅5−8.25
Quartile Deviation = 2
= = 11.625
2

Example 5. Calculate Quartile deviation from the following and hence calculate the coefficient of quartile
deviation.
Class 0-2 2-4 4-6 6-8 8 - 10 10 - 12

Frequency 5 16 13 7 5 4

8
 ISC XI- Applied Mathematics

Solution:
Here, N =50 Class Frequency c.f. (less than)
𝑁𝑁 50
𝑄𝑄1 =� 4 �th item =� 4 �th item =12.5th item 0-2 5 5

2-4 16 21
∴ 𝑄𝑄1 lies in the class = 2 - 4
𝑁𝑁
4-6 13 34
� �−𝑐𝑐𝑐𝑐 12.5−5
𝑄𝑄1 = 𝐿𝐿 + 4
× 𝐼𝐼 = 2 + × 2 = 2.93 6-8 7 41
𝑓𝑓 16

𝑁𝑁 50
𝑄𝑄3 =3 � �th item =3 � �th item =37.5th item 8 - 10 5 46
4 4
10 - 12 4 50
∴ 𝑄𝑄3 lies in the class = 6 - 8
𝑛𝑛
3 � �−𝑐𝑐𝑐𝑐 37.5−34
𝑄𝑄3 = 𝐿𝐿 + 4
× 𝐼𝐼 = 6 + ×2=7
𝑓𝑓 7

𝑄𝑄3 −𝑄𝑄1 7−2.9375

Quartile Deviation = = = 2.031
2 2
𝑄𝑄3 −𝑄𝑄1 7−2.9375 4.0625
Coefficient of Quartile Deviation = = 7+2.9375 = 9.9375 = 0.408
𝑄𝑄3 +𝑄𝑄1

Scenario: An HR team is determining salary ranges for new hires with varying experience and
qualifications.
Application: They use quartile deviation to understand the spread of salaries within different experience
levels. A high quartile deviation might indicate a wide range of salaries for a particular role, while a low
deviation suggests a more uniform salary structure.
Interpretation:
High QD: Suggests a significant difference between the salaries of the most and least experienced
employees within a certain range, potentially indicating a wide range of skills and responsibilities.
Low QD: Indicates that salaries are more closely clustered around the median, suggesting a more
standardized compensation structure.

Limitations of Quartile Deviation

• Focus on the middle 50%: Quartile deviation only considers the range between Q1 and Q3, neglecting the
data below Q1 and above Q3.
• Ignores extreme values: It doesn't reflect the impact of outliers or data points which are significantly
different from the main body of data.
• Not suitable for distribution shape comparison: It provides limited information about the overall shape of
the distribution, making it less informative when comparing datasets with different shapes.
• Dependence on quartiles: The calculation of quartiles, and thus the quartile deviation, can be influenced
by sample size and data distribution.

9
 ISC XI- Applied Mathematics

Two sets of test scores (out of 100):

• Set A: 5, 10, 20, 40, 50, 60, 80, 90, 95
• Set B: 40, 41, 42, 43, 44, 45, 46, 47, 48
For both sets,
80 − 20
• Set A:𝑄𝑄1 ≈ 20, 𝑄𝑄3 ≈ 80, 𝑄𝑄. 𝐷𝐷. = = 30
2
46 − 42
• Set B: 𝑄𝑄1 ≈ 42, 𝑄𝑄3 ≈ 46, 𝑄𝑄. 𝐷𝐷. = = 2
2
Illustration of Limitations:
1. Focus on the middle 50%: In Set A, scores 5, 10, 90, and 95 are ignored completely, as Q.D. only
uses Q1 and Q3.
2. Ignores extreme values: Even though Set A has extreme outliers (5 and 95), the Q.D. = 30 does not
change if these extremes become more extreme (say 0 and 100).
3. Not suitable for distribution shape comparison: Q.D. of Set A (30) is much larger than Set B (2),
but it doesn’t explain that Set A is widely spread with outliers while Set B is tightly clustered. The
shape difference is not conveyed.
4. Dependence on quartiles: If Set A had fewer data points, say only {10, 50, 90}, quartiles would
shift drastically, making Q.D. unstable for small samples.

Mean Deviation About Mean

Mean Deviation about the Mean is a measure of dispersion that quantifies the average absolute deviation of
data points from the mean of the dataset. It helps in understanding how spread out the values are around the
central tendency (mean).

Why do we take absolute values?

Since the average is the central value of the data, some deviations might be positive and some might be
negative. If they are added like that, their sum will not reveal much as they tend to cancel each other’s
effect.
For example, if we consider the dataset: -10, 5, 20

−10+5+20
Mean = =5
3
Now a deviation from the mean for the different values is,
• (-10 -5) = -15
• (5 – 5) = 0
• (20 – 5) = 15
Now adding the deviations, we see that there is zero deviation from the mean which is incorrect. Thus, to
counter this problem only the absolute values of the difference are taken while calculating the mean
deviation.

10
 ISC XI- Applied Mathematics

The Mean Deviation (MD) about the mean (𝑥𝑥̅ ) is the mean of the absolute values of the deviations of the
items from the mean, and is given by
𝜮𝜮𝒇𝒇𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙
�|
𝑴𝑴𝑴𝑴 =
𝜮𝜮𝒇𝒇𝒊𝒊
The coefficient of mean deviation from the mean is a measure of dispersion that indicates the variability of
a dataset around its average. It’s calculated by dividing the mean deviation (average of absolute differences
from the mean) by the mean itself.
𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴 𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅𝒅
Coefficient of Mean Deviation from mean =
�
𝒙𝒙

This ratio helps compare the variability of different datasets, regardless of the scale of the data.

Mean Deviation about mean for ungrouped data

Example 6. Find the mean deviation about mean for 4,8,6,10,12.
Solution:
Consider the dataset: 4,8,6,10,12
4+8+6+10+12
1. Calculate the mean: 𝑥𝑥̅ = =8
5

2. Find the absolute deviations from the mean:

|4−8|=4
|8−8|=0
|6−8|=2
|10−8|=2
|12−8|=4
4+0+2+2+4
3. Compute the Mean Deviation about mean: 𝑀𝑀𝑀𝑀 = = 2.4
5

Example 7. Find the mean deviation about mean for the following and hence calculate the coefficient of
mean deviation.

Marks 5 15 25 35 45
No of students 5 8 15 16 6
Solution:

𝒙𝒙𝒊𝒊 𝒇𝒇𝒊𝒊 𝒇𝒇𝒊𝒊 𝒙𝒙𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙

�| 𝒇𝒇𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙
�|
5 5 25 22 110
15 8 120 12 96
25 15 375 2 30
35 16 560 8 128
45 6 270 18 108
50 1350 472

1350
1. Calculate the mean: 𝑥𝑥̅ = = 27
50
472
2. Compute the Mean Deviation about mean: 𝑀𝑀𝑀𝑀 = 50
= 9.44
9.44
3. Coefficient of Mean Deviation = = 0.3496
27

11
 ISC XI- Applied Mathematics

Mean Deviation about mean for grouped data

Example 8. The following table represents the scores of 50 students in a talent show. Find the mean
deviation about mean for the following:

Scores 0 –10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60 60-70

No of students 6 5 8 15 7 6 3

Solution:

CI 𝒇𝒇𝒊𝒊 𝒙𝒙𝒊𝒊 𝒇𝒇𝒊𝒊 𝒙𝒙𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙

�| 𝒇𝒇𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙
�|
0 – 10 6 5 30 28.4 170.4
10 – 20 5 15 75 18.4 92
20 – 30 8 25 200 8.4 67.2
30 – 40 15 35 525 1.6 24
40 – 50 7 45 315 11.6 81.2
50 – 60 6 55 330 21.6 129.6
60 – 70 3 65 195 31.6 94.8
50 1670 659.2

1670
1. Calculate the mean: 𝑥𝑥̅ = = 33.4
50
659.2
2. Compute the Mean Deviation about mean: 𝑀𝑀𝑀𝑀 = 50
= 13.18
13.18
3. Coefficient of Mean Deviation = = 0.3496
33.4

Example 9. Find the mean deviation about mean using step deviation method for the following:

Height (in cm) 95-105 105-115 115-125 125-135 135-145 145-155

No of students 9 13 26 30 12 10

Solution:

Height (in cm) No of students 𝒙𝒙𝒊𝒊 𝐱𝐱𝐢𝐢 − 𝟏𝟏𝟏𝟏𝟏𝟏 𝒇𝒇𝒊𝒊 𝒖𝒖𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙
�| 𝒇𝒇𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝒙𝒙
�|
𝐮𝐮𝐢𝐢 =
(𝒇𝒇𝒊𝒊 ) 𝟏𝟏𝟏𝟏
95-105 9 100 -2 -18 25.3 227.7
105-115 13 110 -1 -13 15.3 198.9
115-125 26 120 0 0 5.3 137.8
125-135 30 130 1 30 4.7 141
135-145 12 140 2 24 14.7 176.4
145-155 10 150 3 30 24.7 247
100 53 1128.8

1. Take assumed mean A as 120.

𝛴𝛴𝑓𝑓𝑖𝑖 𝑢𝑢𝑖𝑖 53
2. Calculate the mean: 𝑥𝑥̅ = 𝐴𝐴 + × 𝑐𝑐 = 120 + 100 × 10 = 125.3
𝛴𝛴𝑓𝑓𝑖𝑖
1128.8
3. Compute the Mean Deviation about mean: 𝑀𝑀𝑀𝑀 = = 11.29
100

12
 ISC XI- Applied Mathematics

Median
Median is the middle item of a distribution, when items are arranged in order of magnitude. Unlike mean, the
median does not take into account the values of all items in a series.
𝑛𝑛+1 th
For ungrouped data: Median = the value of the � � item
2

For grouped data: When data observations are grouped into

intervals, arrange them in to ascending or descending order.

Find the class interval whose cumulative frequency is equal to or

just more than but closest to the half of the sum of frequencies i.e.
N/2. This class is termed as the median class. Then find the
following from the data corresponding to the median class.
L = lower boundary of the median class
cf = cumulative frequency of the class preceding the median
class
f = frequency of the median class
I = size of class (assuming that all classes have equal size.)
N = total number of items
Then,
𝑁𝑁
𝑁𝑁 −𝑐𝑐𝑐𝑐
Median = the value of the � 2 �th item = 𝐿𝐿 + 2
𝑓𝑓
× 𝐼𝐼

Median for ungrouped data

Example 10. Find the median of
(i) 5, 7, 9, 12, 10, 8, 7, 15, 21
(ii) 94, 33, 86, 68, 32, 80, 48, 70
Solution:
(i) We first arrange in ascending order 5, 7, 7, 8, 9, 10, 12, 15, 21
𝑛𝑛+1 th 9+1 th
Median = � � item = � � item = 5th item = 9
2 2

(ii) We first arrange in ascending order 32, 33, 48, 68, 70, 80, 86, 94
𝑛𝑛+1 th 8+1 th
Median = � � item = � � item = 4.5th item
2 2
1 1
= 2 [4𝑡𝑡ℎ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 + 5𝑡𝑡ℎ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡] = 2 [68 + 70]= 69.

Median for grouped data

Example 11. Find the median of the following frequency distribution.

Weight (in kg) 36 – 40 41 – 45 46 – 50 51 – 55 56 – 60 61 – 65 66 – 70

No of persons 14 26 40 53 50 37 25

13
 ISC XI- Applied Mathematics

Solution:

Class Interval Exclusive CI Frequency Cumulative frequency

36 – 40 35.5 – 40.5 14 14
41 – 45 40.5 – 45.5 26 40
46 – 50 45.5 – 50.5 40 80
51 – 55 50.5– 55.5 53 133
56 – 60 55.5– 60.5 50 183
61 – 65 60.5– 65.5 37 220
66 – 70 65.5– 70.5 25 245
245
N = 245
𝑁𝑁 245 th
Median = � �th item =� � item =122.5th item
2 2

∴ Median class = 50.5 – 55.5

𝑁𝑁
−𝑐𝑐𝑐𝑐 122.5−80
Median = 𝐿𝐿 + 2
× 𝐼𝐼 = 50.5 + × 5 = 54.51 𝑘𝑘𝑘𝑘
𝑓𝑓 53

Missing frequencies when Median is given

Example 12. If the median of the following frequency distribution is 28.5, find the sum of the missing
frequencies.

Class interval 0 -10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60 Total

Frequency 5 a 20 15 b 5 60

Solution:

Class interval Frequency (f) Cumulative frequency (cf)

0 – 10 5 5
10 – 20 a 5+a
20 – 30 20 25 + a
30 – 40 15 40 + a
40 – 50 b 40 + a + b
50 – 60 5 45 + a + b
60
Since, the median is 28.5, then the median class is 20 – 30.

14
 ISC XI- Applied Mathematics

𝑁𝑁
−𝑐𝑐𝑐𝑐
∴ Median = 𝐿𝐿 + 2
× 𝐼𝐼
𝑓𝑓
30 –(5+𝑎𝑎)
⇒ 28.5 = 20 + × 10
20
30 –(5+𝑎𝑎)
⇒ 28.5-20 = 20
× 10
25−𝑎𝑎
⇒ 8.5 = 2

⇒ 17 = 25 – 𝑎𝑎
⇒ 𝑎𝑎 = 8.

Given N = 60
⇒ 45 + 𝑎𝑎 + 𝑏𝑏 =60
⇒ 45 + 8 + b = 60
⇒ 53 + b = 60
⇒ b =7

Hence, the missing frequencies are 𝑎𝑎 = 8 and b = 7.

∴ Sum of missing frequencies is 15.

Mean Deviation About Median

Mean deviation about the median measures the average distance of data points from the median, providing a
measure of data dispersion, where each deviation is calculated as the absolute difference between the data
point and the median.
� ) is given by
The Mean Deviation (MD) about the median (𝑀𝑀

�|
𝛴𝛴𝑓𝑓𝑖𝑖 |𝑥𝑥𝑖𝑖 −𝑀𝑀
𝑀𝑀𝑀𝑀 = , where n is the number of observations.
𝑛𝑛

The coefficient of mean deviation from the median is a statistical measure that indicates the average degree
of dispersion or variability in a dataset, relative to the median. It’s calculated by dividing the mean deviation
from the median by the median itself.

𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑
Coefficient of Mean Deviation from median = �
𝑀𝑀

This helps in comparing the variability of different datasets, regardless of their central tendency.

Mean Deviation about median for ungrouped data

Example 13. Find the mean deviation about median for the following:

Marks 5 7 9 10 12 15
No of students 8 6 2 2 2 6

15
 ISC XI- Applied Mathematics

Solution:

𝒙𝒙𝒊𝒊 𝒇𝒇𝒊𝒊 𝒄𝒄𝒄𝒄 |𝒙𝒙𝒊𝒊 − 𝑴𝑴| 𝒇𝒇𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝑴𝑴|

5 8 8 2 16
7 6 14 0 0
9 2 16 2 4
10 2 18 3 6
12 2 20 5 10
15 6 26 8 48
26 84

1. N = 26
𝑁𝑁+1 26+1
2. Calculate the median: Median = � � th item = � � th item = 13.5 item
2 2

1 1
= [13𝑡𝑡ℎ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 + 14𝑡𝑡ℎ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡] = [7 + 7]= 7.
2 2

84
3. Compute the Mean Deviation about Median: 𝑀𝑀𝑀𝑀 = = 3.23
26

Mean Deviation about median for grouped data

Example 14. The following table represents the scores of 50 students in a talent show. Find the mean deviation
about median for the following and hence calculate the coefficient of mean deviation about the median.

Scores 0 -10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60
No of students 8 10 10 16 4 2

Solution:

Class f x cf |𝒙𝒙𝒊𝒊 − 𝑴𝑴| 𝒇𝒇𝒊𝒊 |𝒙𝒙𝒊𝒊 − 𝑴𝑴|

0 – 10 8 5 8 22 176
10 – 20 10 15 18 12 120
20 – 30 10 25 28 2 20
30 – 40 16 35 44 8 128
40 – 50 4 45 48 18 72
50 – 60 2 55 50 28 56
50 572

1. N = 50
𝑁𝑁 50
2. Median = � �th item =� �th item =25th item
2 2
∴ Median class = 20 – 30
𝑁𝑁
−𝑐𝑐𝑐𝑐
25−18
3. Calculate the median: Median = 𝐿𝐿 + 2 𝑓𝑓 × 𝐼𝐼 = 20 + × 10 = 27
10
572
4. Compute the Mean Deviation about Median: 𝑀𝑀𝑀𝑀 = = 11.44
50
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 11.44
5. Coefficient of Mean Deviation from median = �
= = 0.4237
𝑀𝑀 27

16
 ISC XI- Applied Mathematics

Limitations of Mean Deviation

• Mean deviation ignores the algebraic signs of deviations and is therefore not capable of further
mathematical treatment.

In cricket, calculating the mean deviation of a batsman's average score across a

series of matches helps determine how consistently they perform.
A lower mean deviation indicates more consistency, while a higher deviation
suggests more variability in their scoring.

Variance
Recall that the absolute values were taken while calculating mean deviation, otherwise the deviations may
cancel among themselves. Another way to overcome this difficulty due to signs of deviation is to take squares
of all deviations. The mean of squared deviations is called variance. It is denoted as σ2. This leads to a
limitation for variance as its units are different from the units of the variable ‘x’ making it difficult to compare
datasets that use different units of measurement.
Therefore, it leads us to define standard deviation which is also called the root mean squared deviation and is
considered the square root of variance.

Standard Deviation
Standard deviation is a key statistical measure that quantifies the amount of variation or dispersion in a set of
data values. It tells us how much individual data points deviate from the mean (average) of the dataset.
Standard Deviation is the most reliable measure of dispersion because it is based on all observations in the
dataset and allows further mathematical treatment due to squaring of deviations.
A low standard deviation indicates that the data points are closely clustered around the mean, whereas a high
standard deviation suggests that the data points are spread out over a wider range.
For example, there are five people of different heights
standing against a wall. The average (mean) height of
these five individuals is 106.5 units. The standard
deviation, which measures how much individual It is Root Mean Square Deviation from Mean.
heights deviate from the mean, is 32.8. Since this value As per the name take actions in the reverse order
is relatively high, it indicates that some people in the i.e.
group are significantly taller or shorter than the
average height. 1. Take deviations of all the entries from mean.
2. Take square of the deviations.
Standard Deviation (denoted as σ) is defined as the
3. Get the mean of the squared deviations.
positive square root of the arithmetic mean of the
squares of the deviations of all the observations from 4. Finally take positive square root of this mean.
their arithmetic mean.
1
For ungrouped data: 𝜎𝜎 = � ∑𝑛𝑛
𝑖𝑖=1(𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )
2
𝑛𝑛

1 𝑛𝑛
For grouped data: 𝜎𝜎 = � ∑𝑛𝑛
𝑖𝑖=1 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ ) where 𝑁𝑁 = ∑𝑖𝑖=1 𝑓𝑓𝑖𝑖
2
𝑁𝑁

17
 ISC XI- Applied Mathematics

Mathematically, the standard deviation is calculated using the working formula:

𝛴𝛴𝑥𝑥 𝛴𝛴𝛴𝛴 2 2
For ungrouped data: 𝜎𝜎 = � − � � , where n is the number of observations.
𝑛𝑛 𝑛𝑛

For grouped data:

𝛴𝛴𝑓𝑓𝑓𝑓 2 𝛴𝛴𝛴𝛴𝛴𝛴 2
i. Direct method 𝜎𝜎 = � 𝑁𝑁
−� �
𝑁𝑁

𝛴𝛴𝑓𝑓𝑓𝑓2 𝛴𝛴𝛴𝛴𝛴𝛴 2
ii. Short-cut method 𝜎𝜎 = � −� � , where 𝑢𝑢 = 𝑥𝑥 − 𝐴𝐴.
𝑁𝑁 𝑁𝑁
𝛴𝛴𝑓𝑓𝑓𝑓2 𝛴𝛴𝛴𝛴𝛴𝛴 2 𝑥𝑥−𝐴𝐴
iii. Step deviation method 𝜎𝜎 = � 𝑁𝑁
−� 𝑁𝑁
� × 𝐼𝐼, where 𝑑𝑑 = 𝐼𝐼
.

The coefficient of variation (CV) is the standardised measure of the dispersion of data points around the
mean. It is also called relative standard deviation (RSD). It’s calculated as the ratio of the standard deviation
to the mean and is often expressed as a percentage, making it easy to compare the variability of datasets with
different units or scales.
𝜎𝜎
𝐶𝐶𝐶𝐶 = × 100
𝑥𝑥̅

A lower CV indicates less variability and higher stability, while a higher CV suggests greater variability.

Standard Deviation for ungrouped data

Example 15. Find the standard deviation for the first 10 natural numbers.
Solution:

x x2
1 1
2 4
3 9
4 16
5 25
6 36
7 49
8 64
9 81
10 100
55 385

𝛴𝛴𝑥𝑥 2 𝛴𝛴𝛴𝛴 2 385 55 2

𝜎𝜎 = � −� � = � − � � = 2.87
𝑛𝑛 𝑛𝑛 10 10

Standard Deviation for grouped data

Example 16. Find the standard deviation using direct method for the following:

x 10 20 30 40 50 60
f 9 18 25 27 14 7

18
 ISC XI- Applied Mathematics

Solution:

x f fx fx2
10 9 90 900
20 18 360 7200
30 25 750 22500
40 27 1080 43200
50 14 700 35000
60 7 420 25200
100 3400 134000

𝛴𝛴𝑓𝑓𝑓𝑓 2 𝛴𝛴𝛴𝛴𝛴𝛴 2 134000 3400 2

𝜎𝜎 = � −� � =� −� � = 13.56
𝑁𝑁 𝑁𝑁 100 100

Example 17. Find the standard deviation using the short-cut method for the following:

No of articles produced 18 19 20 21 22 23 24 25 26 27
No of workers 3 7 11 14 18 17 13 8 5 4
Solution:

x f 𝒖𝒖 = 𝒙𝒙 − 𝑨𝑨 fu fu2
18 3 -4 -12 48
19 7 -3 -21 63
20 11 -2 -22 44
21 14 -1 -14 14
22 18 0 0 0
23 17 1 17 17
24 13 2 26 52
25 8 3 24 72
26 5 4 20 80
27 4 5 20 100
100 38 490

Assumed mean A = 22

𝛴𝛴𝑓𝑓𝑓𝑓2 𝛴𝛴𝛴𝛴𝛴𝛴 2 490 38 2

𝜎𝜎 = � −� � =� −� � = 2.18
𝑁𝑁 𝑁𝑁 100 100

Example 18. Find the standard deviation using the step deviation method for the following:

Age 20-25 25-30 30-35 35-40 40-45 45-50 50-55

No of employees 170 110 80 45 40 30 25

19
 ISC XI- Applied Mathematics

Solution:

CI x f 𝐱𝐱 − 𝐀𝐀 fd fd2
𝐝𝐝 =
𝐈𝐈

20-25 22.5 170 -3 -510 1530

25-30 27.5 110 -2 -220 440
30-35 32.5 80 -1 -80 80
35-40 37.5 45 0 0 0
40-45 42.5 40 1 40 40
45-50 47.5 30 2 60 120
50-55 52.5 25 3 75 225
500 -635 2435

Assumed mean A = 37.5

𝛴𝛴𝑓𝑓𝑓𝑓2 𝛴𝛴𝛴𝛴𝛴𝛴 2 2435

𝜎𝜎 = � −� � × 𝐼𝐼 = � – × 5 = 9.02
𝑁𝑁 𝑁𝑁 500

Example 19. The mean and standard deviation of 25 observations is 60 and 3 respectively. Later, it was
decided to omit an observation which was incorrectly recorded as 50. Calculate the mean and standard
deviation of the remaining 24 observations.
Solution:
∑𝑥𝑥
𝑥𝑥̅ = 60 ⇒ = 60 ⇒ ∑𝑥𝑥 = 25 × 60 = 1500
25
∑𝑥𝑥 2 ∑𝑥𝑥 2 ∑𝑥𝑥 2
𝜎𝜎 = 3 ⇒ −� � =9⇒ − (60)2 = 9
𝑛𝑛 𝑛𝑛 25
∑𝑥𝑥 2
= 3609 ⇒ ∑𝑥𝑥 2 = 90225
25

Now as we omit one item 50, new 𝛴𝛴𝛴𝛴 = 1500 − 50 = 1450.

new ∑𝑥𝑥 2 = 90225 − (50)2 = 87725

1450
∴ New mean = = 60 ⋅ 42
24
87725
New Standard deviation = � − (60 ⋅ 42)2 = 2 ⋅ 15
24

20
 ISC XI- Applied Mathematics

Difference between Mean Deviation and Standard Deviation

Mean Deviation Standard Deviation

Mean Deviation does not take into account the Standard Deviation takes into account the algebraic
algebraic signs in its calculations. signs in its calculations.
Mean Deviation can be computed from Mean, Standard Deviation is always computed from the
Median or Mode. Mean.
It is less frequently used. It is one of the most commonly used measures of
variability and frequently used.
When there are a greater number of outliers in the When there are fewer outliers in the data, the
data, mean absolute deviation is used. standard deviation is employed.

Difference between Standard Deviation and Coefficient of Variation

Feature Standard Deviation (SD) Coefficient of Variation (CV)

Measures the spread of data around the Measures the ratio of standard deviation to
Definition
mean. the mean, expressed as a percentage.
Describes the dispersion of data within a Compares the variability of datasets with
Purpose
single dataset. different means or units.
Units Same units as the data. Unitless (expressed as a percentage).
Can be difficult to interpret when More easily interpretable for comparing
Interpretability
comparing datasets with different means. variability across datasets.

Standard deviation can be used to compare the variability of temperatures between two cities. A city
with a higher standard deviation in daily temperatures will have more extreme temperature swings than
a city with a lower standard deviation.
Imagine comparing the average daily maximum temperatures for Kolkata (a coastal city) and Hyderabad
(an inland city). Kolkata might have a lower standard deviation, meaning the daily temperatures are
more consistent, while Hyderabad might have a higher standard deviation, indicating more significant
temperature variations.

Limitations of Standard Deviation

Standard Deviation gives more weight to extreme items and less to those which are near the mean, because
the squares of the deviations, which are larger, would be proportionately greater than the squares of those
deviations which are comparatively smaller.

21
 ISC XI- Applied Mathematics

Combined Mean and Standard Deviation

If there are two groups having the number of observations 𝑛𝑛1 and 𝑛𝑛2 , arithmetic mean ���
𝑥𝑥1 and ���
𝑥𝑥2 and standard
deviations 𝜎𝜎1 and 𝜎𝜎2 respectively, then the combined mean and standard deviation for this group is given as:
𝑛𝑛1 𝑥𝑥1 + 𝑛𝑛2 𝑥𝑥2
𝑥𝑥̅ =
𝑛𝑛1 + 𝑛𝑛2

𝑛𝑛1 �𝜎𝜎12 +𝑑𝑑12 �+𝑛𝑛2 �𝜎𝜎22 +𝑑𝑑22 �

𝜎𝜎 = � where, 𝑑𝑑1 = 𝑥𝑥̅1 − 𝑥𝑥̅ , 𝑑𝑑2 = 𝑥𝑥̅2 − 𝑥𝑥̅
𝑛𝑛1 +𝑛𝑛2

Example 20. A company has two production units. The details of the number of workers and their average
monthly wages (in ₹) are given below:

Production Unit Number of Workers Mean Salary (₹) Standard Deviation (₹)
A 40 25,000 3,000
B 60 30,000 4,000
(a) Find the combined mean salary of all the workers.
(b) Find the combined standard deviation of the salaries.
Solution:
𝑛𝑛1 = 40, 𝑛𝑛2 = 60
𝑥𝑥̅1 = 25,000, 𝑥𝑥̅2 = 30,000
𝜎𝜎1 = 3000 𝜎𝜎2 = 4000

Combined mean
𝑛𝑛1 𝑥𝑥1 + 𝑛𝑛2 𝑥𝑥2 40 × 25000 + 60 × 30000 2800000
𝑥𝑥̅ = = = = ₹ 28,000
𝑛𝑛1 + 𝑛𝑛2 40 + 60 100

Now 𝑑𝑑1 = 25000 − 28000 = −3000, 𝑑𝑑2 = 30000 − 28000 = 2000

Combined standard deviation

𝑛𝑛1 (𝜎𝜎12 + 𝑑𝑑12 ) + 𝑛𝑛2 (𝜎𝜎22 + 𝑑𝑑22 )
𝜎𝜎 = �
𝑛𝑛1 + 𝑛𝑛2

40(9000000 + 9000000) + 60(16000000 + 4000000) 1920000000

=� = �
40 + 60 100

= 4381.78

Example 21. The following table gives the result of three centres of a public examination.

Centre No of examinations Mean score Standard Deviation

A 200 25 3
B 250 10 4
C 300 15 5

Calculate the mean and standard deviation of the combined scores of the three centres.

22
 ISC XI- Applied Mathematics

Solution:

𝑛𝑛1 = 200, 𝑛𝑛2 = 250, 𝑛𝑛3 = 300

𝑥𝑥̅1 = 25, 𝑥𝑥̅2 = 10, 𝑥𝑥̅3 = 15
𝜎𝜎1 = 3, 𝜎𝜎2 = 4, 𝜎𝜎3 = 5
Combined mean
𝑛𝑛1 𝑥𝑥1 + 𝑛𝑛2 𝑥𝑥2 + 𝑛𝑛3 𝑥𝑥3 200 × 25 + 250 × 10 + 300 × 15 12000
𝑥𝑥̅ = = = = 16
𝑛𝑛1 + 𝑛𝑛2 + 𝑛𝑛3 200 + 250 + 300 750

Now 𝑑𝑑1 = 25 − 16 = 9, 𝑑𝑑2 = 10 − 16 = −6, 𝑑𝑑3 = 15 − 16 = −1

Combined standard deviation

𝑛𝑛1 (𝜎𝜎12 + 𝑑𝑑12 ) + 𝑛𝑛2 (𝜎𝜎22 + 𝑑𝑑22 ) + 𝑛𝑛3 (𝜎𝜎32 + 𝑑𝑑32 ) 200(9 + 81) + 250(16 + 36) + 300(25 + 1)
𝜎𝜎 = � =�
𝑛𝑛1 + 𝑛𝑛2 + 𝑛𝑛3 200 + 250 + 300

38800
= � = 7.19
750

Mode
Mode is the value of the item which occurs the maximum number of times i.e. which has the highest frequency.
For ungrouped data: the mode can be obtained by simple observation.
For grouped data: Identify the class interval having the maximum frequency. Call it as Modal class. From
the data set find the following corresponding to the modal class.
L = lower boundary of the modal class
𝑓𝑓𝑚𝑚−1 = frequency of the class preceding the modal class
𝑓𝑓𝑚𝑚 = frequency of the modal class
𝑓𝑓𝑚𝑚+1 = frequency of the class following the modal class
I = size of class (assuming that all classes have equal size).
N = total number of items
𝑓𝑓𝑚𝑚 −𝑓𝑓𝑚𝑚−1
Then, Mode = 𝐿𝐿 + × 𝐼𝐼
2𝑓𝑓𝑚𝑚 −𝑓𝑓𝑚𝑚−1 −𝑓𝑓𝑚𝑚+1
NOTE:

• If a data set has only 1 value that occurs most often, the set is called unimodal.
• If a data set has 2 values that occur with the greatest frequency, the set is called bimodal.
• If a set of data has more than 2 values that occur with the same greatest frequency, the set is called
multimodal.

Example 22. Calculate the mode of the following:

(i) 3, 4, 5, 9, 6, 4, 7, 8, 4
(ii) 3, 4, 5, 9, 6, 4, 7, 8, 3
(iii) 3, 4, 5, 9, 6, 1, 7, 8, 2

23
 ISC XI- Applied Mathematics

Solution:
(i) 4 occurs three times – which is the maximum number of occurrences.
∴ Mode = 4
(ii) Both 3 and 4 occurs twice – which is the maximum number of occurrences.
∴ Mode = 3 and 4
(iii) All numbers occur the same number of times. So, there is no mode.

Example 23. The following table represents the number of minutes that students spend studying for a math
test:

Studying time (in minutes) 1-10 11- 20 21- 30 31- 40 41- 50

Number of students 2 10 6 4 3

What is the mode of the amount of time spent studying for the math test?

Solution: Changing the given data set into continuous class intervals as shown below:

Studying time (in minutes) Frequency

0.5 - 10.5 2
10.5 - 20.5 10
20.5 - 30.5 6
30.5 - 40.5 4
40.5 - 50.5 3

Since, the highest frequency is 10, therefore the modal class is 10.5 - 20.5.

𝑓𝑓𝑚𝑚 −𝑓𝑓𝑚𝑚−1 10−2 8

Mode = 𝐿𝐿 + × 𝐼𝐼 = 20.5 + 2(10)−2−6 × 10 = 20.5 + 12 × 10
2𝑓𝑓𝑚𝑚 −𝑓𝑓𝑚𝑚−1 −𝑓𝑓𝑚𝑚+1

= 17.17 minutes

Missing frequencies when Mode is given

Example 24. The modal marks for 94 students are 54 in a test marked out of 100. If ten students got marks
between 0 and 20, thirty students got marks between 40 and 60 and fourteen students got marks between 80
and 100, find how many students scored between 20-40 and 60-80?

Solution:

Marks No of students
0 - 20 10

20 – 40 x
40 -–60 30

60 – 80 y

80 -–100 14

24
 ISC XI- Applied Mathematics

Since, the mode is 54, therefore the modal class is 40 - 60.

𝑓𝑓𝑚𝑚 −𝑓𝑓𝑚𝑚−1
Mode = 𝐿𝐿 + 2𝑓𝑓 × 𝐼𝐼
𝑚𝑚 −𝑓𝑓𝑚𝑚−1 −𝑓𝑓𝑚𝑚+1

30−𝑥𝑥
⇒ 54 = 40 + × 20
2(30)−𝑥𝑥−𝑦𝑦
30−𝑥𝑥
⇒ 54 - 40 = × 20
60−𝑥𝑥−𝑦𝑦

⇒ 14(60 − 𝑥𝑥 − 𝑦𝑦) = 20(30 − 𝑥𝑥)

⇒ 7y – 3x = 120…………………………………….(i)
Given n = 94
⇒ 10 + x + 30 + y + 14 = 94
⇒ x + y = 40………………………………………..(ii)
Solving (i) and (ii) we get, x = 16 and y = 24.
So, 16 students scored between 20 and 40 and 24 students scored between 60 and 80.

Case Study based Question

Example 25. A business organisation is planning to rent a venue for a series of monthly events and has
shortlisted two options:
• Venue A: Mean monthly rental charge = ₹80,000; Standard Deviation = ₹8,000
• Venue B: Mean monthly rental charge = ₹60,000; Standard Deviation = ₹12,000
Find which option is better investment in terms of rental charges?
Solution: We can decide the better option by finding the coefficient of variation (CV).

For Venue A:
𝜎𝜎 8000
𝐶𝐶𝐶𝐶 = × 100 = × 100 = 10%
𝑥𝑥̅ 80000

For Venue B:
𝜎𝜎 12000
𝐶𝐶𝐶𝐶 = × 100 = × 100 = 20%
𝑥𝑥̅ 60000

Since Venue A has a lower CV, its rental charges are more consistent, making it a better investment from a
stability standpoint.

Choice of a Measure of Dispersion

The choice of a suitable measure of dispersion depends on the type of data, the level of accuracy required, and
the nature of the statistical analysis. Below is a comparison of Range, Quartile Deviation, Mean Deviation,
and Standard Deviation along with when to use each of them:

25
 ISC XI- Applied Mathematics

Measure of Formula Best Used Advantages Disadvantages

Dispersion When
Range (R) Range Quick Simple and easy Affected by
= largest value – smallest comparison of to calculate extreme values
value variability in (outliers)
small datasets
Quartile Quartile Deviation =
𝑄𝑄3 −𝑄𝑄1 Data has Not affected by Ignores extreme
2
Deviation outliers. extreme values values and uses
(QD) only middle 50%
of data
Mean �|
𝛴𝛴𝑓𝑓𝑖𝑖 |𝑥𝑥𝑖𝑖 − 𝑀𝑀 When Simpler than Absolute
Deviation 𝑀𝑀𝑀𝑀 = absolute standard deviations are used
𝑛𝑛
(MD) deviations are deviation; less
needed. affected by
extreme values.
Standard 𝑛𝑛
When precise Most reliable and Sensitive to
Deviation (SD) 1 dispersion widely used outliers
𝜎𝜎 = � � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )2 measurement measure
𝑁𝑁
𝑖𝑖=1 is needed.

Interdisciplinary Aspects of Measures of Dispersion

• Mathematics & Statistics: Core tools in data analysis and probability theory.
• Economics & Finance: Risk assessment and economic modelling.
• Engineering & Manufacturing: Process optimization and quality assurance.
• Social Sciences & Psychology: Behavioural studies and survey analysis.
• Environmental Science: Measuring pollution level variations and climate change effects.
• Medicine & Public Health: Assessing variability in disease outbreaks and treatment outcomes.

REAL- LIFE APPLICATIONS

Real-World Applications of Measures of Dispersion

• Economics & Business: Analysing income inequality, stock market risk, and price stability in
markets.
• Quality Control & Manufacturing: Ensuring product consistency by monitoring variability in
production processes.
• Education & Psychology: Assessing student performance consistency using standard deviation of
test scores.
• Healthcare & Medicine: Evaluating the effectiveness of treatments by analysing the variation in
patient responses.
• Weather & Climate Science: Studying temperature variations over time to understand climate
change patterns.
• Sports & Performance Analysis: Comparing player performances by analysing score variability.
Today, measures of dispersion like range, variance, standard deviation, and interquartile range play a
crucial role in various scientific, economic, and industrial applications.

26
 ISC XI- Applied Mathematics

Exercises
1. Choose the correct option:
(i) The mean of 100 observations is 50 and their standard deviation is 10. If 5 is added to each
observation, then new mean and new standard deviation respectively will be
(a) 50, 10 (c) 55, 10
(b) 55, 15 (d) 60, 10

(ii) The mean of 5 observations is 3 and their variance is 2. If three of the observations are 1, 3 and 5
then the other two observations are
(a) 2 and 4 (c) 7 and 2
(b) 2 and 6 (d) 7 and 9

(iii) The median of a set of 9 distinct observations is 20.5. If each of the largest 4 observations of the set
is increased by 2, then the median of the new set
(a) is increased by 2
(b) is decreased by 2
(c) is two times the original median
(d) remains the same as that of the original set

(iv) Consider the following statements:

Statement 1: Positional measure of dispersion describe the position that a particular data value has
within a data set.
Statement 2: Quartiles are positional measure of dispersion.
Which of the statement(s) given above is/are true?
(a) Only 1 (c) Both 1 and 2
(b) Only 2 (d) Neither 1 nor 2

(v) Assertion (A): If the standard deviation of a dataset is zero, then all the values in the dataset are
equal.
Reason (R): Standard deviation is the square root of the arithmetic mean of the squares of deviations
from the actual mean.
(a) Both A and R individually true and R is the correct explanation of A.
(b) Both A and R individually true but R is not the correct explanation of A.
(c) A is true but R is false.
(d) A is false but R is true.

2. Given below are the prices of shares of a company for the last 10 days.
172,164,188,214,190,237,200,195,208,230
Find the quartile deviation.

27
 ISC XI- Applied Mathematics

3. Calculate the range, quartile deviation and coefficient of quartile deviation of wages.
Wages (Rs) 30 - 32 32 - 34 34 - 36 36 - 38 38 - 40 40 - 42 42 - 44
No of members 12 18 16 14 12 8 6

4. For a group of 50 male workers the mean and S.D of their daily wages are Rs 630 and Rs 90 respectively
for a group of 40 female workers these are Rs 540 and Rs 60 respectively. Find the mean and S.D of these
90 workers.

5. The marks of 10 students in a particular examination are as follows:

10, 12, 15, 12, 16, 20, 13, 17, 15, 10
Later, it was decided that 5 bonus marks will be awarded to each student.
i. Calculate the variance and standard deviation in the two cases.
ii. Draw an inference from the values calculated.

6. The distribution of weight of 100 students is given below:

Weight (in Kg) 50 −55 55 −60 60 −65 65 −70 70 −75 75 −80
No. of students 5 13 35 25 17 5
Calculate the mean deviation and coefficient of mean deviation about mean.

7. Find the mean deviation and coefficient of mean deviation about median for the following data:
Heights (in cm) 95-105 105-115 115-125 125-135 135-145 145-155
Number of Girls 9 15 23 30 13 10

8. A gardener buys 10 packets of seeds from two different companies. Each pack contains 20 seeds, and he
records the number of plants which grow from each pack.
Company A 20 5 20 20 20 6 20 20 20 8
Company B 17 18 15 16 18 18 17 15 17 18
(i) Find the median and mode for each company’s seeds.
(ii) Which company does the mode suggest is best?
(iii) Find the range for each company’s seeds.

9. The diameter of circles (in mm) drawn in a design are given below:
Diameter (in mm) 33 – 36 37 – 40 41 – 44 45 – 48 49 – 52
No. of circles 15 17 21 22 25
Calculate the standard deviation and mean diameter of the circles.

10. A teacher wants to analyse the performance of 50 students

in a Mathematics test to identify patterns and trends in
their marks distribution. The teacher collects the marks of
all 50 students, organizes them into class intervals such as
0-10, 10-20, 20-30, and so on, and counts the number of
students whose marks fall within each interval. Using this
data, the teacher creates a histogram to represent the
frequency distribution.
(i) Find the mean deviation about median.
(ii) Find the mean deviation about mean.

28
 ISC XI- Applied Mathematics

11. A catering company is comparing the weekly transportation costs of two delivery vehicles - a diesel van
and an electric van - over a few months. The data shows:
• Diesel Van: Mean weekly cost = ₹5,000; Standard Deviation = ₹750
• Electric Van: Mean weekly cost = ₹4,200; Standard Deviation = ₹1,050
(i) Calculate the coefficient of variation (CV) for both vehicles.
(ii) Which vehicle shows more consistent costs and would be better for budgeting?

Answers
1. (i) (c) 55, 10

(ii) (a) 2 and 4

(iii) (d) remains the same as that of the original set

(iv) (c) Both 1 and 2

(v) (a) Both A and R individually true and R is the correct explanation of A.

2. 17
3. range = 14, quartile deviation = 2.85, coefficient of quartile deviation = 0.0792
4. mean = 590, S.D = 90
5. (i) Case 1. Variance = 9.2, standard deviation = 3.03
Case 2. Variance = 9.2, standard deviation = 3.03
(ii) There is no change in variance and standard deviation of the given data if the origin is changed i.e., if a
constant is added to each observation.
6. 5.003 and 0.0769
7. 11.54 and 0.0916
8. (a) Company A: Median = 20, Mode = 20 & Company B: Median = 17, Mode = 18
(b) The mode for Company A is 20 and the mode for Company B is 18. Since a larger number of seeds
grew from the packs in Company A, the mode suggests that Company A is better.
(c) Company A's range: 15, Company B's range: 3
9, Mean = 43.5 Standard deviation = 5.55
10. (i) 12.575 (ii) 12.74
11. (i) CV of Diesel Van = 15%
CV of Electric Van = 25%
(ii) Since the Diesel Van has a lower CV, its weekly costs are more consistent. Therefore, it would be a
better choice for budgeting and planning, despite slightly higher average costs.

29
 ISC XI- Applied Mathematics

Project Work
"Can Measures of Dispersion Be Manipulated to Misrepresent Data?"
Case Study:
A leading consumer electronics company, TechTrend, recently launched a new smartphone. To promote their
product, they conducted a survey comparing customer satisfaction scores between their phone and a
competitor’s model.
The company published the following statistical summary in their advertisement:

Smartphone Model Mean Satisfaction Score (out of 10) Standard Deviation

TechTrend X100 8.5 1.2

Competitor Y200 8.3 2.8

TechTrend claimed that their smartphone had a much higher customer satisfaction rating. However, a
group of independent analysts questioned their use of dispersion measures.
Discussion Questions:
1. What role does Standard Deviation play in understanding customer satisfaction in this case?
2. Does TechTrend’s claim seem misleading? Why or why not?
3. How could a company selectively use measures of dispersion to create a biased representation of data?
4. What other statistical measures could be considered before making a judgment?
5. How can businesses ensure ethical use of statistics in advertising?

30
 ISC XI- Applied Mathematics

HISTORICAL NOTE

The concept of dispersion in statistics has evolved over centuries, with contributions from various
mathematicians and statisticians:
• 17th Century – Beginnings of Probability Theory:
− Blaise Pascal and Pierre de Fermat laid the foundation for probability theory, which later
influenced statistical measures.
• 18th Century – Early Statistical Concepts:
− Abraham de Moivre introduced the normal distribution in his book The Doctrine of Chances
(1733), hinting at the importance of variability in data.
• 19th Century – Formal Development of Dispersion Measures:
− Carl Friedrich Gauss (1809) developed the method of least squares, closely linked to variance
and standard deviation.
− Francis Galton (late 1800s) introduced the concept of statistical correlation and spread,
influencing modern dispersion measures.
• Early 20th Century – Standard Deviation and Variance:
− Karl Pearson formally introduced standard deviation as a measure of spread in 1894.
− Ronald Fisher (1920s) developed analysis of variance (ANOVA), further expanding
applications of dispersion in research.
• Modern Era – Computational Advancements:
− The widespread use of computers in the 20th century enabled large-scale data analysis, making
dispersion measures essential in fields like finance, healthcare, and machine learning.

31
 ISC XI- Applied Mathematics

(ii) Skewness and Kurtosis

SCOPE

• Define Skewness and Kurtosis using graphical representation of a data set.

• Interpret Skewness and Kurtosis of a frequency distribution by plotting the graph.
• Calculate coefficient of Skewness and interpret the results.
• Empirical Relation between Mean, Median and Mode.
• Moments
• Measures of skewness – Absolute and Relative
• Relative measures –
− Karl Pearson’s Coefficient of Skewness.
− Bowley's Coefficient of Skewness.
− β and γ Coefficient of Skewness
• Karl Pearson’s Measures of Kurtosis

LEARNING OUTCOMES

Students will be able to:

• define and calculate coefficient of Skewness.
• define and interpret Skewness and Kurtosis graphically.
• chose appropriate measure of dispersions based on the given data and draw inferences.

PRE-REQUISITES

• Classification and tabulation of data.

• Concept of frequency distributions
• Measures of Central Tendency
• Quartiles
• Standard Deviation

32
 ISC XI- Applied Mathematics

Introduction to Moments
In Statistics, moments are used to describe the characteristic of a distribution. They represent a convenient and
unifying method for summarizing many of the most commonly used statistical measures.
Generally, in any frequency distribution, four moments are obtained which are known as first, second, third
and fourth moments. These four moments describe the information about mean, variance, skewness, and
kurtosis of a frequency distribution.
Calculation of moments gives some features of a distribution which are of statistical importance. Moments
can be classified as raw and central moment. Raw moments are measured about any arbitrary point A (say). If
A is taken to be zero, then raw moments are called moments about origin. When A is taken to be arithmetic
mean we get central moments. The first raw moment about origin is the mean whereas the first central moment
is zero. The second central moment is variance and the third and fourth moments are useful in measuring
skewness and kurtosis.

Moments for Ungrouped Data

If X1, X2, ……………….Xn are n values of the variable X, the rth raw moment of X about any point A is
defined as
𝑛𝑛
1
𝜇𝜇𝑟𝑟 = �(𝑥𝑥𝑖𝑖 − 𝐴𝐴)𝑟𝑟 ; 𝑟𝑟 = 0,1,2, ….
𝑛𝑛
𝑖𝑖=1

𝑥𝑥̅ is the first moment of X about the origin.

Moments for Grouped Data

If X1, X2, ……………….Xn are n values of the variable X, the rth raw moment of X about any point A is
defined as
𝑛𝑛
1
𝜇𝜇𝑟𝑟 = � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝐴𝐴)𝑟𝑟 ; 𝑟𝑟 = 0,1,2, ….
𝑛𝑛
𝑖𝑖=1

Central Moments
If the arbitrary origin of moments of variable X is taken as arithmetic mean i.e. A = 𝑥𝑥̅ the moments are called
central moments.
For ungrouped data, the central moment of variable X is given by
𝑛𝑛
1
𝜇𝜇𝑟𝑟 = �(𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )𝑟𝑟 ; 𝑟𝑟 = 0,1,2, …
𝑛𝑛
𝑖𝑖=1

For grouped data, the central moment of variable X is given by

𝑛𝑛
1
𝜇𝜇𝑟𝑟 = � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )𝑟𝑟 ; 𝑟𝑟 = 0,1,2, ⋯
𝑛𝑛
𝑖𝑖=1

When r = 0,
𝑛𝑛
1 𝛴𝛴𝑓𝑓𝑖𝑖
𝜇𝜇0 = � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )0 = =1
𝑛𝑛 𝑛𝑛
𝑖𝑖=1

33
 ISC XI- Applied Mathematics

When r = 1,
𝑛𝑛
1 𝛴𝛴𝑓𝑓𝑖𝑖 𝑥𝑥𝑖𝑖 𝛴𝛴𝛴𝛴
𝜇𝜇1 = � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )1 = − 𝑥𝑥̅ = 𝑥𝑥̅ − 𝑥𝑥̅ = 0
𝑛𝑛 𝑛𝑛 𝑛𝑛
𝑖𝑖=1

When r = 2,
𝑛𝑛
1
𝜇𝜇2 = � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )2 = 𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉
𝑛𝑛
𝑖𝑖=1

Moments can be extended to higher powers but in actual practice the first four moments are enough to describe
a series.

Introduction to Skewness
In statistics, when we collect and organize data, we often create a graph to see how the data is distributed.
A normal curve, also known as a Gaussian or bell curve, is a symmetrical, bell-shaped probability
distribution [Link] normal curve is symmetrical around its mean. In a normal distribution, the mean,
median, and mode are all equaland located at the centre of the curve. The curve has a characteristic bell shape,
with the highest point at the mean and the tails tapering off towards both ends.
But real-life data is not always symmetrical. Sometimes one side of the graph stretches out more than the
other. This is called skewness.

What is Skewness?
A distribution is said to be 'skewed' when the mean and the median fall at different points in the distribution,
and the balance (or centre of gravity) is shifted to one side or the other-to left or right. Measures of skewness
tell us the direction and the degree of asymmetry.
In symmetrical distribution the mean, median and mode are identical. The more the mean moves away from
the mode, the larger the asymmetry or skewness.

Types of Skewness
1. Symmetrical Distribution

• Mean = Median = Mode

• Both sides of the curve are equal
• Example: Perfectly fair exam results

2. Positively Skewed Distribution (Right-Skewed)

• Tail is longer on the right

• Mean > Median > Mode
• Example: Income distribution (most people earn less but a few people like
CEOs, actors, cricketers, businessmen earn very high)

34
 ISC XI- Applied Mathematics

3. Negatively Skewed Distribution (Left-Skewed)

• Tail is longer on the left

• Mean < Median < Mode
• Example: Age at retirement (most people retire around 60–65 but
nowadays a few retire very early by taking voluntary retirement)

Empirical Relation between Mean, Median and Mode

In cases of moderately skewed distributions, the values of the mean, median and mode have an empirical
relationship given by Karl Pearson
Mean – Mode = 3( Mean – Median)
When mode is ill defined this formula can be used to calculate mode. An ill-defined mode occurs when there
are two or more values in the data set that have the same highest frequency.

Real life example of positive skewness

The Tale of Online Orders and the Waiting Game

Once upon a time in the busy world of online shopping, most customers were happy. They would place
their orders and, like clockwork, their packages would arrive within 1 to 3 days. Quick, efficient, and right
on time — it was almost magical.
But every now and then, a few unlucky souls had to wait… and wait… and wait. Their orders didn’t show
up on time sometimes getting delayed by even 10 days.
This unusual pattern made the delivery data look a bit lopsided. On a graph, there was a big hump where
most of the fast deliveries clustered but there was also a tail stretching off to the right, marking those rare
but frustrating delays.
And so, the story of delivery times became a lesson in statistics. Because of those few long delays, the
distribution wasn’t balanced. It was positively skewed with a longer tail on the right, reminding everyone
that while most packages arrive on time, some take their sweet time to show up!

Why is Skewness Important?

Skewness helps us to:
• understand the shape of the data distribution.
• know whether the mean is being pulled to one side.
• decide which measures of central tendency (mean, median, mode) to trust.

Measures of Skewness
To study the extent of asymmetry and direction in a series, various measures of skewness are employed. These
measures of skewness can be both absolute or relative.

35
 ISC XI- Applied Mathematics

Absolute Measures of Skewness

Absolute measures tell us the extent of asymmetry and whether it is positive or negative.
1) The first absolute measure of skewness is based on the difference between mean and mode or mean and
median.
(i) Absolute Sk = Mean − Mode or
(ii) Absolute Sk = Mean − Median.

If the value of mean is greater than the mode or median, skewness is

positive, otherwise it is negative.
It may be noted that for a positively skewed distribution; the value of
the mean is the greatest and the value of mode is the least of the three
measures. Likewise, for a negatively skewed distribution, mode has the
maximum value and mean has the least value. In both the cases median
is between, the mean and mode.
2) The second measure of skewness, based on quartiles and, depends upon
the fact that for a symmetrical distribution Q1 andQ3 are equidistant from
the median,
i.e., Q3-Md = Md-Q1

But if a distribution is asymmetrical, then one quartile lying on the

longer tail side will be farther from the median than the other quartile. In such a case absolute measure
of skewness can be, measured by the following formula:
Absolute Sk = (Q3−Md) – (Md−Q1) = Q3 + Q1 − 2Md

The formula shows if (Q3−Md) > (Md−Q1), skewness is positive otherwise it is negative. This is true
because (Q3−Md) > (Md−Q1) implies that there is a longer tail on the right hand side i.e., skewness is
right handed or positive.

Rules to know the Degree of Skewness

Value of skewness Inference

−0.5 to 0.5 Fairly symmetrical

−1 to −0.5 (negatively skewed) Moderately skewed

0.5 to 1 (positively skewed)
Less than −1(negatively skewed) Highly skewed
More than 1(positively skewed)

Relative Measures of Skewness

In order to make comparison between the skewness in two or more distributions, coefficient of
skewness is computed for the given series or distributions.
The following are the important methods of measuring relative skewness:
Karl Pearson’s Coefficient of Skewness
This method is most frequently used for measuring skewness and is based on first absolute measure. The
Pearson’s coefficients of skewness can be measured as:

𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 − 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀
𝑠𝑠𝑘𝑘 =
𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷
36
 ISC XI- Applied Mathematics

When mode is ill defined,

3(𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 − 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀)
𝑠𝑠𝑘𝑘 =
𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷
These two measures are called, respectively, Pearson’s first and second coefficients of [Link] value
will be free of units of the data.

Mean = Median = Mode 𝑆𝑆𝑘𝑘 = 0 Data is symmetrical

𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀<𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 or 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀<𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑆𝑆𝑘𝑘< 0 Data is negatively skewed

𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀>𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 or 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀>𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑆𝑆𝑘𝑘> 0 Data is positively skewed

The limits of Karl Pearson’s coefficient of skewness is between -3 to 3.

Bowley's Coefficient of Skewness

This method is based on quartiles, i.e., second absolute measure of skewness. The Bowley's Coefficient of
Skewness can be calculated as :

(𝑄𝑄3 − 𝑀𝑀𝑑𝑑 ) − (𝑀𝑀𝑑𝑑 − 𝑄𝑄1 ) 𝑄𝑄3 + 𝑄𝑄1 − 2𝑀𝑀𝑑𝑑

𝑆𝑆𝑘𝑘 = =
(𝑄𝑄3 − 𝑀𝑀𝑑𝑑 ) + (𝑀𝑀𝑑𝑑 − 𝑄𝑄1 ) 𝑄𝑄3 − 𝑄𝑄1

This method is particularly useful in case of open end distributions and where extreme values are present or
when class-intervals are unequal. Skewness should be measured by Bowley's method also when positional
measures are called for.

𝑄𝑄3−𝑄𝑄2 = 𝑄𝑄2 − 𝑄𝑄1 𝑆𝑆𝑘𝑘 = 0 Data is symmetrical

𝑄𝑄3+𝑄𝑄1< 2𝑄𝑄2 𝑆𝑆𝑘𝑘< 0 Data is negatively skewed

𝑄𝑄3+𝑄𝑄1> 2𝑄𝑄2 𝑆𝑆𝑘𝑘> 0 Data is positively skewed

𝑄𝑄1 = 𝑄𝑄2 𝑆𝑆𝑘𝑘= +1 Data is perfectly positively skewed

𝑄𝑄2 = 𝑄𝑄3 𝑆𝑆𝑘𝑘 = −1 Data is perfectly negatively skewed

The limits of Bowley’s coefficient of skewness is between −1 to 1.

But the main drawback of this measure is that it is based on central 50% of the data and it ignores the
remaiaing 50% of the data i.e., 25% of the data below Q1 and 25% of the data above Q3.

β and γ Coefficient of Skewness

Karl Pearson defined the following β and γ coefficients of skewness, based upon the second and third central
moments:
𝜇𝜇32
𝛽𝛽1 = 3
𝜇𝜇2
Where 𝜇𝜇2 and 𝜇𝜇3 are second and third moments. It is used as measure of skewness. For a symmetrical
distribution, β1= 0.
But β1 as a measure of skewness does not tell us about the direction of skewness, i.e. positive or negative.
Though µ3 being the sum of cubes of the deviations from mean may be positive or negative but µ32 is always
positive. Also, µ2 being the variance always positive. Hence, β1 will always be positive.
37
 ISC XI- Applied Mathematics

This drawback is removed if we calculate Karl Pearson’s Gamma coefficient γ1.

𝜇𝜇3
𝛾𝛾1 = ±�𝛽𝛽1 =
�𝜇𝜇23

Then the sign of skewness would depend upon whether the value of µ3 and is positive or negative. That is
why γ1 is preffered to measure skewness.

Example 1. The table below shows the weekly pocket money (₹) received by 103 students:
Pocket Money 100 200 300 400 500 600 700 800 900 1000
No of students 3 5 8 10 12 15 20 15 10 5

Plot the data on the graph and interpret it.

Solution:

From the graph we can infer that for the curve the tail is longer on the left. So the data is negatively skewed.

Example 2. From the marks secured by 120 students in Sections A and B of a class, the following measures
are obtained:
Section A : 𝑥𝑥̅ = 46.83, 𝜎𝜎 = 14.8, Mode = 51.67
Section B : 𝑥𝑥̅ = 47.83, 𝜎𝜎 = 14.8, Mode = 47.07
Find the skewness of marks in both sections. Which section has more skewed marks and in which direction?
Solution:
𝑥𝑥̅ −𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 46.83−51.67
For Section A: 𝑆𝑆𝑘𝑘 = = = −0.327
𝜎𝜎 14.8

𝑥𝑥̅ −𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 47.83−47.07

For Section B: 𝑆𝑆𝑘𝑘 = = = 0.051
𝜎𝜎 14.8

So, the data is fairly symmetrical for both cases.

But the distribution of marks in Section A is more skewed. The skewness for Section [Link] negative, while that
of B is positive.

38
 ISC XI- Applied Mathematics

Example 3. For a given data, Q1 = 58, Md = 59 and Q3 = 61, find the coefficient of skewness.
Solution . We know
𝑄𝑄3 + 𝑄𝑄1 − 2𝑀𝑀𝑑𝑑 61 + 58 − 2 × 59
𝑆𝑆𝑘𝑘 = = = 0.33
𝑄𝑄3 − 𝑄𝑄1 61 − 58

So, the data is fairly symmetrical.

Example 4. In a frequency distribution, the coefficient of skewness based upon quartiles is 0.6. If the sum of
the upper and lower quartiles is 100 and the median is 38, find the value of the upper quartile.
Solution. We know,
𝑄𝑄3 + 𝑄𝑄1 − 2𝑀𝑀𝑑𝑑
𝑆𝑆𝑘𝑘 =
𝑄𝑄3 − 𝑄𝑄1
Substituting the given values,
100−2×38 100−2×38
0.6 = 𝑄𝑄3 −𝑄𝑄1
⇒ 𝑄𝑄3 − 𝑄𝑄1 = 0.6
= 40
Now,
(𝑄𝑄3 + 𝑄𝑄1 ) + (𝑄𝑄3 − 𝑄𝑄1 ) = 100 + 40 ⇒ 2𝑄𝑄3 = 140 ⇒ 𝑄𝑄3 = 70

Hence the upper quartile is 70.

Example 5. Assume that a firm has selected a random sample of 100 employees from its production line
representing their monthly working hours and has obtained the data shown in table below.
Monthly 130 - 134 135 - 139 140 - 144 145 - 149 150 - 154 155 - 159 160-164
working
hours
No of 3 12 21 28 19 2 5
employees
Compute:
(i) the arithmetic mean
(ii) the standard deviation
(iii) Karl Pearson’s Coefficient of Skewness.
Solution:From the given data we obtained

CI x f d fd fd2

130 - 134 132 3 -3 9 27

135 - 139 137 12 -2 4 48

140 - 144 142 21 -1 1 21

145 - 149 147 28 0 0 0

150 - 154 152 19 1 1 19

155 - 159 157 12 2 4 48

160-164 162 5 3 9 45
100 4 208
39
 ISC XI- Applied Mathematics

A = 147

4 208 4 2
𝑥𝑥̅ = 147 + � × 5� = 147 ⋅ 2 𝝈𝝈 = � −� � × 5 = 7 ⋅ 21
100 100 100

Modal Class = 144.5 – 149.5

28−21
Mode = 144.5 + × 5 = 146.69
2(28)−21−19

𝑥𝑥̅ −𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 147.2−146.69

Karl Pearson’s Coefficient of Skewness 𝑆𝑆𝑘𝑘 = = = 0.07
𝜎𝜎 7.21

Example 6. Calculate the Bowley’s Coefficient of Skewness from the data given below:
CI 30 - 40 40 - 50 50 - 60 60 - 70 70 - 80 80 - 90 90 - 100
Frequency 1 3 11 21 43 32 9
Solution:From the given data we obtained,
CI f Cf
30 - 40 1 1
40 - 50 3 4
50 - 60 11 15
60 - 70 21 36
70 - 80 43 79
80 - 90 32 111
90 - 100 9 120
n = 120
𝑛𝑛 120
𝑄𝑄1 =�4�th item =� 4 �th item =30th item
∴ 𝑄𝑄1 lies in the class = 60 – 70
𝑛𝑛
� �−𝑐𝑐𝑐𝑐 30−15
𝑄𝑄1 = 𝐿𝐿 + 4
× 𝐼𝐼 = 60 + × 10 = 67.14
𝑓𝑓 21

𝑛𝑛 120
Median =2 �4�th item =2 � 4 �th item = 60th item
∴𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 lies in the class = 70 – 80
𝑛𝑛
2 � �−𝑐𝑐𝑐𝑐 60−36
𝑄𝑄2 = 𝐿𝐿 + 4
× 𝐼𝐼 = 70 + × 10 = 75.5
𝑓𝑓 83

𝑛𝑛 120 th
𝑄𝑄3 =3 �4�th item =3 � 4
� item =75th item

∴ 𝑄𝑄3 lies in the class = 80 – 90

𝑛𝑛
3 � �−𝑐𝑐𝑐𝑐 90−79
𝑄𝑄3 = 𝐿𝐿 + 4
× 𝐼𝐼 = 80 + × 10 = 83.44
𝑓𝑓 32
𝑄𝑄3 +𝑄𝑄1 −2𝑀𝑀𝑑𝑑 83.44+67.14−2×75.5
Bowley’s Coefficient of Skewness 𝑆𝑆𝑘𝑘 = = = −0.035
𝑄𝑄3 −𝑄𝑄1 83.44−67.14

40
 ISC XI- Applied Mathematics

Example 7. Find out if there is skewness in the distribution and whether the skewness is positive or negative
given that 𝜇𝜇1 = 0, 𝜇𝜇2 = 40, 𝜇𝜇3 = −100, 𝜇𝜇4 = 200.
Solution:We know

𝜇𝜇32 (−100)2 1
𝛽𝛽1 = 3 = 3
=
𝜇𝜇2 (40) 64
1
Since, 𝜇𝜇3 is negative we can infer that skewness is also negative and of the value
64
.

Example 8. You are given the following information before and after the settlement of an industrial dispute:

Before the dispute After the dispute

No of workers 1000 950
Average wage 308 325
Standard deviation 25 20
Median wage 320 300
Draw inferences using the mean, median, mode, and skewness, Comment on the gains and losses from the
point of view of workers and that of the management.
Solution:
(i) The number of workers have declined from 1000 to 950. Hence, some workers have become
unemployed or changed the job.
(ii) The average wage of workers have gone up from Rs 308 to Rs 325.
Total amount spent on wages before dispute = 308 × 1000 = 3,08,000.
Total amount spent on wages after dispute = 325 × 950 = 3,08,750.
Thus, the management has satisfied the workers with increased pay without substantially increasing
the wage bills due to the lessing of the number of workers.
(iii) The median wage of Rs 320 has gone down to Rs 300. This implies that before the dispute 50% of
workers were getting more than Rs 320 per head but after the settlement they are now getting only Rs
300. This means that the wages of the higher paid staff have decreased.
(iv) Using the empirical formula,
Mode before dispute = (3 × 320) − (2 × 308) = 𝑅𝑅𝑅𝑅 344
Mode after settlement = (3 × 300) − (2 × 325) = 𝑅𝑅𝑅𝑅 250
This signifies that the wages have become concentrated around a lower value implying thata large
number of workers will now be getting lower wages.
3(𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀) 3(308−320)
(v) Before settlement 𝑠𝑠𝑘𝑘 =
𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷
= 25
= −1.44

3(𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀) 3(325−300)
After settlement 𝑠𝑠𝑘𝑘 = = = +3.75
𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 20

Thus, from negative skewness before the settlement the skewness has become positive so there is no longer a
tail towards the right of the curve. This indicates that the number of workers getting more wages has gone
down.

Remarks about Skewness

1. If the value of mean, median and mode are same in any distribution, then the skewness does not exist in
that distribution. Larger the difference in these values, larger the skewness;
2. If sum of the frequencies are equal on the both sides of mode then skewness does not exist;
3. If the distance of first quartile and third quartile are same from the median then a skewness does not
[Link] is no asymmetry;
41
 ISC XI- Applied Mathematics

4. If the sums of positive and negative deviations obtained from mean, median or mode are equal then there
is no asymmetry;
5. If a graph of a data become a normal curve and when it is folded at middle and one part overlap fully on
the other one then there is no asymmetry.

Introduction to Kurtosis
In statistics, kurtosis is a measure that describes the "peakedness" or flatness of a distribution curve
compared to a normal distribution.
While skewness tells us how stretched the curve is to the left or right, kurtosis tells us how tall and sharp
or flat and wide the peak of the curve is, which further indicates the presence and extent of outliers by
assessing the tails and peaks.

What is Kurtosis?
Kurtosis is another measure of the shape of a frequency curve.
It is a Greek word, which means bulginess. While skewness
signifies the extent of asymmetry, kurtosis measures the degree
of peakedness of a frequency distribution. Karl Pearson
classified curves into three types on the basis of the shape of
their peaks. These are Mesokurtic, leptokurtic and platykurtic.
These three types of curves are shown in figure below:

Types of Kurtosis

1. Mesokurtic (Normal Kurtosis)

• The curve is normal-shaped (bell curve)
• Moderate peak and tails
• Reference level: kurtosis = 3

2. Leptokurtic (High Kurtosis)

• Curve is taller and sharper than normal
• Heavy tails, meaning more data is in the extremes
• Indicates more outliers
• Kurtosis > 3

3. Platykurtic (Low Kurtosis)

• Curve is flatter and wider than normal
• Light tails, meaning fewer extreme values
• Indicates less variation in extreme values
• Kurtosis < 3

42
 ISC XI- Applied Mathematics

William Gosset, an English statistician, in his 1927 paper, used the visual analogy of a platypus for
platykurtic and two leaping kangaroos for leptokurtic distributions to aid in remembering the shapes.
(Source: ‘Student’, 1972, Errors of routine [Link],19,[Link] p.160)

Why is Kurtosis Important?

Kurtosis helps in:
• understanding risk and outliers in data.
• identifying how extreme values affect the distribution.
• deciding whether the data follows a normal pattern or not.
Real life examples of types of Kurtosis:

Imagine three classes took the same 100-mark test:

Class A (Leptokurtic – High Kurtosis):
• Most students scored around 50 marks.A few students scored very high (95+) or very low (below
10). This means:
• The distribution has a sharp peak near 50 but also has thick tails (lots of extreme values).This is
leptokurtic — tall peak and heavy tails → kurtosis > 3
Class B (Platykurtic – Low Kurtosis):
• Students are spread out evenly from 30 to [Link] are hardly any extreme scores.
This means:
• The graph is flatter with few or no [Link] is platykurtic — flat peak and light tails →
kurtosis < 3
Class C (Mesokurtic – Normal Kurtosis):
• Scores are distributed like a bell curve. Most students scored between 40 and 60. A few scored
higher or [Link] means:
This is mesokurtic — standard normal shape → kurtosis = 3.

43
 ISC XI- Applied Mathematics

Karl Pearson defined the following β and γ coefficients of kurtosis, based upon the second and fourth central
moments:
𝜇𝜇4
𝛽𝛽2 =
𝜇𝜇22

Or, 𝛾𝛾2 = 𝛽𝛽2 − 3

𝜷𝜷𝟐𝟐 = 3 or 𝜸𝜸𝟐𝟐 = 0 curve is said to be mesokurtic

𝜷𝜷𝟐𝟐 < 3 or 𝜸𝜸𝟐𝟐 < 0 curve is said to be platykurtic

𝜷𝜷𝟐𝟐 > 3 or 𝜸𝜸𝟐𝟐 > 0 curve is said to be leptokurtic

𝜷𝜷𝟐𝟐 = 3 or 𝜸𝜸𝟐𝟐 = 0 curve is said to be mesokurtic

𝜷𝜷𝟐𝟐 < 3 or 𝜸𝜸𝟐𝟐 < 0 curve is said to be platykurtic

𝜷𝜷𝟐𝟐 > 3 or 𝜸𝜸𝟐𝟐 > 0 curve is said to be leptokurtic

Example 9. The table below shows the marks obtained by 23 students

Marks 2 3 4 5 6 7 8
No of Students 1 1 4 12 3 1 1

Plot the data on the graph and interpret it.

Solution:

Since the curve is taller and sharper than normal, it is a leptokurtic curve.

44
 ISC XI- Applied Mathematics

Example 10. The first four moments of a distribution are 0, 2.5, 0.7 and 18.75. Comment on the Kurtosis of
the distribution.
Solution:

𝜇𝜇4 18.75
𝛽𝛽2 = = =3
𝜇𝜇22 (2.5)²
The curve is Mesokurtic – Normal Kurtosis.

Example 11. Calculate Skewness and Kurtosis from the following grouped data.

Class Frequency

0-2 5

2-4 16

4-6 13

6-8 7

8 - 10 5

10 - 12 4

Solution: From the given data we obtained,

x f fx �
𝒙𝒙 − 𝒙𝒙 �)𝟐𝟐
𝒇𝒇(𝒙𝒙 − 𝒙𝒙 �)𝟑𝟑
𝒇𝒇(𝒙𝒙 − 𝒙𝒙 �)𝟒𝟒
𝒇𝒇(𝒙𝒙 − 𝒙𝒙
Class
0-2 1 5 5 -4.12 84.872 -349.6726 1440.6513

2-4 3 16 48 -2.12 71.9104 -152.45 323.1941

4-6 5 13 65 -0.12 0.1872 -0.0225 0.0027

6-8 7 7 49 1.88 24.7408 46.5127 87.4439

8 - 10 9 5 45 3.88 75.272 292.0554 1133.1748

10 - 12 11 4 44 5.88 138.2976 813.1899 4781.5565

50 256 395.28 649.6128 7766.0233

45
 ISC XI- Applied Mathematics

Now,
∑𝒇𝒇𝒇𝒇 256
x� = = = 5 ⋅ 12,
Σf 50
1 𝑛𝑛 395.28
𝜇𝜇2 = �𝑖𝑖=1 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )2 = = 7.9056,
𝑛𝑛 50
𝑛𝑛
1 649.6128
𝜇𝜇3 = � 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )3 = = 12.99,
𝑛𝑛 50
𝑖𝑖=1
1 𝑛𝑛 7766.0233
𝜇𝜇4 = 𝑛𝑛 �𝑖𝑖=1 𝑓𝑓𝑖𝑖 (𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )3 = 50
= 155.32,

𝜇𝜇2 (12.99)2
𝛽𝛽1 = 𝜇𝜇33 = (7.9056)3
= 0.342,
2

Since, 𝜇𝜇3 is positive we can infer that skewness is also positive and of the value 0.342.
𝝁𝝁𝟒𝟒 155.32
𝜷𝜷𝟐𝟐 = 𝟐𝟐
= = 2.4852
𝝁𝝁𝟐𝟐 (7.9056)²
The curve is Platykurtic – Low Kurtosis.

Comparison chart for Skewness and Kurtosis

Feature Skewness Kurtosis
Measures the "tailedness" or peakedness
Measures the asymmetry or lack of
Definition of a distribution compared to a normal
symmetry in a distribution
distribution
Focus Asymmetry of the distribution Shape of the distribution's peak and tails
Indicates the direction and extent of Indicates the presence and extent of
Interpretation
asymmetry (left or right) outliers by assessing the tails and peaks
Positive (right skew) Leptokurtic (heavy tails),
Types Negative (left skew) Mesokurtic (normal)
Zero (symmetrical) Platykurtic (light tails)
Value > 3: Leptokurtic (heavy tails)
Positive value: Right skew
Value Value = 3: Mesokurtic (normal
Negative value: Left skew
Interpretation distribution)
Zero value: Symmetrical distribution
Value < 3: Platykurtic (light tails)

46
 ISC XI- Applied Mathematics

REAL- LIFE APPLICATIONS

Real-World Applications of Skewness and Kurtosis

Finance:
• Investment strategies:Skewness and kurtosis help assess the risk and potential returns of
investments.
• Identifying risky assets:High kurtosis (heavy tails) may indicate higher risk as extreme events
are more likely.
• Portfolio diversification:Understanding the shape of asset returns helps in building diversified
portfolios.
Quality Control:
• Production processes:Skewness and kurtosis can help identify errors or defects in production,
aiding in quality improvement.
• Understanding deviations from the norm:These measures help pinpoint deviations from the
expected distribution of quality characteristics.
Psychological Research:
• Understanding traits and behaviors: Skewness and kurtosis can help analyze the distribution of
psychological traits or behaviors among different groups.
• Identifying outliers: Kurtosis helps identify individuals with extreme scores or responses.
Environmental Studies:
• Climate patterns: Skewness and kurtosis can aid in understanding the distribution of climate
variables like temperature and precipitation.
• Identifying unusual events: These measures can help identify unusual or extreme climate
events.

Exercises
1. Choose the correct option:
(i) In Uni-model distribution, if the mode = 50 and mean = 60, then the distribution will be
(a) Negatively skewed (c) Positively skewed
(b) Symmetrical (d) Not possible to draw inference

(ii) If the third moment about mean is zero, then the distribution is:
(a) Negatively skewed (c) Positively skewed
(b) Symmetrical (d) Not possible to draw inference

(iii) In a symmetrical distribution, Q3 – Q1 = 20, median = 15. Q3 is equal to:

(a) 5 (c) 20
(b) 15 (d) 25

47
 ISC XI- Applied Mathematics

(iv) Consider the following statements:

Statement 1: When most of the data are concentrated on the left of the distribution, it is called a
positively skewed distribution.
Statement 2: A symmetric distribution is one in which the left and right sides mirror each other.
Which of the statement(s) given above is/are true?
(a) Only 1 (c) Both 1 and 2
(b) Only 2 (d) Neither 1 nor 2

(v) Assertion: The second central moment is always non-negative.

Reason: The second central moment is the arithmetic mean of the squared deviations of
values from the mean.
(a) Both A and R individually true and R is the correct explanation of A
(b) Both A and R individually true but R is not the correct explanation of A
(c) A is true but R is false
(d) A is false but R is true

2. For 40 observations, it is given that ∑𝑓𝑓𝑓𝑓 = 322, 𝛴𝛴𝛴𝛴𝑥𝑥 2 = 2780 and mode = 8. Calculate the Karl
Pearson’s coefficient of skewness with the given data and comment on the degree of skewness.

3. The standard deviation of symmetrical distribution is 3. What must be the value of the fourth moment
about the mean in order that the distribution be mesokurtic?

4. In a frequency distribution the coefficient of skewness based on quartiles is 0.6. If the sum of the upper
and the lower quartiles is 100 and the median is 38, find the value of the upper quartile.

5. Calculate the Karl Pearson’s coefficient of skewness for the following data:

Age (in years) 20 - 30 30 - 40 40 - 50 50 - 60 60 - 70 70 - 80 80 - 90

No of persons 3 61 132 153 140 51 2

6. Calculate Bowley’s coefficient of skewness for the following distribution of weekly wage of workers.
Wages Below 300 300 - 400 400 - 500 500 - 600 600 - 700 Above 700
5 8 18 35 27 7
No of workers

7. The first four central moments of distribution are 0, 2.5, 0.7 and 18.75. Comment on the skewness and
kurtosis of the distribution.

8. Calculate the moment coefficient of skewness β1 from the following data.

Marks 10 - 20 20 - 30 30 - 40 40 - 50 50 - 60
No. of marks 2 4 7 6 1

48
 ISC XI- Applied Mathematics

9. For two distributions the following moments about mean is given:

First distribution: 𝜇𝜇2 = 25, 𝜇𝜇3 = −10, 𝜇𝜇4 = 1560
Second distribution: 𝜇𝜇2 = 36, 𝜇𝜇3 = −28.8, 𝜇𝜇4 = 4000
(i) Which of the two distributions, is more skewed to the left.
(ii) From the point of view of peakedness, which distribution more nearly approximates a normal
distribution.

10. Compute the moment coefficient of kurtosis from the following data.

x 5 10 15 20 25 30 35
f 8 15 18 29 23 17 5

11.
(i) Calculate the Pearson's coefficient of skewness
for Curve B.
(ii) Identify which curve is symmetrical, positively
skewed, and leptokurtic, respectively. Justify
your answer.
(iii) Based on the given standard deviations, which
curve is most dispersed?

12.
(i) Which of the three curves shows no skewness?
(ii) Which curve indicates positive skewness?
(iii) Which curve has highest kurtosis?

Answers
1. (i) (c) Positively skewed

(ii) (b) Symmetrical

(iii) (d) 25

(iv) (c) Both 1 and 2

(v) (a) Both A and R individually true and R is the correct explanation of A.

2. 0.023(fairly symmetrical)
3. 243
4. Upper quartile is 70
5. – 0.122

49
 ISC XI- Applied Mathematics

6. Bowley's coefficient of skewness is −0.0515.

7. 𝛽𝛽1 =0.031, the distribution is slightly skewed.𝛽𝛽2 = 3, the distribution is mesokurtic.
8. 0.068
9. (i) The second distribution is more skewed to the left.
(ii) The second distribution more nearly approximates a normal distribution.
10. 2.26 (Platykurtic)
11. (i) 0.67 (Positive skewness)
(ii) Symmetrical: Curve A, Positively skewed: Curve B, Leptokurtic: Curve C
(iii) Curve B (most dispersed, highest SD)
12. (i) Curve A
(ii) Curve B
(iii) Curve C

Project Work
"Analysing Health and Lifestyle Patterns Using Skewness and Kurtosis"
Objective:
To collect real-life data, calculate skewness and kurtosis, and interpret what they tell us about health and
lifestyle trends in a selected group.
Steps to Follow:
1. Choose a Dataset
Pick any one of the following and try to collect data from 20–30 individuals for better analysis.

• Number of steps walked per day by Class XII students.

• Daily screen time (in hours) of Class XII students.
• Sleep duration per night of Class XII students.
• Water intake per day (in litres or glasses) by Class XII students.
2. Organize the Data
Make a frequency table.
3. Calculate:
• Mean, Median, Mode
• Standard deviation
• Skewness
• Kurtosis
4. Graphical Representation
Include Histogram and Graphs
5. Interpret the Results.

50
 ISC XI- Applied Mathematics

6. Conclusion
• Are Class XII students healthy?
• Is the lifestyle consistent or highly varied?
• Are there outliers?
• What could improve?

HISTORICAL NOTE

• Skewness and kurtosis, measures of distribution shape, were introduced by Karl Pearson, an English
biostatistician and mathematician, in the late 19th and early 20th centuries. Karl Pearson introduced
the concept of kurtosis in 1905 and developed a system of probability distributions that varied in
skewness, measured relative to the mode.
• Ronald Fisher, British mathematical statistician and biologist, with his 1925 text "Statistical
Methods for Research Workers," helped introduce skewness and kurtosis to many generations of
statisticians.
• While Thiele, a Danish astronomer, actuary, and mathematician, did not use the terms "skewness"
and "kurtosis”, but his work laid the groundwork for these concepts.

51
 ISC XI- Applied Mathematics

(iii) Correlation Analysis

SCOPE

• Definition and meaning of covariance.

• Coefficient of Correlation by Karl Pearson.
𝐼𝐼𝐼𝐼 𝑥𝑥 − 𝑥𝑥, 𝑦𝑦 − 𝑦𝑦 𝑎𝑎𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑛𝑛𝑛𝑛𝑛𝑛 − 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛, 𝑤𝑤𝑤𝑤 𝑢𝑢𝑢𝑢𝑢𝑢

∑ ( x - x )( y - y ) ∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
r= = ∑𝑑𝑑 2
×∑𝑑𝑑𝑥𝑥 2
∑ (x - x ) ∑(y - y)
2 2 𝑥𝑥

If x and y are small numbers, we use

1
∑ xy − ∑ x∑ y
r= N
1
2
∑x − (∑ x )2 ∑ y 2 − 1 (∑ y )2
N N

Otherwise, we use assumed means

A and B, where u = x-A, v = y-B

1
( ∑ u )( ∑ v )
∑ uv -
r= N
2 1 2 2 1 2
∑ u − (∑ u) ∑ v − (∑ v)
N N

• Differentiate between causation and correlation.

• Rank Correlation by Spearman’s
(Correction included)

LEARNING OUTCOMES

Students will be able to:

• define Covariance and Correlation.
• relate Coefficient of correlation and Causation.

PRE-REQUISITES

• Organizing and presenting data using tables, graphs, and charts.

• Concept of how data is analyzed and interpreted, including the concept of variability.
• Mean and Standard Deviation.

52
 ISC XI- Applied Mathematics

Introduction
In statistics, earlier methods such as measures of central tendency and dispersion are primarily used to analyse
a single variable at a time. This kind of analysis, where only one variable is involved, is known as Univariate
Distribution.
However, in many real-world situations, data involves two variables simultaneously. For example, we may
study relationships between income and expenditure, price and demand, or height and weight. When two
variables are considered together, the data is referred to as a Bivariate Distribution.
In such cases, it becomes essential to examine whether and how the variables are related. This leads us to the
concept of Correlation, which is a statistical technique used to measure the strength and direction of a
relationship between two variables. Understanding correlation helps in drawing meaningful conclusions and
making informed decisions based on patterns observed in the data.
Before understanding the idea of correlation let us first learn about the covariance.

Covariance
Covariance is a statistical representation of the degree to which the two given variables vary together.
Basically, Covariance is a number reflecting the degree to which the two variables vary together.
• If both variables increase or decrease together, the covariance is positive.
• If one variable increase while the other decreases, the covariance is negative.
• If there is no consistent pattern, the covariance is close to zero.
The symbol of Covariance of two variables (say X and Y) is denoted by Cov (X, Y).

𝛴𝛴(𝑥𝑥𝑖𝑖 − 𝑥𝑥̅ )(𝑦𝑦𝑖𝑖 − 𝑦𝑦�) 𝛴𝛴𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦

Cov(𝑋𝑋, 𝑌𝑌) = =
𝑛𝑛 𝑛𝑛

where 𝑑𝑑𝑥𝑥 = deviation of X from Mean = 𝑋𝑋 − 𝑋𝑋�

𝑑𝑑𝑦𝑦 = deviation of Y from Mean = 𝑌𝑌 − 𝑌𝑌�
𝑛𝑛 = number of pair of observations

Note: Covariance is independent of the choice of the origin, but it depends on the scale of measurement.
We can also say,
1 1 1
Cov(𝑋𝑋, 𝑌𝑌) = �𝛴𝛴𝑥𝑥𝑖𝑖 𝑦𝑦𝑖𝑖 − Σ𝑥𝑥𝑖𝑖 ∑𝑦𝑦𝑖𝑖 �= 2 [𝑛𝑛𝑛𝑛𝑥𝑥𝑖𝑖 𝑦𝑦𝑖𝑖 − Σ𝑥𝑥𝑖𝑖 ∑𝑦𝑦𝑖𝑖 ]
𝑛𝑛 𝑛𝑛 𝑛𝑛

Even though, the value of covariance indicates whether statistical relationship exists between two variables or
not, it is not often used as its value depends on the dimensions(units) used for variables. The value of
covariance may be very large if some particular units of measurement for X and Y are used, while it may be
small if some other units of measurements are used.
Therefore, we define a new quantity which does not depend on dimensions of variables X and Y, that is we
define a non-dimensional quantity to study correlation. This measure of correlation is called Coefficient of
Correlation.

Correlation
Correlation is a statistical tool used to measure and analyze the relationship between two variables. It helps us
understand how one variable behaves in relation to another, which is particularly useful in studying economic
and social behavior. By examining correlation, we can identify whether changes in one variable are associated
with changes in another. For example, suppose we want to study the relationship between hours studied and
marks obtained by students. If students who study more hours generally score higher marks, then there is a
positive correlation between study hours and marks.

53
 ISC XI- Applied Mathematics

When Are Two Variables Said to Be Correlated?

Two variables are said to be correlated when a change in one variable results in a corresponding change in the
other.

Why Study Correlation?

Studying correlation allows economists, researchers, and decision-makers to uncover patterns and make
informed predictions.

Causation (Cause and Effect)

Causation means that a change in one variable directly results in a change in another. It implies a mechanistic
link — one event is the cause, and the other is the effect. That is, one event causes another to happen. It's a
direct cause-and-effect relationship.
Examples:
• Rain causes wet roads.
• A virus causes an illness.
• Studying hard causes better exam performance.
• Applying fertilizer causes plants to grow taller.
• Eating spoiled food causes food poisoning.
• A drop in interest rates causes more borrowing.
Note that Causation cannot be expressed as a single number. It must be established through evidence,
experiments, or deep analysis (e.g., randomized control trials, longitudinal studies).

Key differences between Correlation and Causation

Feature Correlation Causation

Statistical relationship between One variable directly produces
Definition variables change in another
Can it be measured Yes – by correlation coefficient No – cannot be summarized with a
numerically? (e.g., Pearson’s r) single number
Between −1 and +1 Not applicable
Range of values

Implied Direction of No direction implied Definite direction: cause → effect

Influence
Only data patterns Requires mechanism, control of
Evidence Required confounders, experimental proof
High – correlation may be due Low – causation must rule out other
Possibility of third factor? to a hidden variable explanations

Correlation does NOT imply causation

For example, In summer, sales of ice cream vs sales of cotton clothes.
In summer, both ice cream sales go up and cotton cloth sales go up.
But does eating ice cream cause cotton sales to go up? No!
The more likely explanation is that more people consume ice cream and also prefer to wear cotton cloth as it
is warmer outside in summer, which explains why these two variables are so highly correlated.
Although these two variables are correlated, one does not cause the other.
High School Graduates vs. Pizza Consumption
If we collect data for the total number of high school graduates and total pizza consumption in India each year,
we would find that the two variables are highly correlated.
54
 ISC XI- Applied Mathematics

Does this mean that an increased number of high school graduates is leading to more pizza consumption in
India? Not quite!
The more likely explanation is that India’s population has been increasing over time, which means that the
number of people receiving a high school degree and the total pizza being consumed are both increasing as
population increases.
Although these two variables are correlated, one does not cause the other.

Example of neither correlation nor causation

A classic example of neither correlation nor causation is the relationship between ice cream sales and homicide
rates in summer.
Correlation:
Ice cream sales and homicide rates tend to increase together in summer, suggesting a positive correlation.
Causation:
In summer, the increase in ice cream sales doesn't directly cause an increase in homicide rates, nor does the
increase in homicide rates cause an increase in ice cream sales.
Underlying Factor:
In summer, specifically high temperatures, is the third variable influencing both ice cream sales and homicide
rates. Warmer weather leads to more people outdoors, creating more opportunities for both ice cream
consumption and criminal activity.

Types and Degree of Correlation

Correlation can be either positive, negative, perfect or zero (no correlation).
• Positive Correlation:
When two variables move in the same direction, i.e., when one increases the other also increases and
vice-versa, then such a relation is called a Positive Correlation.
For example,
− As employment levels rise, output or production typically increases.
− With a rise in income, consumption expenditure usually increases as well.

• Negative Correlation:
When two variables move in opposite directions, i.e., when one increases the other decreases, and vice-
versa, then such a relation is called a Negative Correlation.
For example,
− An increase in the price of a commodity often leads to a decrease in the quantity demanded.
− A decrease in temperature immediately leads to an increased sale of woollen garments.

• Perfect Correlation:
If the relationship between the two variables is in such a way that it varies in equal proportion (increase
or decrease) it is said to be perfectly correlated. This can be of two types:
− Positive Correlation: When the proportional change in two variables is in the same direction, it is said
to be positively correlated. In this case, the Coefficient of Correlation is shown as +1.
− Negative Correlation: When the proportional change in two variables is in the opposite direction, it is
said to be negatively correlated. In this case, the Coefficient of Correlation is shown as -1.
• Zero Correlation:

55
 ISC XI- Applied Mathematics

If there is no relation between two series or variables, it is said to have zero or no correlation. It means
that if one variable varies and it does not have any impact on the other variable, then there is a lack of
correlation between them. In such cases, the Coefficient of Correlation will be 0.

• Limited Degree of Correlation:

There is a situation with a limited degree of correlation between perfect correlation and absence of
correlation. The coefficient of correlation, in this case, lies between +1 and -1.
Correlation is limited and negative when there are unequal changes in the opposite direction.
Correlation is limited and positive when there are unequal changes in the same direction.

Degree of Correlation
Degree of Correlation Positive Correlation Negative Correlation

Perfect +1 -1

Very High + 0.9 - 0.9

Fairly High Between + 0.75 and + 0.9 Between - 0.75 and - 0.9

Moderate Between + 0.25 and + 0.75 Between - 0.25 and - 0.75

Low Between 0 and + 0.25 Between 0 and - 0.25

Zero 0 0

Methods of Measurements of Correlation

The three different methods of measuring correlation between two variables are:
• Scatter Diagram
• Karl Pearson's Coefficient of Correlation
• Spearman's Rank Correlation Coefficient

Scatter Diagram
A simple and attractive method of measuring correlation by diagrammatically representing bivariate
distribution for determination of the nature of the correlation between the variables is known as the Scatter
Diagram Method. This method gives the investigator/analyst a visual idea of the nature of the association
between the two variables. The values of the two variables are drawn in a cartesian plane by points/dots. Then
by the scatter or closeness of the points the degree of the correlation between the variables can be discussed.

Interpretation of Scatter Diagram

After observing the pattern of dots, one can know the presence or absence of correlation and its type. Besides,
it also gives an idea of the nature and intensity of the relationship between the two variables.
The scatter diagram can be interpreted in the following ways:

56
 ISC XI- Applied Mathematics

Perfect Positive Correlation Perfect Negative Correlation

Positive Correlation

Negative Correlation

57
 ISC XI- Applied Mathematics

No Correlation

Merits of a Scatter Diagram

• Easy to draw.
• Easily understood and interpreted.
• Extreme values do not affect this method.

Demerits of a Scatter Diagram

• Only gives a visual picture of the relationship of the two variables.
• Does not give any idea about the precise degree of relationship.

Karl Pearson's Coefficient of Correlation

The first person to give a mathematical formula for the measurement of the degree of relationship between
two variables in 1890 was Karl Pearson, a prominent statistician. Karl Pearson's Coefficient of Correlation is
also known as Product Moment Correlation or Simple Correlation Coefficient. This method of measuring the
coefficient of correlation is the most popular and is widely used. It is denoted by 'r', where r is a pure number
which means that r has no unit. This a non-dimensional quantity which does not depends on dimensions.

Karl Pearson's Coefficient of Correlation and Covariance

Karl Pearson's method of determining coefficient of correlation is based on the covariance of the given
variables.
Karl Pearson’s coefficient of correlation between two quantitative variables X and Y is defined as:
𝐶𝐶𝐶𝐶v(𝑋𝑋,𝑌𝑌)
𝑟𝑟(𝑋𝑋, 𝑌𝑌) =
�𝑉𝑉𝑉𝑉𝑉𝑉 (𝑋𝑋)𝑉𝑉𝑉𝑉𝑉𝑉(𝑌𝑌)

The formula for calculating Karl Pearson's Coefficient of Correlation can be transformed into a working
formula:

𝛴𝛴𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 𝛴𝛴𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 1 1

𝑟𝑟 = 𝑛𝑛×𝜎𝜎 = × 𝜎𝜎 × 𝜎𝜎
𝑥𝑥 ×𝜎𝜎𝑦𝑦 𝑛𝑛 𝑥𝑥 𝑦𝑦

𝛴𝛴𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
⇒ 𝑟𝑟 =
2 2
𝑑𝑑 𝑑𝑑𝑦𝑦
𝑛𝑛 × � 𝑛𝑛𝑥𝑥 × � 𝑛𝑛
𝛴𝛴𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
∴ 𝑟𝑟 =
∑𝑑𝑑𝑥𝑥 2 × ∑𝑑𝑑𝑦𝑦 2
58
 ISC XI- Applied Mathematics

This method of determining Coefficient of Correlation should be applied only when the deviations of items
are taken from actual means and not from assumed means.
𝑟𝑟 2 , the square of the coefficient of correlation is called coefficient of determination. So, 0 ≤ 𝑟𝑟 2 ≤ 1.
Variation between X and Y is indicated by 𝑟𝑟 2 and not r.
For example, imagine you are studying the relationship between hours a student studies and marks scored in
an exam. You collect data and find that 𝑟𝑟 = 0 ⋅ 8 and 𝑟𝑟 2 = 0 ⋅ 64. As 𝑟𝑟 = 0 ⋅ 8 there is a strong positive
relationship between study time and exam [Link] more importantly, 𝑟𝑟 2 = 0 ⋅ 64 means that 64% of the
variation in exam marks is explained by the variation in study time. The remaining 36% variation is due to
other factors (like sleep, health, test anxiety, etc.).

Example 1. Determine the Coefficient of Correlation between X and X Y

Y if the summation of the product of deviations of Series X and Y from Number of items 30 30
their respective means is 200. Standard Deviation 4 3
Solution:
Given 𝑛𝑛 = 30, 𝜎𝜎𝑥𝑥 = 4, 𝜎𝜎𝑦𝑦 = 3, ∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 = 200

∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 200

𝑟𝑟 = = = 0.5
𝑛𝑛 × 𝜎𝜎𝑥𝑥 × 𝜎𝜎𝑦𝑦 30 × 4 × 3

Coefficient of Correlation = 0.5 (positive, moderate correlation)

Example 2. If the Covariance between two variables X and Y is 9.4 and the variance of Series X and Y are
10.6 and 12.5, respectively, then calculate the coefficient of correlation.
Solution:
𝛴𝛴𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
Cov(𝑋𝑋, 𝑌𝑌) = = 9.4
𝑛𝑛
Variance of X = 𝜎𝜎𝑥𝑥 2 = 10.6 ⇒ 𝜎𝜎𝑥𝑥 = 3 ⋅ 25
Variance of Y = 𝜎𝜎𝑦𝑦 2= 12.5 ⇒ 𝜎𝜎𝑦𝑦 = 3 ⋅ 53
∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 ∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 1 1 1 1
𝑟𝑟 = = × × = 9.4 × × = 0.816
𝑛𝑛 × 𝜎𝜎𝑥𝑥 × 𝜎𝜎𝑦𝑦 𝑛𝑛 𝜎𝜎𝑥𝑥 𝜎𝜎𝑦𝑦 3.25 3.53
Coefficient of Correlation = 0.816 (positive, fairly high correlation)

Example 3. If ∑𝑥𝑥𝑖𝑖 = 15, ∑𝑦𝑦𝑖𝑖 = 40, ∑(𝑥𝑥𝑖𝑖 − 2)(𝑦𝑦𝑖𝑖 − 3) = 20 and n = 10, find Cov(x,y).
Solution:
∑(𝑥𝑥𝑖𝑖 − 2)(𝑦𝑦𝑖𝑖 − 3) = 20
⇒ ∑xi yi − 3∑xi − 2Σyi + 6n = 20
⇒ ∑xi yi − 3 × 15 − 2 × 40 + 6 × 10 = 20
⇒ ∑xi yi = 85
1 1 1 1
Cov(𝑥𝑥, 𝑦𝑦) = �𝛴𝛴𝛴𝛴𝛴𝛴 − Σx∑y� = �85 − (15)(40)� = 2.5
𝑛𝑛 𝑛𝑛 10 10

Assumptions of Karl Pearson's Coefficient of Correlation

• Linear Relationship: The first assumption in this method is that there is a linear relationship between the
given two variables.
• Causal Relationship: Another assumption is that there is no cause-and-effect relationship between the
given two variables.

59
 ISC XI- Applied Mathematics

Methods of Calculating Karl Pearson's Coefficient of Correlation

• Actual Mean Method
• Direct Method
• Short-Cut Method/Assumed Mean Method
• Step-Deviation Method

Actual Mean Method

This method is used when both the actual means are integer values.

∑ ( x - x )( y - y ) ∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
r= =
∑𝑑𝑑𝑥𝑥 2 ×∑𝑑𝑑𝑥𝑥 2
∑ (x - x ) ∑(y - y)
2 2

Example 4. Using Actual Mean Method, determine the coefficient of correlation for the following data:

X 12 16 20 24 28 32 36
Y 6 9 12 15 18 21 24

Solution:

X �
𝑑𝑑𝑥𝑥 = 𝑿𝑿 − 𝑿𝑿 𝑑𝑑𝑥𝑥 2 Y 𝑑𝑑𝑦𝑦 � 𝑑𝑑𝑦𝑦 2
= 𝒀𝒀 − 𝒀𝒀 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
12 -12 144 6 -9 81 108
16 -8 64 9 -6 36 48
20 -4 16 12 -3 9 12
24 0 0 15 0 0 0
28 4 16 18 3 9 12
32 8 64 21 6 36 48
36 12 144 24 9 81 108
168 448 105 252 336

� = ∑X = 168 = 24, Y
X � = ∑Y = 105 = 15
n 7 n 7

∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 336

𝑟𝑟 = 2 2 = =1
∑𝑑𝑑𝑥𝑥 × ∑𝑑𝑑𝑦𝑦 √448 × 252

Coefficient of Correlation = 1 (positive, perfect correlation).

Direct Method
Though this method can be used for all situations, it is mainly used when the values of X and Y are small.

1
∑XY − n ΣX∑Y 𝑛𝑛∑XY − ΣX∑Y
𝒓𝒓 = =
�ΣX 2 − 1 (∑X)2 �∑Y 2 − 1 (∑Y)2 �nΣX 2 − (∑X)2 �n∑Y 2 − (∑Y)2
n n

60
 ISC XI- Applied Mathematics

Example 5. Using Direct Method, determine the coefficient of correlation for the following data:

X 5 4 3 2 1
Y 4 2 10 8 6
Solution:

X X2 Y Y2 XY
5 25 4 16 20
4 16 2 4 8
3 9 10 100 30
2 4 8 64 16
1 1 6 36 6
15 55 30 220 80

𝑛𝑛∑XY − ΣX∑Y 5 × 80 − (15)(30)

r= = = −0.5
�nΣX 2 − (∑X)2 �n∑Y 2 − (∑Y)2 �5 × 55 − (15)2 �5 × 220 − (30)2

Coefficient of Correlation = -0.5 (negative, moderate correlation)

Short-Cut Method/Assumed Mean Method

Actual Mean can sometimes come in fractions which can make the calculation of standard deviation
complicated and difficult. In those cases, it is suggested to use Short-Cut Method to simplify the calculations.

1
∑uv − n Σu∑v
𝒓𝒓 =
�Σ𝑢𝑢2 − 1 (∑u)2 �∑𝑣𝑣 2 − 1 (∑v)2
n n

where u = X – Assumed mean of X

v = Y – Assumed mean of Y

Any value can be considered as assumed mean but for ease of calculations the integer value closest to the
average of the highest and the lowest value for a particular variable is taken as the assumed mean.

Example 6. Using Short-Cut Method, determine the coefficient of correlation for the following data:

X 78 89 96 69 59 79 68 62
Y 125 137 156 112 107 136 123 104

Assume 80 and 130 as the mean values for x and y respectively.

Solution:

X u = 𝑿𝑿 − 𝟖𝟖𝟖𝟖 u2 Y v = 𝒀𝒀 − 𝟏𝟏𝟏𝟏𝟏𝟏 v2 uv
78 -2 4 125 -5 25 10
89 9 81 137 7 49 63
96 16 256 156 26 676 416
69 -11 121 112 -18 324 198
59 -21 441 107 -23 529 483
79 -1 1 136 6 36 -6
68 -12 144 123 -7 49 84
62 -18 324 104 -26 676 468
-40 1372 -40 2364 1716

61
 ISC XI- Applied Mathematics

1 1
∑uv − n Σu∑v 1716 − 8 (−40)(−40)
𝒓𝒓 = = = 0.95
�Σ𝑢𝑢2 1 1
− n (∑u)2 �∑𝑣𝑣 2 − n (∑v)2 �1372 − 1 (−40)2 �2364 − 1 (−40)2
8 8

Coefficient of Correlation = 0.95 (positive, very high correlation)

Example 7. Calculate the value of the correlation coefficient for the following data:
(1, 13), (2, 23), (3, 33), (4, 43), (5, 53), (6, 63), (7, 73), (8, 83), (9, 93), (10, 103), (11, 10.5), (12, 11), (13,
11.5), (14, 12), (15, 12.5), (16, 13), (17, 13.5), (18, 14), (19, 14.5), (20, 15)
Solution:
We take 10 and 13 as the assumed mean for X and Y respectively.

X 𝒖𝒖 = 𝑿𝑿 − 𝟏𝟏𝟏𝟏 u2 Y 𝒗𝒗 = 𝒀𝒀 − 𝟏𝟏𝟏𝟏 v2 uv
1 -9 81 13 0 0 0
2 -8 64 23 10 100 -80
3 -7 49 33 20 400 -140
4 -6 36 43 30 900 -180
5 -5 25 53 40 1600 -200
6 -4 16 63 50 2500 -200
7 -3 9 73 60 3600 -180
8 -2 4 83 70 4900 -140
9 -1 1 93 80 6400 -80
10 0 0 103 90 8100 0
11 1 1 10.5 -2.5 6.25 -2.5
12 2 4 11 -2 4 -4
13 3 9 11.5 -1.5 2.25 -4.5
14 4 16 12 -1 1 -4
15 5 25 12.5 -0.5 0.25 -2.5
16 6 36 13 0 0 0
17 7 49 13.5 0.5 0.25 3.5
18 8 64 14 1 1 8
19 9 81 14.5 1.5 2.25 13.5
20 10 100 15 2 4 20
10 447.5 28521.25 -1172.5

1 1
∑uv − n Σu∑v −1172.5 − 20 (10)(447.5)
𝒓𝒓 = = = −0.39798
�Σ𝑢𝑢2 1 1
− n (∑u)2 �∑𝑣𝑣 2 − n (∑v)2 �670 − 1 (10)2 �28521.25 − 1 (447.5)2
20 20

Coefficient of Correlation = −0.39798 (negative, moderate correlation)

Merits of Karl Pearson's Coefficient of Correlation

• Popular Method.
• Summarizes both the Degree and Direction of Correlation.

Demerits of Karl Pearson's Coefficient of Correlation

• Affected by Extreme Values
• Assumption of Linear Relationship
• Time-Consuming Calculation
• Possibility of Wrong Interpretation
62
 ISC XI- Applied Mathematics

Why study Spearman’s Rank Correlation Coefficient?

It is possible to avoid making any assumptions about the populations being studied by ranking the
observations according to size and basing the calculations on the ranks rather than upon the original
observations. It does not matter which way the items are ranked; item number one may be the largest or it
may be the smallest. Using ranks rather than actual observations gives the coefficient of rank correlations.
This method of finding out co-variability or the lack of it between two variables was developed by the
British Psychologist Charles Edward Spearman in 1904.

What is Ranking?
A ranking is a relationship between a set of items such that, for any two items, the first is either 'ranked
higher than', 'ranked lower than' or 'ranked equal to' the second.
In mathematics, this is known as a weak order or total pre order of objects. It is not necessarily a total order
of objects because two different objects can have the same ranking. The rankings themselves are totally
ordered.

What is Spearman’s Rank Correlation Coefficient?

Spearman’s Rank Correlation Coefficient or Spearman’s Rank Difference Method or Formula is a method of
calculating the correlation coefficient of qualitative variables. In other words, the formula determines the
correlation coefficient of variables like beauty, ability, honesty, etc., whose quantitative measurement is not
possible. Therefore, these attributes are ranked or put in the order of their preference.
Spearman’s correlation coefficient is defined as:

6∑𝐷𝐷 2
𝑅𝑅 = 1 −
𝑛𝑛(𝑛𝑛2 − 1)

In the given formula,

R = Coefficient of rank correlation
D = Rank differences
n = Number of variables

Types of Rank Methods

There are three different cases of Spearman’s Rank Correlation Coefficient:
(a) When ranks are given
(b) When ranks are not given
(c) When there are equal ranks

(a) When actual ranks are given

Example 8. 10 competitors in a competition were ranked by two judges X and Y in the following manner:

Ranked by X 1 6 5 10 3 2 4 9 7 8

Ranked by Y 6 4 9 8 1 2 3 10 5 7

Calculate the rank correlation coefficient.

63
 ISC XI- Applied Mathematics

Solution:

Ranked by X(R1) Ranked by Y(R2) D = R1 - R2 D2

1 6 -5 25
6 4 2 4
5 9 -4 16
10 8 2 4
3 1 2 4
2 2 0 0
4 3 1 1
9 10 -1 1
7 5 2 4
8 7 1 1
60

6∑𝐷𝐷 2 6×60
𝑅𝑅 = 1 − 𝑛𝑛(𝑛𝑛2−1) = 1 − 10(102−1) = 0.64 (positive, moderate correlation)

(b) When actual ranks are not given

Example 9. The marks obtained by nine students in Accounts and Commerce are given below:

Accounts marks 35 23 47 17 10 43 9 6 28
Commerce marks 30 33 45 23 8 49 12 4 31

Calculate the rank correlation coefficient and interpret the result.

Solution:
Accounts Commerce Rank in Rank in D = R1 - R2 D2
marks marks Accounts(R1) Commerce(R2)
35 30 3 5 -2 4
23 33 5 3 2 4
47 45 1 2 -1 1
17 23 6 6 0 0
10 8 7 8 -1 1
43 49 2 1 1 1
9 12 8 7 1 1
6 4 9 9 0 0
28 31 4 4 0 0
12

Method to Assign Ranks:

The smallest value is assigned Rank 1, the next value Rank 2 and so on…………

6∑𝐷𝐷 2 6×12
𝑅𝑅 = 1 − 𝑛𝑛(𝑛𝑛2−1) = 1 − 9(92−1) = 0.9 (positive, very high correlation)

This implies that a student scoring high marks in accounts is expected to score high marks in commerce as
well.

64
 ISC XI- Applied Mathematics

(c) When there are equal ranks

Sometimes where there is more than one item with the same value, a common rank is given to such items.
This rank is the average of the ranks which these items would have got had they differed slightly from each
other. When this is done, the coefficient of rank correlation needs some correction, because the above formula
is based on the supposition that the ranks of various items are different and that no rank is given to more than
one item.
If in a series, there are ‘m’ items whose ranks are common, then for correction of the coefficient of rank
1
correlation (𝑚𝑚3 − 𝑚𝑚) is added to the value of ∑𝐷𝐷 2. If there are more than one such groups of items with
12
common rank, this value is added as many times as the number of groups.

1
Correction factor = CF = ∑ (𝑚𝑚3 − 𝑚𝑚) … …
12

The formula for Spearman’s correlation coefficient is thus modified as follows:

6[∑𝐷𝐷2 + 𝐶𝐶𝐶𝐶]
𝑅𝑅 = 1 −
𝑛𝑛(𝑛𝑛2 − 1)
Example 10. Calculate the Rank Coefficient from the sales (in thousand units) and expenses (in thousand
Rupees) of 10 firms as given below:

Sales 50 50 55 60 65 65 65 60 60 50
Expenses 11 13 14 16 16 15 15 14 13 13

Solution:

Sales Expenses Rank in Rank in D = R1 - R2 D2

Sales(R1) Expenses(R2)
50 11 9 10 -1 1
50 13 9 8 1 1
55 14 7 5.5 1.5 2.25
60 16 5 1.5 3.5 12.25
65 16 2 1.5 0.5 0.25
65 15 2 3.5 -1.5 0.25
65 15 2 3.5 -1.5 2.25
60 14 5 5.5 -0.5 0.25
60 13 5 8 -3 9
50 13 9 8 1 1
31.5

Now,
1 1 1 1 1 1
CF = 12 (33 − 3) + 12
(33 − 3) +
12
(33 − 3) +
12
(23 − 2) +
12
(23 − 2) +
12
(23 − 2) +
1
(33 − 3) = 9.5
12

6[∑𝐷𝐷2 + 𝐶𝐶𝐶𝐶] 6[31.5 + 9.5] 6[31.5 + 9.5]

𝑅𝑅 = 1 − 2
=1− 2
=1− = 0.75
𝑛𝑛(𝑛𝑛 − 1) 10(10 − 1) 990

65
 ISC XI- Applied Mathematics

Example 11. The coefficient of rank correlation of the marks obtained by 10 students in Mathematics and
Statistics was found to be 0.5. It was then detected that the difference in ranks in the two subjects for one
particular student was wrongly taken to be 3 in place of 7. What should be the correct rank correlation
coefficient?
Solution:
n = 10, R = 0.5
wrong value of D = 3
correct value of D = 7

6∑𝐷𝐷2 6∑𝐷𝐷2
𝑅𝑅 = 1 − ⇒ 0.5 = 1 − ⇒ ∑𝐷𝐷2 = 82.5
𝑛𝑛(𝑛𝑛2 − 1) 10(102 − 1)

Corrected value for ∑𝐷𝐷 2 = 82.5 − 32 + 72 = 122.5

6 × 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑜𝑜𝑜𝑜 ∑𝐷𝐷 2 6 × 122.5

Hence, the corrected 𝑅𝑅 = 1 − = 1 − 10(102−1) = 0.258
𝑛𝑛(𝑛𝑛2 −1)

Merits of Spearman’s Rank Correlation Coefficient

• Ideal for Data where the exact values of the variables are not known, but the order of preference is known.
• Less sensitive to Outliers.
• Captures relationships even if the relationship is not linear.
• Applicable to a wider range of datasets, including those with skewed distributions.
• Easy to understand and apply.

Demerits of Spearman’s Rank Correlation Coefficient

• Not applicable for grouped data.
• Can handle only a limited number of observations or items.
• Only considers the ranks of the data points and ignores the actual magnitude of differences between the
values of the variables.
• May result in a loss of information in the data, especially if the variables of the data have meaningful
magnitudes or units.

66
 ISC XI- Applied Mathematics

REAL- LIFE APPLICATIONS

Real-World Applications of Correlation

• Finance:
− Identifying correlations between assets helps manage portfolio risk and diversification.
− Understanding correlations can improve financial forecasting and decision-making.
• Healthcare:
− Analyzing correlations between patient characteristics and health outcomes can lead to better
treatments.
− Correlation analysis can help identify factors that contribute to disease or health conditions.
• Social Sciences:
− Epidemiology
− Behavioral Research
• Engineering and Manufacturing:
− Identifying factors that affect product quality and improving processes.
− Correlating machine data with failure events to predict maintenance needs.
• Science:
− Environmental Research
− Data Analysis

Exercises
1. Match List-I with List-II:
List I List-II
(A) If you water your plants then they grow (I) Correlation but not Causation
(B) Shoe size of an employee and his salary (II) Causation
(C) The people who smoke and alcoholism. (III) Neither correlation nor Causation
Choose the correct answer from the options given below:
(a) (A) - (I), (B) - (II), (C) - (III)
(b) (A) - (II), (B) - (III), (C) - (I)
(c) (A) - (II), (B) - (I), (C) - (III)
(d) (A) - (III), (B) - (II), (C) - (I)

2. Which of the following is true?

Statement 1: Cool drinks sales and drowning incidents both increase in summer.
Statement 2: Injecting insulin lowers blood sugar in diabetic patients.
(a) Statement 1 is only causation.
(b) Statements 1 & 2 only causation
(c) Statement 1 only correlation and Statement 2 is only causation.
(d) Statement 1 only causation and Statement 2 is only correlation.

67
 ISC XI- Applied Mathematics

3. Which of the following is true?

Statement 1: Increasing fertilizer use and crop yield on a farm.
Statement 2: The number of books in a person’s home and their shoe brand preference.
(a) Statement 1 only correlation and Statement 2 is only causation.
(b) Statement 1 only causation and Statement 2 is only correlation.
(c) Statement 1 is correlation, also causation and Statement 2 is neither correlation nor causation.
(d) Statement 1 is neither correlation nor causation and Statement 2 is correlation, also causation.

4. For each of the following statements, state whether it describes a correlation, causation, or neither:
(i) When people increase the number of hours they exercise, they tend to lose weight.
(ii) A survey finds that students who eat breakfast regularly score higher on math tests.
(iii) Turning up the volume on your speaker makes the sound louder.
(iv) People who own more books tend to have higher income.
(v) A new medicine is found to directly lower blood pressure in patients.
(vi) It is observed that students who wear red shirts tend to perform better in exams.
(vii) As ice cream sales increase, the number of drowning incidents also increases.
(viii) Scientists discover that smoking damages lung tissues and leads to cancer.

5. Each situation describes a correlation. Suggest whether causation is likely or not and explain briefly.
(i) Students who spend more time studying tend to get better grades.
(ii) People who sleep less than 5 hours a night often reports higher levels of stress.
(iii) Cities with more police officers also have higher crime rates.
(iv) People who own luxury cars are more likely to take vacations abroad.

6. Choose the correct option:

(i) The coefficient of correlation is independent of
(a) change of scale but not of origin (c) change of scale and origin both
(b) change of origin but not of scale (d) neither change of scale nor of origin

(ii) If ∑𝑢𝑢𝑢𝑢 = 50 and n = 15 where u and v are deviations of X and Y series from their respective mean,
then Cov(X, Y) is
(a) 2.43 (c) 3.33
(b) 3.24 (d) 3.63

(iii) If the values of two variables move in opposite directions, their correlation is said to be
(a) linear (c) positive
(b) non-linear (d) negative

(iv) Consider the following statements:

Statement 1: If the dots in a scatter diagram fall on a narrow band, it indicates a low degree of
correlation.
Statement 2: Coefficient of correlation explains both the degree and direction of the relationship
between two variables.
Which of the statement(s) given above is/are true?
(a) Only 1 (c) Both 1 and 2
(b) Only 2 (d) Neither 1 nor 2

68
 ISC XI- Applied Mathematics

(v) Assertion: A perfect positive correlation means that all data points lie exactly on a straight line
sloping upward.
Reason: When r = 1, the values of the two variables increase in exact proportion to each other.
(a) Both A and R individually true and R is the correct explanation of A.
(b) Both A and R individually true but R is not the correct explanation of A.
(c) A is true but R is false.
(d) A is false but R is true.

7. Can you directly conclude that drinking more coffee causes better productivity just because they are
correlated? What else would you need to check?

8. Scientists found a correlation between kids who play a lot of video games and better hand-eye
coordination.
(i) Does this necessarily prove that playing video games causes better hand-eye coordination?
(ii) Suggest a way to find out if causation is true.

9. A study shows a correlation between eating more green vegetables and a lower risk of heart disease. How
could researchers test if this is truly causation?

10. Find the correlation coefficient if,

(i) 𝜎𝜎𝑥𝑥 = 6, 𝜎𝜎𝑦𝑦 = 10, cov(𝑥𝑥, 𝑦𝑦) = −30.
(ii) 𝑋𝑋 = 𝑥𝑥 − 𝑥𝑥̅ , 𝑌𝑌 = 𝑦𝑦 − 𝑦𝑦�, 𝑛𝑛 = 10, ∑𝑥𝑥 2 = 90, 𝜎𝜎𝑦𝑦 = 2.5, ∑𝑥𝑥𝑥𝑥 = 60

11. If r = 0.4, Cov(x, y) = 10, σy = 5, find the value of 𝜎𝜎𝑥𝑥 .

12. Calculate the coefficient of correlation by Karl Pearson’s method from the following data relating to
overhead expenses and cost of production.
Overhead in ’000 Rs 80 90 100 110 120 130 140 150 160
Cost in ’000 Rs 15 15 16 19 17 18 16 11 19

13. Calculate the Pearson coefficient of correlation between advertisement cost and sales as per the data
given below:
Advertisement cost 39 65 62 90 82 75 25 98 36 78
in ’000 Rs
Sales in lakhs Rs 47 53 58 86 62 68 60 91 51 84

14. A computer while calculating the correlation coefficient between two variables x and y obtained the
following results:
𝑛𝑛 = 30, ∑𝑥𝑥 = 120, ∑𝑦𝑦 = 90, ∑𝑥𝑥 2 = 600, ∑𝑦𝑦 2 = 250, ∑𝑥𝑥𝑥𝑥 = 356
It was, however, later discovered that two pairs of observations (8,10) and (12,7) had been wrongly
entered. The correct values were (8,12) and (10,8). Obtain the correct value of the correlation coefficient
between x and y.

15. Calculate the rank correlation coefficient for the following data:
x 92 89 87 86 83 77 71 63 53 50
y 86 83 91 77 68 85 52 82 37 57

69
 ISC XI- Applied Mathematics

16. In a contest, two judges ranked eight candidates in order of their preference as given below. Find the
rank correlation coefficient.
Judge 1 5 2 8 1 4 6 3 7
Judge 2 4 5 7 3 2 8 1 6

17. Calculate rank correlation coefficient from the following data:

Expenditure on 10 15 14 25 14 14 20 22
advertisement
Profit 6 25 12 18 25 40 10 7

18. The coefficient of rank correlation between the marks in Economics and Statistics obtained by a group
9
of students is and the sum of squares of the difference in rank is 30. Find the number of students in
11
the group.

Answers
1. (b) (A) - (II), (B) - (III), (C) - (I)
2. (c)Statement 1 only correlation and Statement 2 is only causation
3. (c)Statement 1 is correlation, also causation and Statement 2 is neither correlation nor causation.
4. (i) Correlation — Exercise and weight are related, but other factors may influence
weight loss.
(ii) Correlation — Breakfast and better scores are linked, but no proof breakfast causes better scores.
(iii) Causation — Direct cause-and-effect: turning up volume causes louder sound.
(iv) Correlation — Book ownership and income are related, but one doesn't
necessarily cause the other.
(v) Causation — Medicine directly causes a drop in blood pressure (assuming a properly done clinical
trial).
(vi) Correlation — Wearing red shirts and exam performance are just observed together; no proven
cause.
(vii) Correlation — Ice cream and drowning rise together (due to summer), but one does not cause the
other.
(viii) Causation — Smoking has been proven to cause lung cancer scientifically.
5. (i) Causation is likely — More studying probably helps grades, but need controlled
experiments to be sure.
(ii) Causation is possible — Lack of sleep may cause stress, but stress might also cause poor sleep;
more study needed.
(iii) Correlation only — High crime may lead to hiring more police; not that police cause crime.
(iv) Correlation only — Wealth might cause both luxury car ownership and more
vacations — a third factor (wealth).
6. (i) (c) change of scale and origin both
(ii) (c) 3.33
(iii) (d) negative
(iv) (b) only 2
(v) (a) Both A and R individually true and R is the correct explanation of A

70
 ISC XI- Applied Mathematics

7. No, you cannot directly conclude causation. You would need to check for:
• Other factors (like overall health, motivation, job type).
• Experimental studies or control groups.
8. (a) No, correlation does not prove causation.
(b) Conduct an experiment — randomly assign some kids to play video games and compare hand-eye
coordination over time.
9. Researchers could do a long-term controlled study, tracking people who eat more vs. fewer green
vegetables, while controlling other lifestyle factors like exercise, smoking, genetics, etc.
(Or conduct a randomized controlled trial, if ethical.)
10. (i) -0.5 (ii) 0.8
11. 5
12. 0.055
13. 0.78
14. 0.05
15. 0.73
16. 0.67
17. - 0.024
18. 10

Classroom Discussion
Give one example of your own where two things are correlated but not causally related. Explain the hidden
third factor that could cause both.

Project Work
Project Title: "Does Study Time Affect Academic Performance?"
Objective:
To investigate whether there is a statistical relationship between the number of hours students spend studying
and their academic performance (measured by test scores).
Data Collection:
• Participants: 30 students from the same grade or class.
• Variables:
− Independent Variable: Number of hours studied per week.
− Dependent Variable: Scores obtained in the most recent mathematics test.
• Tool: Use a questionnaire or Google Form to collect data.
Steps:
1. Collect Data:
Ask each student:
− How many hours they studied in the last week before the test.
− Their score on the test (out of 100).

71
 ISC XI- Applied Mathematics

2. Organize Data in a Table:

Student Hours Studied Test Score

A 4 55
B 6 70
... ... ...
− Calculate Correlation using Karl Pearson’s correlation coefficient.
3. Interpret Results:
− A positive correlation suggests more study leads to higher scores.
− A negative correlation means more study might not be helping.
− A zero correlation implies no relationship.
Presentation:
• Use graphs like scatter plots to visually represent the correlation.
• Provide a written analysis and conclusion.
• Reflect on possible reasons behind the results (motivation, distractions, study method).

HISTORICAL NOTE

• Auguste Bravais's work on the bivariate normal distribution in the mid-19th century laid a foundation
for understanding correlation.
• Francis Galton is credited with formally defining the concept of correlation in 1885. He defined
correlation as the relationship between two variables, where a change in one is associated with a
change in the other.
• Karl Pearson developed the Pearson correlation coefficient, a widely used statistical measure that
quantifies the strength and direction of a linear relationship between two variables.
• Charles Spearman introduced his rank correlation coefficient in [Link] method was particularly
useful when data did not meet the assumptions of parametric tests.

72
 ISC XI- Applied Mathematics

(iv) Linear Regression

SCOPE

• Lines of regression of x on y and y on x.

• Scatter diagrams.
• The method of least squares.
• Lines of best fit.
• Regression coefficient of x on y and y on x.
• b xy × b yx = r 2 , 0 ≤ b xy × b yx ≤ 1
• Identification of regression equations.
• Properties of regression lines.
• Estimation of the value of one variable using the value of other variable from appropriate line
of regression.

LEARNING OUTCOMES

Students will be able to:

• derive regression equation of x on y and y on x.
• identify the appropriate regression equation to estimate the unknown variable.

PRE-REQUISITES

• Organizing and presenting data using tables, graphs, and charts.

• Concept of how data is analyzed and interpreted, including the concept of variability.
• Mean and Standard Deviation
• Correlation Analysis

73
 ISC XI- Applied Mathematics

Introduction
The dictionary meaning of the word regression is ‘stepping or going back’. The use of this word dates to the
times of early studies made by Francis Galton in the later half of the nineteenth century. He studied the
relationship between the heights of fathers and sons and arrived at the following conclusions:
(i) Tall fathers have tall sons and short fathers have short sons.
(ii) If fathers are tall, mean height of the sons < mean height of fathers.
(iii) If fathers are short, mean height of the sons > mean height of fathers.
(iv) Deviations in the mean height of sons from the mean height of the race < Deviations in the mean
height of fathers from the mean height of the race.
So, he concluded that when the fathers move above the mean or below the mean, the sons tend to go back or
regress towards the mean.
Galton studied the average relationship between two variables graphically and called the line describing the
relationship the line of regression.

What is Regression?
When two or more variables are correlated, a
statistical method to formulate an algebraic
relationship between two or more variables in the
form of equation to estimate the value of one
variable when the values of all other variables are
known or specified, is called Regression.
The variable whose value is predicted is called
response or dependent variable. The variables
used for prediction the response variable or
dependent variable is called predictors or
independent variables.
A regression shows the changes in a dependent
variable on the y-axis to the changes in the
independent variable on the x-axis. Basically,
regression is the task of predicting a continuous
quantity.

What is Regression Analysis?

Regression analysis is a form of predictive modelling technique which investigates the relationship between a
dependent variable and an independent variable. It involves graphing a line over a set of data points that most
closely fits the overall shape of the data.
There are many regression analysis techniques, but the three most widely used regression models are:
1. Linear Regression
2. Logistic Regression
3. Polynomial Regression

74
 ISC XI- Applied Mathematics

What Is Linear Regression?

Linear regression is a predictive modelling technique used to predict
a continuous dependent variable, given one or more independent
variables. A model with exactly one independent variable is a simple
linear regression.
In a linear regression model, the relationship between the dependent
and independent variable is always linear, Thus, when we try to plot
their relationship, we observe more of a straight line than a curved
one.

Comparison between Correlation and Regression

Correlation Regression
Regression is used to numerically describe how a
Correlation is used to determine whether variables
dependent variable changes with a change in an
are related or not.
independent variable
Correlation tries to establish a linear relationship It finds the best-fitted regression line to estimate an
between variables. unknown variable on the basis of the known variable.

The variables can be used interchangeably The variables cannot be interchanged.

Correlation uses a signed numerical value to
Regression is used to show the impact of a unit change
estimate the strength of the relationship between
in the independent variable on the dependent variable.
the variables.

Methods of studying Regression

Regression can be studied either:
(i) Graphically
(ii) Algebraically

Graphical study of Regression – Scatter diagrams

A scatter diagram contains one point for one pair of values of x and y variables. The x variable(independent)
is represented on the horizontal axis and the y variable(dependent) is represented on the vertical axis. When
all related pairs have been plotted on the scatter diagram, we visualize a smooth curve approximating the data.
This is called curve fitting.

In our scope of study, we will be dealing only with Linear Regression. The main purpose of curve fitting is
to predict or estimate the value of the dependent variable from the value of the independent variable. The
regression line which is used to predict the values of y for the given values of x is called the regression line of
y on x and the regression line which is used to predict the values of x for the given values of y is called the
regression line of x on y.

75
 ISC XI- Applied Mathematics

Regression lines can be drawn by free hand curve method but it is a very cumbersome and difficult method.
It is also dependent on individual judgement and different observers will obtain different curves and equations.

Algebraic study of Regression- the method of Least Squares

In the method of Least Squares, we minimise
the sum of squares of the deviations between
the given values of a variable and its estimated
values given by the best fitting curve. In linear
regression, the line of best fit is a straight line
as shown in the given diagram.
The given data points are minimized by the
method of reducing deviations or offsets of
each point from the line. The vertical offsets
are generally used in surface, polynomial and
hyperplane problems, while perpendicular
offsets are utilized in common practice. For
any point that is fitted accurately, its
perpendicular offset is zero.
If the line of regression is so chosen that the sum
of squares of deviations parallel to the y-axis is
minimised, it is called the regression line of y
on x. Its equation is 𝑦𝑦 = 𝑎𝑎 + 𝑏𝑏𝑏𝑏. The slope of
the line, b, is called the regression coefficient
of y on x and is represented as 𝑏𝑏𝑦𝑦𝑦𝑦 . (Fig. A)

If the line of regression is so chosen that the sum of squares of deviations parallel to the x-axis is minimised,
it is called the regression line of x on y. Its equation is 𝑥𝑥 = 𝑐𝑐 + 𝑑𝑑𝑑𝑑 and b is called the regression coefficient
of x on y and is represented as 𝑏𝑏𝑥𝑥𝑥𝑥 . (Fig. B)

Fig . A Fig. B

For y on x, the equation of least square line is given by 𝑦𝑦 = 𝑎𝑎 + 𝑏𝑏𝑏𝑏. The corresponding normal equations
are:
∑𝑦𝑦 = 𝑛𝑛𝑛𝑛 + 𝑏𝑏∑𝑥𝑥
∑𝑥𝑥𝑥𝑥 = 𝑎𝑎∑𝑥𝑥 + 𝑏𝑏∑𝑥𝑥 2

Solving these two normal equations we can get the required trend line equation.
76
 ISC XI- Applied Mathematics

Thus, we can get the line of best fit with formula 𝑦𝑦 = 𝑎𝑎 + 𝑏𝑏𝑏𝑏.
For x on y, the equation of least square line is given by 𝑥𝑥 = 𝑐𝑐 + 𝑑𝑑𝑑𝑑. The corresponding normal equations
are:
∑𝑥𝑥 = 𝑛𝑛𝑛𝑛 + 𝑑𝑑∑𝑦𝑦
∑𝑥𝑥𝑥𝑥 = 𝑐𝑐∑𝑦𝑦 + 𝑑𝑑∑𝑦𝑦 2

Solving these two normal equations we can get the required trend line equation.
Thus, we can get the line of best fit with formula 𝑥𝑥 = 𝑐𝑐 + 𝑑𝑑𝑑𝑑.

Example 1. Fit the regression lines of y on x and x on y for the following data.

x 2 4 6 8 10
y 5 7 9 8 11

Solution:

x y x2 y2 xy
2 5 4 25 10
4 7 16 49 28
6 9 36 81 54
8 8 64 64 64
10 11 100 121 110
∑𝑥𝑥 =30 ∑𝑦𝑦 =40 ∑𝑥𝑥 2 =220 ∑𝑦𝑦 2 =340 ∑𝑥𝑥𝑥𝑥 = 266

∑𝑦𝑦 = 𝑛𝑛𝑛𝑛 + 𝑏𝑏∑𝑥𝑥 ⇒ 40 = 5𝑎𝑎 + 30𝑏𝑏

∑𝑥𝑥𝑥𝑥 = 𝑎𝑎∑𝑥𝑥 + 𝑏𝑏∑𝑥𝑥 2 ⇒ 266 = 30𝑎𝑎 + 220𝑏𝑏
Solving we get, a = 4.1 and b = 0.65.
∴ 𝑦𝑦 = 4.1 + 0.65𝑥𝑥 is the y on x regression line.
∑𝑥𝑥 = 𝑛𝑛𝑛𝑛 + 𝑑𝑑∑𝑦𝑦 ⇒ 30 = 5𝑐𝑐 + 40𝑑𝑑
∑𝑥𝑥𝑥𝑥 = 𝑐𝑐∑𝑦𝑦 + 𝑑𝑑∑𝑦𝑦 2 ⇒ 266 = 40𝑐𝑐 + 340𝑑𝑑
Solving we get, c = - 4.4 and d = 1.3.
∴ 𝑥𝑥 = −4.4 + 1.3𝑦𝑦 is the x on y regression line.

77
 ISC XI- Applied Mathematics

Why two regression lines?

In general, there are two regression equations unless there
exists a perfect linear relationship between x and y. This is
because two regression equations are derived under two
different conditions. The regression equation of y on x is
obtained by minimising the sum of squares of deviations
parallel to the y-axis and the regression equation of x on y is
obtained by minimising the sum of squares of deviations
parallel to the x-axis.
In linear regression, the choice between a regression line of
y on x and x on y depends on which variable is given and
which is being estimated. The regression line of y on x is
used when we want to estimate the value of y given a value
of x, while the regression line of x on y is used when we
want to estimate the value of x given a value of
y. Essentially, the independent variable (the value which is given) goes on the x-axis, and the dependent
variable (the value which is estimated) goes on the y-axis.

Regression Coefficients
Solving the normal equations,

∑𝑦𝑦 = 𝑛𝑛𝑛𝑛 + 𝑏𝑏∑𝑥𝑥 and ∑𝑥𝑥𝑥𝑥 = 𝑎𝑎∑𝑥𝑥 + 𝑏𝑏∑𝑥𝑥 2

∑𝑥𝑥𝑥𝑥 ∑𝑥𝑥 ∑𝑦𝑦

𝑛𝑛∑𝑥𝑥𝑥𝑥 − ∑𝑥𝑥∑𝑦𝑦 − � �� � 𝐶𝐶𝐶𝐶𝐶𝐶(𝑥𝑥,𝑦𝑦) 𝜎𝜎𝑦𝑦
we get 𝑏𝑏 = = 𝑛𝑛
∑𝑥𝑥2
𝑛𝑛 𝑛𝑛
∑𝑥𝑥 2
= = 𝑟𝑟 𝜎𝜎
𝑛𝑛∑𝑥𝑥 2 − (∑𝑥𝑥)2 −� � 𝜎𝜎𝑥𝑥2 𝑥𝑥
𝑛𝑛 𝑛𝑛

Since, 𝑏𝑏𝑦𝑦𝑦𝑦 = 𝑏𝑏 the regression equation of y on x can be written as 𝑦𝑦 − 𝑦𝑦� = 𝑏𝑏𝑦𝑦𝑦𝑦 (𝑥𝑥 − 𝑥𝑥̅ ).

Similarly, solving the normal equations,

∑𝑋𝑋 = 𝑛𝑛𝑛𝑛 + 𝑏𝑏∑𝑌𝑌 and ∑𝑋𝑋𝑋𝑋 = 𝑎𝑎∑𝑌𝑌 + 𝑏𝑏∑𝑌𝑌 2

∑𝑥𝑥𝑥𝑥 ∑𝑥𝑥 ∑𝑦𝑦

𝑛𝑛∑𝑥𝑥𝑥𝑥 − ∑𝑥𝑥∑𝑦𝑦 − � �� � 𝐶𝐶𝐶𝐶𝐶𝐶(𝑥𝑥,𝑦𝑦) 𝜎𝜎
we get 𝑏𝑏 = = 𝑛𝑛
∑𝑦𝑦2
𝑛𝑛 𝑛𝑛
∑𝑦𝑦 2
= = 𝑟𝑟 𝜎𝜎𝑥𝑥
𝑛𝑛∑𝑦𝑦 2 − (∑𝑦𝑦)2 −� � 𝜎𝜎𝑦𝑦2 𝑦𝑦
𝑛𝑛 𝑛𝑛

Since, 𝑏𝑏𝑥𝑥𝑥𝑥 = 𝑏𝑏 the regression equation of x on y can be written as 𝑥𝑥 − 𝑥𝑥̅ = 𝑏𝑏𝑥𝑥𝑥𝑥 (𝑦𝑦 − 𝑦𝑦�)

Let 𝑑𝑑𝑥𝑥 = 𝑥𝑥 − 𝑥𝑥̅ and 𝑑𝑑𝑦𝑦 = 𝑦𝑦 − 𝑦𝑦� be the deviations of the variables x and y from the actual mean, then
∑𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 = ∑(𝑥𝑥 − 𝑥𝑥̅ )(𝑦𝑦 − 𝑦𝑦�) = ∑𝑥𝑥𝑥𝑥 − 𝑥𝑥̅ (𝑦𝑦1 +𝑦𝑦2 + ⋯ + 𝑦𝑦𝑛𝑛 ) − 𝑦𝑦�(𝑥𝑥1 + 𝑥𝑥2 + ⋯ + 𝑥𝑥𝑛𝑛 ) + 𝑛𝑛𝑥𝑥̅ 𝑦𝑦�
= (∑𝑥𝑥𝑥𝑥 + 𝑛𝑛𝑥𝑥̅ 𝑦𝑦�) − 𝑥𝑥.
� 𝑛𝑛𝑦𝑦� − 𝑦𝑦.
� 𝑛𝑛𝑥𝑥̅
= ∑𝑥𝑥𝑥𝑥 + 𝑛𝑛𝑥𝑥̅ 𝑦𝑦� − 2𝑛𝑛𝑥𝑥̅ 𝑦𝑦� = ∑𝑥𝑥𝑥𝑥 − 𝑛𝑛𝑥𝑥̅ 𝑦𝑦�
∑𝑥𝑥 ∑𝑦𝑦 ∑𝑥𝑥∑𝑦𝑦
= ∑𝑥𝑥𝑥𝑥 − 𝑛𝑛. ⋅ = ∑𝑥𝑥𝑥𝑥 −
𝑛𝑛 𝑛𝑛 𝑛𝑛

(∑𝑥𝑥)2 (∑𝑦𝑦)2
Also, 𝑛𝑛𝜎𝜎𝑥𝑥2 = ∑𝑥𝑥 2 − and 𝑛𝑛𝜎𝜎𝑦𝑦2 = ∑𝑦𝑦 2 −
𝑛𝑛 𝑛𝑛

78
 ISC XI- Applied Mathematics

∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 ∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦

Then, we get 𝑏𝑏𝑦𝑦𝑦𝑦 = and 𝑏𝑏𝑥𝑥𝑥𝑥 =
𝑛𝑛𝜎𝜎𝑥𝑥2 𝑛𝑛𝜎𝜎𝑦𝑦2
Or we can say,

∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 ∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦

𝑏𝑏𝑦𝑦𝑦𝑦 = and 𝑏𝑏𝑥𝑥𝑥𝑥 =
∑ 𝑑𝑑𝑥𝑥2 2
∑ 𝑑𝑑𝑦𝑦

Properties of Regression Lines

1. The point of intersection of the two regression lines
represents the means of the x and y variables. This point is
denoted as (𝑥𝑥̅ , 𝑦𝑦�) where 𝑥𝑥̅ is the mean of the x values and
𝑦𝑦� is the mean of the y values.
2. The acute angle 𝜃𝜃 between two regression lines is
given by
1 − 𝑟𝑟 2
tan 𝜃𝜃 = � �
𝑏𝑏𝑦𝑦𝑦𝑦 + 𝑏𝑏𝑥𝑥𝑥𝑥

3. If two lines of regression, when plotted on a graph paper, coincide, there is a perfect correlation between
the two series.

Perfect Positive Correlation Perfect Negative Correlation

When 𝒓𝒓 = ±𝟏𝟏, 𝒕𝒕𝒕𝒕𝒕𝒕 𝜽𝜽 = 𝟎𝟎. This implies θ is either 0 or π. Hence, the lines of regression coincide and
there is perfect correlation between the variables x and y.
4. Greater the angle between the regression lines, lesser is the correlation between the two series.

High Degree Positive Correlation High Degree Negative Correlation

79
 ISC XI- Applied Mathematics

Low Degree Negative Correlation

Low Degree Positive Correlation

5. If the two regression lines cut each other at right angles there is no
correlation between them.

𝝅𝝅
When 𝒓𝒓 = 𝟎𝟎, 𝜽𝜽 = . Hence, the estimated value of y will be the same for
𝟐𝟐
all values of x and vice- versa.

Properties of Regression Coefficients

1. Correlation coefficient is the geometric mean between the regression coefficients.

𝑟𝑟 = �𝑏𝑏𝑦𝑦𝑦𝑦 . 𝑏𝑏𝑥𝑥𝑥𝑥
2. If one of the regression coefficients is greater than unity, then the other must be less than unity numerically.
0 ≤ 𝑏𝑏𝑦𝑦𝑦𝑦 ⋅ 𝑏𝑏𝑥𝑥𝑥𝑥 ≤ 1
3. Arithmetic mean of regression coefficients is greater than the correlation coefficient.
𝑏𝑏𝑦𝑦𝑦𝑦 + 𝑏𝑏𝑥𝑥𝑥𝑥 > 2𝑟𝑟
4. The correlation coefficient and the two regression coefficients have the same sign.
5. Regression coefficients are independent of the change in origin but are dependent on the change in scale.

Working formulas for regression coefficients

If x, y are small numbers,
1 1
∑𝑥𝑥𝑥𝑥− ∑𝑥𝑥∑𝑦𝑦 ∑𝑥𝑥𝑥𝑥− ∑𝑥𝑥∑𝑦𝑦
𝑏𝑏𝑦𝑦𝑦𝑦 = 𝑛𝑛
1 and 𝑏𝑏𝑥𝑥𝑥𝑥 = 𝑛𝑛
1
∑𝑥𝑥 2 − (∑𝑥𝑥)2 ∑𝑦𝑦 2 − (∑𝑦𝑦)2
𝑛𝑛 𝑛𝑛

If 𝑥𝑥 − 𝑥𝑥̅ , 𝑦𝑦 − 𝑦𝑦� are small integer numbers,

∑(𝑥𝑥−𝑥𝑥̅ )(𝑦𝑦−𝑦𝑦�) ∑(𝑥𝑥−𝑥𝑥̅ )(𝑦𝑦−𝑦𝑦�)
𝑏𝑏𝑦𝑦𝑦𝑦 = and 𝑏𝑏𝑥𝑥𝑥𝑥 =
∑(𝑥𝑥−𝑥𝑥̅ )2 ∑(𝑦𝑦−𝑦𝑦�)2

Otherwise, we use assumed means A and B such that 𝑢𝑢 = 𝑥𝑥 − 𝐴𝐴, 𝑣𝑣 = 𝑦𝑦 − 𝐵𝐵,

1 1
∑𝑢𝑢𝑢𝑢− ∑𝑢𝑢∑𝑣𝑣 ∑𝑢𝑢𝑢𝑢− ∑𝑢𝑢∑𝑣𝑣
𝑏𝑏𝑦𝑦𝑦𝑦 = 𝑛𝑛
1 and 𝑏𝑏𝑥𝑥𝑥𝑥 = 𝑛𝑛
1
∑𝑢𝑢2 − (∑𝑢𝑢)2 ∑𝑣𝑣 2 − (∑𝑣𝑣)2
𝑛𝑛 𝑛𝑛

80
 ISC XI- Applied Mathematics

Example 2. From the following data find the two regression lines and predict the value of y when x= 2.5
x 1 2 3 4 5
y 2 3 5 4 6

Solution:
Method 1:

x y x2 y2 xy
1 2 1 4 2
2 3 4 9 6
3 5 9 25 15
4 4 16 16 16
5 6 25 36 30
15 20 55 90 69

∑𝑥𝑥 15 ∑𝑦𝑦 20
𝑥𝑥̅ = = = 3, 𝑦𝑦� = = =4
𝑛𝑛 5 𝑛𝑛 5
1 1
∑𝑥𝑥𝑥𝑥 − 𝑛𝑛 ∑𝑥𝑥∑𝑦𝑦 69 − 5 (15)(20)
𝑏𝑏𝑦𝑦𝑦𝑦 = = = 0.9
1 1
∑𝑥𝑥 − (∑𝑥𝑥)
2 2 55 − (15) 2
𝑛𝑛 5
1 1
∑𝑥𝑥𝑥𝑥 − 𝑛𝑛 ∑𝑥𝑥∑𝑦𝑦 69 − 5 (15)(20)
𝑏𝑏𝑥𝑥𝑥𝑥 = = = 0.9
1 1
∑𝑦𝑦 2 − 𝑛𝑛 (∑𝑦𝑦)2 90 − (20)2
5

Method 2:

x y 𝒅𝒅𝒙𝒙 𝒅𝒅𝒚𝒚 𝒅𝒅𝒙𝒙 𝒅𝒅𝒚𝒚 𝒅𝒅𝒙𝒙 ² 𝒅𝒅𝒚𝒚 ²

1 2 -2 -2 4 4 4
2 3 -1 -1 1 1 1
3 5 0 1 0 0 1
4 4 1 0 0 1 0
5 6 2 2 4 4 4
15 20 9 10 10

∑𝑥𝑥 15 ∑𝑦𝑦 20
𝑥𝑥̅ = = = 3, 𝑦𝑦� = = =4
𝑛𝑛 5 𝑛𝑛 5
∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 ∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦
The actual mean are integer values, hence we use 𝑏𝑏𝑦𝑦𝑦𝑦 = and 𝑏𝑏𝑥𝑥𝑥𝑥 = .
∑ 𝑑𝑑𝑥𝑥2 2
∑ 𝑑𝑑𝑦𝑦
∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 9
𝑏𝑏𝑦𝑦𝑦𝑦 = == 0.9
∑ 𝑑𝑑𝑥𝑥2 10
∑ 𝑑𝑑𝑥𝑥 𝑑𝑑𝑦𝑦 9
𝑏𝑏𝑥𝑥𝑥𝑥 = 2 = 10 = 0.9
∑ 𝑑𝑑𝑦𝑦
The regression equation of y on x is 𝑦𝑦 − 𝑦𝑦� = 𝑏𝑏𝑦𝑦𝑦𝑦 (𝑥𝑥 − 𝑥𝑥̅ ) ⇒ 𝑦𝑦 − 4 = 0.9(𝑥𝑥 − 3)
⇒ 𝑦𝑦 = 0.9𝑥𝑥 + 1.3
The regression equation of x on y is 𝑥𝑥 − 𝑥𝑥̅ = 𝑏𝑏𝑥𝑥𝑥𝑥 (𝑦𝑦 − 𝑦𝑦�) ⇒ 𝑥𝑥 − 3 = 0.9(𝑦𝑦 − 4)
⇒ 𝑥𝑥 = 0.9𝑦𝑦 − 0.6
When x = 2.5, from 𝑦𝑦 = 0.9𝑥𝑥 + 1.3 we get y = 3.55
81
 ISC XI- Applied Mathematics

Example 3. The grade of 9 students for the half yearly and the annual examination in Mathematics are as
follows:
x 77 50 71 72 81 94 96 99 67
y 82 66 78 84 47 85 99 99 68

Find a suitable linear regression equation for the data and then estimate the grade in the annual examination
for a student who received 85 in half yearly but was sick at the time of annual examinations and hence was
absent.
Solution:

x y u = x -72 v = y - 78 𝒖𝒖𝟐𝟐 uv
77 82 5 4 25 20
50 66 - 22 -12 484 264
71 78 -1 0 1 0
72 84 0 6 0 0
81 47 9 - 31 81 -279
94 85 22 7 484 154
96 99 24 21 576 504
99 99 27 21 729 567
67 68 -5 -10 25 50
707 708 59 6 2405 1280

∑x 707 ∑y 708
x� = n
= = 78.56, y� = n = 9 = 78.67
9
1 1
∑𝑢𝑢𝑢𝑢 − 𝑛𝑛 ∑𝑢𝑢∑𝑣𝑣 1280 − 9 (59)(6)
𝑏𝑏𝑦𝑦𝑦𝑦 = = = 0.615
1 1
∑𝑢𝑢2 − 𝑛𝑛 (∑𝑢𝑢)2 2405 − 𝑛𝑛 (59)2

The regression equation of y on x is 𝑦𝑦 − 𝑦𝑦� = 𝑏𝑏𝑦𝑦𝑦𝑦 (𝑥𝑥 − 𝑥𝑥̅ ) ⇒ 𝑦𝑦 − 78.67 = 0.615(𝑥𝑥 − 78.56)
⇒ 𝑦𝑦 = 0.615𝑥𝑥 + 30.36
When x = 85, we get y = 82.635≈ 83.
His Mathematics grade in the annual examination will be 83(approx.)

Example 4. The coefficient of correlation between the ages of husbands and wives in a community was found
to be 0.8; the mean of the husband’s age was 25 years and that of the wives 22 years. Their standard deviations
were 4 and 5 years respectively. Find the two lines of regression and hence obtain:
(a) Husband’s age when wife’s age is 22.
(b) Wife’s age when husband’s age is 33.
Solution: r = 0 ⋅ 8, x� = 25, y
� = 22, σx = 4, σy = 5
𝜎𝜎𝑦𝑦 5
𝑏𝑏𝑦𝑦𝑦𝑦 = 𝑟𝑟 = 0.8 × = 1
𝜎𝜎𝑥𝑥 4
𝜎𝜎𝑥𝑥 4
𝑏𝑏𝑥𝑥𝑥𝑥 = 𝑟𝑟 = 0.8 × = 0.64
𝜎𝜎𝑦𝑦 5

The regression equation of y on x is 𝑦𝑦 − 𝑦𝑦� = 𝑏𝑏𝑦𝑦𝑦𝑦 (𝑥𝑥 − 𝑥𝑥̅ ) ⇒ 𝑦𝑦 − 22 = 1(𝑥𝑥 − 25)

⇒ 𝑦𝑦 = 𝑥𝑥 − 3
When x = 33, we get y = 33 – 3 = 30 years

82
 ISC XI- Applied Mathematics

The regression equation of x on y is 𝑥𝑥 − 𝑥𝑥̅ = 𝑏𝑏𝑥𝑥𝑥𝑥 (𝑦𝑦 − 𝑦𝑦�) ⇒ 𝑥𝑥 − 25 = 0.64(𝑦𝑦 − 22)

⇒ 𝑥𝑥 = 0.64𝑦𝑦 + 10.92
When y = 32, we get x = 25 years.
(a) When wife’s age is 22, husband’s age is 25 years.
(b) When husband’s age is 33, wife’s age is 30 years.

Example 5. The following results were obtained from the record of age(x) and blood pressure(y) of a group
of 10 women.
x y
Mean 53 142
Variance 130 165

Given 𝛴𝛴(𝑥𝑥 − 𝑥𝑥̅ )(𝑦𝑦 − 𝑦𝑦�) = 1220, find the regression equation of y on x and use it to estimate the blood
pressure of a 47-year-old woman.
Solution:

1 1
∑(𝑥𝑥 − 𝑥𝑥̅ )(𝑦𝑦 − 𝑦𝑦�) 𝑛𝑛 ∑(𝑥𝑥 − 𝑥𝑥̅ )(𝑦𝑦 − 𝑦𝑦�) 10 × 1220
𝑏𝑏𝑦𝑦𝑦𝑦 = = = = 0 ⋅ 938
∑(𝑥𝑥 − 𝑥𝑥̅ )2 1 130
) 2
𝑛𝑛 ∑(𝑥𝑥 − 𝑥𝑥̅

∴𝑦𝑦 − 𝑦𝑦� = 𝑏𝑏𝑦𝑦𝑦𝑦 (𝑥𝑥 − 𝑥𝑥̅ ) ⇒ 𝑦𝑦 − 142 = 0 ⋅ 938(𝑥𝑥 − 53) ⇒ 𝑦𝑦 = 0 ⋅ 938𝑥𝑥 + 92 ⋅ 29

When x = 47, y = 136.376 ≈ 136
The blood pressure of a 47-year-old woman will be 136(approx.)

Example 6. If two regression lines of a bivariate distribution are 4𝑥𝑥 − 5𝑦𝑦 + 33 = 0 and 20𝑥𝑥 −
9𝑦𝑦 − 107 = 0, calculate
(i) 𝑥𝑥̅ and 𝑦𝑦� , the arithmetic means of x and y respectively.
(ii) Estimate the value of x when y = 7.
(iii) Find the variance of y when 𝜎𝜎𝑥𝑥 = 3.

Solution:

(i) The given regression lines are

4𝑥𝑥 − 5𝑦𝑦 + 33 = 0 and 20𝑥𝑥 − 9𝑦𝑦 − 107 = 0
The regression lines intersect at the point (𝑥𝑥̅ , 𝑦𝑦�).
Solving the two equations we get 𝑥𝑥̅ = 13, 𝑦𝑦� = 17.
(ii) Let 4𝑥𝑥 − 5𝑦𝑦 + 33 = 0 be the regression line of y on x.
4 33 4
Rewriting, 𝑦𝑦 = 𝑥𝑥 + ⇒ 𝑏𝑏𝑦𝑦𝑦𝑦 = .
5 5 5

Let 20𝑥𝑥 − 9𝑦𝑦 − 107 = 0 be the regression line of x on y.

9 107 9
Rewriting, 𝑥𝑥 = 𝑦𝑦 + ⇒ 𝑏𝑏𝑥𝑥𝑥𝑥 =
20 20 20
4 9 9
Now, 𝑏𝑏𝑦𝑦𝑦𝑦 ⋅ 𝑏𝑏𝑥𝑥𝑥𝑥 = × = 25 < 1
5 20
83
 ISC XI- Applied Mathematics

which means our assumption is correct.

20𝑥𝑥 − 9𝑦𝑦 − 107 = 0 is the regression line of x on y.
9 107
When y = 7, from 𝑥𝑥 = 𝑦𝑦 + we get x = 8.5.
20 20

(iii) Given 𝜎𝜎𝑥𝑥 = 3, then 𝜎𝜎𝑥𝑥2 = 9.

Since, 𝑏𝑏𝑥𝑥𝑥𝑥 , 𝑏𝑏𝑦𝑦𝑦𝑦 is positive so

4 9 3
𝑟𝑟 = +�𝑏𝑏𝑦𝑦𝑦𝑦 ⋅ 𝑏𝑏𝑥𝑥𝑥𝑥 = +� × =+
5 20 5

𝜎𝜎𝑦𝑦 3 𝜎𝜎𝑦𝑦 4 3 𝜎𝜎𝑦𝑦

𝑏𝑏𝑦𝑦𝑦𝑦 = 𝑟𝑟 = × ⇒ = × ⇒ 𝜎𝜎𝑦𝑦 = 4
𝜎𝜎𝑥𝑥 5 3 5 5 3

Variance of y = 𝜎𝜎𝑦𝑦2 = 16.

Example 7. Calculate from the equations of the two regression lines.

4𝑥𝑥 + 3𝑦𝑦 + 7 = 0 and 3𝑥𝑥 + 4𝑦𝑦 + 8 = 0
(i) Mean of x and y
(ii) Regression coefficients
(iii) Coefficient of correlation.
Solution:
(i) The given regression lines are
4𝑥𝑥 + 3𝑦𝑦 + 7 = 0 and 3𝑥𝑥 + 4𝑦𝑦 + 8 = 0
The regression lines intersect at the point (𝑥𝑥̅ , 𝑦𝑦�).
4 11
Solving the two equations we get 𝑥𝑥̅ = − , 𝑦𝑦� = −
7 7
.
(ii) Let 4𝑥𝑥 + 3𝑦𝑦 + 7 = 0 be the regression line of y on x.
4 7 4
Rewriting, 𝑦𝑦 = − 𝑥𝑥 − ⇒ 𝑏𝑏𝑦𝑦𝑦𝑦 = − .
3 3 3

Let 3𝑥𝑥 + 4𝑦𝑦 + 8 = 0 be the regression line of x on y.

4 8 4
Rewriting, 𝑥𝑥 = − 𝑦𝑦 − ⇒ 𝑏𝑏𝑥𝑥𝑥𝑥 = −
3 3 3
4 4 16
Now, 𝑏𝑏𝑦𝑦𝑦𝑦 ⋅ 𝑏𝑏𝑥𝑥𝑥𝑥 = − × − = >1
3 3 9

which means our assumption is incorrect.

Therefore, we conclude that,
4𝑥𝑥 + 3𝑦𝑦 + 7 = 0 is the regression line of x on y and 3𝑥𝑥 + 4𝑦𝑦 + 8 = 0 is the regression line
of y on x.
3 3
∴ 𝑏𝑏𝑥𝑥𝑥𝑥 = − 4 , 𝑏𝑏𝑦𝑦𝑦𝑦 = − 4.
(iii) Since, 𝑏𝑏𝑥𝑥𝑥𝑥 , 𝑏𝑏𝑦𝑦𝑦𝑦 is negative so

3 3 3
𝑟𝑟 = −�𝑏𝑏𝑦𝑦𝑦𝑦 ⋅ 𝑏𝑏𝑦𝑦𝑦𝑦 = −��− � × �− � = −
4 4 4

84
 ISC XI- Applied Mathematics

REAL- LIFE APPLICATIONS

Real-World Applications of Regression

• Finance:
− Predicting stock prices based on market data.
− Forecasting sales based on factors like weather, GDP, or previous sales.
− Using the Capital Asset Pricing Model (CAPM) for asset pricing.
• Business:
− Predicting sales based on advertising spending or other factors.
− Optimizing inventory management and staffing levels in e-commerce by predicting peak
shopping periods.
− Analyzing the impact of marketing campaigns.
• Healthcare:
− Identifying risk factors for diseases, like heart disease.
− Predicting blood pressure based on age, gender, weight, etc.
• Other Fields:
− Predicting house prices based on size, location, and other features.
− Forecasting sales of a product based on historical data.
− Predicting the number of road accidents based on reckless driving.
− Identifying the effect of advertising on sales.
• Technology:
− Predicting device malfunctions in Internet of Things (IoT) systems.
− Developing recommendation systems in gaming.
− Analyzing risk in financial software.

Exercises
1. Choose the correct answer.

(i) The slope of the regression line of y on x is also referred to as the?

(a) Regression coefficient of x on y (c) Regression coefficient of y on x
(b) The correlation coefficient of x on y (d) Correlation coefficient of y on x.

(ii) Which of the below statements concerning two regression coefficients’ arithmetic mean is correct?
(a) It is greater than the correlation (c) It is equal to or greater than the
coefficient. correlation coefficient.
(b) It is less than that of the correlation (d) It’s equal to the correlation coefficient.
coefficient.

85
 ISC XI- Applied Mathematics

(iii) What is the sum of squares of error if all the actual and estimated values of Y on the regression line
are the same?
(a) Maximum (c) Zero
(b) Minimum (d) Unknown

(iv) Which of the following statement(s) is/are correct with respect to regression coefficients?
Statement 1. It measures the degree of linear relationship between two variables.
Statement 2. It gives the value by which one variable changes for a unit change in the other variable.
(a) Only 1 (c) Both 1 and 2
(b) Only 2 (d) Neither 1 nor 2

(v) Assertion: Regression coefficient is affected by change of scale.

Reason: Changing scale affects the value of the standard deviation and regression coefficient depends
on the standard deviations.
(a) Both A and R individually true and R is the correct explanation of A.
(b) Both A and R individually true but R is not the correct explanation of A.
(c) A is true but R is false.
(d) A is false but R is true.

2. The two lines of regressions are x + 2y - 5 = 0 and 2x + 3y - 8 = 0 and the variance of x is 12. Find the
variance of y and the coefficient of correlation.
3. The following table shows the sales and advertisement expenditure of a firm.

Sales Advertisement Expenditures

(In Crores ₹) (In Crores ₹)
Mean 40 6
Standard Deviation 10 1.5

The correlation coefficient is 0.9. Estimate the likely sales for a proposed expenditure of Rs 10 crores.

4. Find the line of regression of y on x from the following table.

x 1 2 3 4 5
y 7 6 5 4 3

Hence, estimate the value of y when x = 6.

5. The following observations are given: (1,4), (2,8), (3,2), (4,12), (5,10), (6,14), (7,16), (8,6), (9,18).
Estimate the values of
(i) y when the value of x is 10.
(ii) x when the value of y is 5.
6. The following data represents the number of hours 12 different students used mobile phones during the
weekend and the scores of each student who took a test the following Monday.
(i) Find the equation of the suitable regression line.
(ii) Use the equation to find the expected test score for a student who used the mobile phone for 9
hours.
Hours 0 1 2 3 3 5 5 5 6 7 7 10
Test score 96 85 82 74 95 68 76 84 58 65 75 50

86
 ISC XI- Applied Mathematics

7. Calculate the two regression equations of x on y and y on x from the data given below, taking deviations
from actual means of x and y.

Price (₹) 10 12 13 12 16 15
Demand(kg) 40 38 43 45 37 43

Estimate the likely demand when the price is ₹.20.

8. Obtain regression equation of y on x and estimate y when x = 55 from the following

x 40 50 38 60 65 50 35
y 38 60 55 70 60 48 30

9. Find the means of x and y and the coefficient of correlation between them from the following two
regression equations:
2y – x – 50 = 0
3y – 2x – 10 = 0.

10. Consider the time series data given below:

x 8 3 2 10 11 3 6 5 6 8

y 4 12 1 12 9 4 9 6 1 14

Use the least square method to determine the equation of line of best fit (y on x) for the data.

Answers
1 (i) (c) Regression coefficient of y on x
(ii) (a) It is greater than the correlation coefficient.
(iii) (c) Zero
(iv) (b) Only 2
(v) (a) Both A and R individually true and R is the correct explanation of A.
2. Variance of y = 4, r = - 0.866
3. ₹ 64 crores.
4. y+ x = 8, y = 2
50 10
5. y = , x=
3 3

6. y = - 4.07x + 93.97, Expected score = 57.34

7. x = - 0.12y + 17.92, y = - 0.25x + 44.25, y = 39.25 kg
8. y = 0.942x + 6.08, y = 57.89
9. 𝑥𝑥̅ = 130, 𝑦𝑦� = 90, 𝑟𝑟 = 0.866
10. y = 3.0026 + 0.677x

87
 ISC XI- Applied Mathematics

Project Work
Project Title: "Predicting Monthly Electricity Bill Based on Number of Units Consumed"
Objective:
To apply regression analysis to predict the monthly electricity bill of a household based on the number of
electricity units consumed, using real data.
Concepts Applied:
• Regression line (linear regression)
• Scatter diagram
• Correlation coefficient
• Regression coefficients
• Equation of regression line
Real-Life Context:
Electricity bills usually depend on the number of units used. Understanding this relationship helps families
predict bills, budget wisely, or adopt energy-saving strategies.
Data Collection:
Collect data from your or a few households over 6–12 months. Record:

Month Units Consumed (x) Electricity Bill (y in ₹)

(If real data isn’t available, simulate realistic values.)

Analysis Steps:
1. Draw a scatter diagram of units consumed (x) vs. electricity bill (y).
2. Find:
• Mean of x and y
• Standard deviations σx ,σy
• Correlation coefficient r
3. Calculate regression coefficient.
4. Form the regression equation of y on x.
5. Use the equation to predict the bill for any given number of units.
Example: If a household uses 250 units in a month, what would be the expected bill? Use the regression
equation to estimate.
Conclusion:
• Discuss whether the relationship is strong or weak (based on correlation).
• Comment on whether the model is realistic and useful.
• Suggest how it could help families plan better.

88
 ISC XI- Applied Mathematics

HISTORICAL NOTE

• Early Roots: Isaac Newton's work in 1700 on equinoxes is considered an early form of regression
analysis, including averaging data and forcing the regression line through the average point.
• Method of Least Squares: Legendre (1805) and Gauss (1809) independently developed the method
of least squares, which is fundamental to modern regression analysis. They applied it to astronomical
observations, like determining orbits of celestial bodies.
• Francis Galton's Contribution: Galton, a cousin of Charles Darwin, studied heredity, particularly
the inheritance of height. He noticed that tall parents tended to have somewhat shorter offspring, and
vice versa. He coined the term "regression" to describe this "reversion" towards the mean.
• Evolution of Regression: Galton's work laid the groundwork for understanding the relationship
between variables and predicting one variable based on another. Since then, numerous researchers
have expanded on these concepts.

89
 ISC XI- Applied Mathematics

(v) Probability

SCOPE

Random experiments; outcomes, sample spaces (set representation). Events; occurrence of events, 'not',
'and' and 'or' events, exhaustive events, mutually exclusive events, Axiomatic (set theoretic) probability,
connections with other theories studied in earlier classes. Probability of an event, probability of' not',
'and' and 'or' events.
• Random experiments and their outcomes.
• Events: sure events, impossible events, mutually exclusive and exhaustive events.
- Definition of probability of an event
- Laws of probability addition theorem.

LEARNING OUTCOMES

Students will be able to:

• express the sample space and events for a random experiment.
• classify events as exhaustive events and mutually exclusive events.
• describe the idea of probability.
• find the probability of events by using classical and axiomatic theories.

PRE-REQUISITES

• Basic Understanding of Sets and Logic.

• Venn Diagrams.
• Permutation and Combination.

Introduction
In common parlance the term Probability refers to the chance of happening or not happening of an event. The
use of the word chance in any statement indicates that there is an element of uncertainty about that statement.
For example, consider the statements,
(a) The probability that it will rain tomorrow is low.
(b) The chance that the prices of Company A will go up in the next financial year is very high.
There is an element of uncertainty associated with each one of the above statements as there is no numerical
value assigned to these statements.
Probability provides a numerical measure of the element of uncertainty. It enables us to take decisions under
conditions of uncertainty with a calculated risk.

90
 ISC XI- Applied Mathematics

Random Experiment
It is an experiment which if conducted repeatedly under homogeneous conditions does not give the same
result. It has more than one possible outcome and it is not possible to predict the outcome in advance. The
result maybe any one of the various possible outcomes. For example, if an unbiased die is thrown any one of
the six numbers can appear as an outcome.

Sample space
This is a set of all possible outcomes of a random
experiment. It is also called event space or
probability set. It is denoted by S. Each outcome is
called an element or a sample point of the sample
space.

Sample Space using Tree Diagrams

When a coin is tossed three times then all possible
outcomes can be represented by drawing a tree
diagram as follows:
In case of a single toss, the number of possible
outcomes is 2. In case of two tosses, the number of outcomes
is 2 × 2 = 22 = 4. In case of three tosses, it is 2 × 2 × 2 =
23 = 8.
Therefore, for n tosses, the number of outcomes will be 𝟐𝟐𝒏𝒏 .
Now, in a different scenario, if the coin shows a head, we toss it
again but if it shows a tail, then we roll a die. In this case, the
events do not have the same chances of occurrence that is they
are not equally likely events.

Trial and Event

The performance of a random experiment is called a trial and the
outcome is the event. Thus, the throwing of a die would be called
a trial and the result of getting any one of the six outcomes will
be called an event.

Events
Events could be either simple/elementary or compound/composite. An event is called simple if it corresponds
to a single possible outcome. Thus, in throwing a die, the chance of getting a 3 is a simple event. However,
the chance of getting an odd number is a compound event.
A compound event can be further decomposed into simple events. So, for the above example for compound
event, the decomposed simple events are getting either 1 or 3 or 5 on throwing the die.
Exhaustive Cases: All possible outcomes of an event are known as exhaustive cases. So, in the case of a
throwing a die there are 6 exhaustive cases but if two dice are thrown there are 36 exhaustive cases to consider.

Favourable Cases: The number of outcomes which result in the occurrence of a desired event are called
favourable cases. In drawing a card from a pack, the cases favourable to getting a spade are 13.

Mutually Exclusive Events: Two or more events are said to be mutually exclusive if the occurrence of one
of them excludes the occurrence of the others. So, when tossing a coin getting a head or a tail are mutually
exclusive events.
Equally likely Events: Two or more events are said to be equally likely if their chances of happening are
equal. For the tossing of a coin, it is equally likely to get a head or a tail as a result.
91
 ISC XI- Applied Mathematics

Not Equally likely Events: Not equally likely events are those where each possible
outcome of an experiment does not have the same chances of occurring. For example, a
spinner with unequal sections will have different probabilities for landing on each
colour.
Sure Event: Since the set S is a subset of itself, it represents an event. Since all the
outcomes of a random experiment belong to its sample space S, the event S represents
the sure event.
Impossible Event: Since the null set is also a subset of S, therefore, it also represents an event. As no outcomes
of the experiment belongs to 𝜙𝜙, the event 𝜙𝜙 can never occur when the experiment is performed. For this
reason, the event 𝜙𝜙 is called the impossible event.
Complementary Event: Let S be the sample space associated with a random experiment and let E be an
event. The event of non-occurrence of E is called the complementary event of E. It is represented by 𝐸𝐸 ′ or 𝐸𝐸� .
We also call this event as not E.

Algebra of Events using Set Theory

Let A and B be any two events associated with a random experiment.
(i) Union of two events: A ∪ B is the event that occurs if A occurs, or B occurs or both occurs.
(ii) Intersection of two events:A ∩ B is the event that occurs if A occurs, and B occurs.
(iii) Complement of an event: The complement of an event A is the event that occurs if A does not occur.
It is represented by 𝐴𝐴′ or 𝐴𝐴̅.It is a set of all sample points of the sample space S that is not contained
in A.
The complement of 𝐴𝐴 ∪ 𝐵𝐵 is 𝐴𝐴 ∪ 𝐵𝐵 = 𝐴𝐴̅ ∩ 𝐵𝐵�.
�������
(iv) Mutually exclusive events: Two or more events are mutually exclusive if the events cannot occur
together that is they are disjoint sets and have no common elements together. Hence, 𝐴𝐴 ∩ 𝐵𝐵 = 𝜙𝜙.
(v) Mutually exclusive and exhaustive events: If S is a sample space associated with a random
experiment and 𝐸𝐸1 , 𝐸𝐸2, … 𝐸𝐸𝑛𝑛 are elementary events associated with the random experiment then if
𝐸𝐸𝑖𝑖 ∩ 𝐸𝐸𝑗𝑗 = 𝜙𝜙 𝑓𝑓𝑓𝑓𝑓𝑓 𝑖𝑖 ≠ 𝑗𝑗 𝑎𝑎𝑎𝑎𝑎𝑎 𝐸𝐸1 ∪ 𝐸𝐸2 ∪ ⋯ 𝐸𝐸𝑛𝑛 = 𝑆𝑆
then 𝐸𝐸1 , 𝐸𝐸2, … 𝐸𝐸𝑛𝑛 form a mutually exclusive and exhaustive system of events.

Notations

Terms in probability Set notations

Sample space S
Event A has occurred A

Event A has not occurred/ Not A 𝐴𝐴̅ or 𝐴𝐴′

Event A or B or both 𝐴𝐴 ∪ 𝐵𝐵
Event A and B 𝐴𝐴 ∩ 𝐵𝐵
Event A and B are mutually exclusive 𝐴𝐴 ∩ 𝐵𝐵 = 𝜙𝜙
Event A and B are not mutually exclusive 𝐴𝐴 ∩ 𝐵𝐵 ≠ 𝜙𝜙
Only Event A occurs 𝐴𝐴 ∩ 𝐵𝐵�
Only Event B occurs 𝐴𝐴̅ ∩ 𝐵𝐵
Neither Event A occurs nor Event B 𝐴𝐴̅ ∩ 𝐵𝐵�

92
 ISC XI- Applied Mathematics

What is Probability?
The probability of an event denotes the likelihood of its occurrence. The value of probability ranges from 0 to
1. If an event is certain to happen its probability would be 1 and it is certain that it will not take place its
probability would be 0.

Classical or Mathematical or Priori definition of Probability

This is the earliest approach to the theory of probability. Laplace defined this as the ‘ratio of the number of
favourable cases to the total number of equally likely cases.’ This theory assumes that various outcomes of an
event are equally likely and so the probability of their happening is also equal. In a way the theory determines
the probability of an event even before it has taken place. It is for this reason that these probabilities are called
‘priori probabilities’.

𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜

𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑜𝑜𝑜𝑜 𝑎𝑎𝑎𝑎 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 =
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛 𝑜𝑜𝑜𝑜 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜

𝑛𝑛(𝐴𝐴)
For an event A we can write this as, 𝑃𝑃(𝐴𝐴) = ,
𝑛𝑛(𝑆𝑆)
where,
𝑃𝑃(𝐴𝐴) is the probability of an event A,
𝑛𝑛(𝐴𝐴) = the number of favourable outcomes for event A,
𝑛𝑛(𝑆𝑆) is the total number of outcomes in the sample space.

Range of Probability Values

• For a sure event, 𝑛𝑛(𝐴𝐴) = 𝑛𝑛(𝑆𝑆) ⇒ 𝑃𝑃(𝐴𝐴) = 1.
• For an impossible event, 𝑛𝑛(𝐴𝐴) = 0 ⇒ 𝑃𝑃(𝐴𝐴) = 0.
• For a probable event, 0 < 𝑛𝑛(𝐴𝐴) < 𝑛𝑛(𝑆𝑆) ⇒ 0 < 𝑃𝑃(𝐴𝐴) < 1.
Complementary Rule
The probability of the complementary event 𝐴𝐴̅ (i.e., the event NOT A) is given by:
𝑃𝑃(𝐴𝐴̅) = 1 − 𝑃𝑃(𝐴𝐴)

Odds in Favour and Against an Event

𝑃𝑃(𝐴𝐴)
• Odds in favour of event A are given by
𝑃𝑃(𝐴𝐴̅)
𝑃𝑃(𝐴𝐴̅)
• Odds against event A are given by
𝑃𝑃(𝐴𝐴)

Axiomatic Approach to Probability

The classical definition of probability is only applicable if the outcomes are mutually exclusive, exhaustive
and equally likely which is rarely the case for a real-world scenario.
To address these issues, the axiomatic approach to probability, developed by A.N. Kolmogorov
in 1933, provides a rigorous mathematical framework that applies to a wide range of real-world scenarios.
In the axiomatic approach, probability is treated as a function P that assigns a real number to each event in
a sample space. This function must satisfy the following three axioms:
Axiom 1: Non-Negativity
For any event E, 0 ≤ 𝑃𝑃(𝐸𝐸) ≤ 1

93
 ISC XI- Applied Mathematics

Axiom 2: Total Probability

For the entire sample space S, 𝑃𝑃(𝑆𝑆) = 1
Axiom 3: Additivity (for Mutually Exclusive Events)
If A and B are mutually exclusive events, 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵)
This means that the probability of either A or B occurring is the sum of their individual probabilities.

Do you know?
Artificial intelligence (AI) systems rely heavily on probability. When we type a search query in Google,
it predicts what we might type next. The next word is not equally likely. Rather it depends on past
searches and user behaviour. Classical probability cannot be used because all words are not equally
likely. AI systems use probability models to calculate the likelihood of different words.

Why Axiomatic Probability is Better

• It does not assume equal likelihood.
• It allows different probabilities based on real-world observations.
• It works for both fair and biased situations.

Classical Probability as a Special Case of Axiomatic Probability

Let us consider random experiment whose sample space S has finite number of points, say N,
which are mutually disjoint events.
Let us now suppose that all the outcomes of the random experiment are equally likely so that the probability
1
of occurrence for each of the events is . Further, let an event A be such that 𝑛𝑛(𝐴𝐴) = 𝑘𝑘.
𝑁𝑁

Then by the axiom of additivity (Axiom 3), we get,

𝑘𝑘
𝑖𝑖 𝑘𝑘 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑖𝑖𝑖𝑖 𝐴𝐴 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑡𝑡𝑡𝑡 𝐴𝐴 𝑛𝑛(𝐴𝐴)
𝑃𝑃(𝐴𝐴) = � = = = =
𝑁𝑁 𝑁𝑁 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑖𝑖𝑖𝑖 𝑆𝑆 Number of sample points in S 𝑛𝑛(𝑆𝑆)
𝑖𝑖=1

which is the result obtained by the classical definition of probability.

Hence, the classical probability may be regarded as a special case of axiomatic probability.

Probability Vs Odds
Odds represent the likelihood of an event happening compared to it not happening, while probability is the
chance of an event occurring out of all possible outcomes. Odds are expressed as a ratio, while probability is
usually a fraction or percentage.
Given the probability of an event ‘E’, i.e., P(E),
𝑃𝑃(𝐸𝐸)
Odds for 𝐸𝐸 =
1−𝑃𝑃(𝐸𝐸)

1−𝑃𝑃(𝐸𝐸)
Odds against 𝐸𝐸 = 𝑃𝑃(𝐸𝐸)
94
 ISC XI- Applied Mathematics

Aspect Probability Odds

Definition A mathematical expression of how A ratio comparing the likelihood of an
likely an event is to occur, between 0 event happening to it not happening.
and 1.
Range Between 0 and 1 Expressed as a ratio
Formula 𝑃𝑃(𝐴𝐴) 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂
𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 = 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜
=
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 ∶ 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜

Usage Used in statistics and probability Often used in betting, gambling, and
theory. risk assessments.
Interpretation Represents how likely an event will Compares the occurrence of an event to
happen based on a ratio to all possible the non-occurrence.
outcomes.

Example 1. A spinner with three equal sectors coloured red, green and blue is spun
three times.
(i) Represent the sample space on a tree diagram.
Use the tree diagram to work out the probability of getting:
(ii) the same colour on all three spins.
(iii) a green and two reds in any order.
(iv) a different colour on all three spins.
Solution:
(i) Total number of outcomes = 27

(ii) These outcomes are highlighted in yellow below:

RRR, GGG, BBB
3 1
P (getting same colour) = =
27 9

95
 ISC XI- Applied Mathematics

(iii) These outcomes are shown in green below:

GRR, RGR, RRG
3 1
P (GRR in any order) = =
27 9

(iv) These outcomes are highlighted in blue below:

RGB, RBG, GRB, GBR, BGR, BRG
6 2
P (three different colours) = =9
27

Example 2. A fair six-sided die is rolled once. Find the probability of getting:
(i) a multiple of 3
(ii) a number greater than 4
(iii) an odd number
Solution:
The sample space 𝑆𝑆 = {1,2,3,4,5,6}
Total number of outcomes = 6
(i) Multiples of 3 = {3,6} ∴ 2 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜
2 1
P (Multiple of 3) = =
6 3
(ii) Numbers greater than 4 = {5,6} ∴ 2 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜
2 1
P (Number > 4) = =
6 3
(iii) Odd numbers = {1,3,5} ∴ 3 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜
3 1
P (Odd number) = =
6 2

96
 ISC XI- Applied Mathematics

Example 3. A card is drawn randomly from a standard deck of 52 playing cards. Find the probability that
the card is:
(i) A king
(ii) A red card
(iii) A face card (King, Queen, Jack)
Solution:
(i) There are 4 Kings in a deck.
4 1
P(King)= = 13
52
(ii) There are 26 red cards (13 hearts + 13 diamonds).
26 1
P (Red card) = =2
52
(iii) Face cards: 12 total (4 Kings, 4 Queens, 4 Jacks)
12 3
P (Face card) = = 13
52

Example 4. Sahil makes a game involving tossing three coins and noting the sequence of heads and tails. If
more heads than tails appear, he wins. If more tails than heads appear, you win. Is this game fair?
Solution:
To see if the game is fair, we calculate Sahil's chance of winning and your chance of winning. For
the experiment of tossing three coins, the sample space is
{𝐻𝐻𝐻𝐻𝐻𝐻, 𝐻𝐻𝐻𝐻𝐻𝐻, 𝐻𝐻𝐻𝐻𝐻𝐻, 𝑇𝑇𝑇𝑇𝑇𝑇, 𝐻𝐻𝐻𝐻𝐻𝐻, 𝑇𝑇𝑇𝑇𝑇𝑇, 𝑇𝑇𝑇𝑇𝑇𝑇, 𝑇𝑇𝑇𝑇𝑇𝑇}.
4 1
P (Sahil winning) = =2
8
4 1
P (you winning) = =2
8
This is a fair game.

Example 5. Consider the experiment of rolling a pair of dice. Let A be the event that the numbers on the dice
have a sum of 8. Let B be the event that the two numbers on the dice are a 3 and a 5.
(i) What is P(A)?
(ii) What is P(B)?
Solution:
The sample space for this experiment has 36 outcomes.
𝑆𝑆 = {(1,1), (1,2), (1,3), (1,4), (1,5), (1,6), (2,1), (2,2), (2,3), (2,4), (2,5), (2,6),
(3,1), (3,2), (3,3), (3,4), (3,5), (3,6), (4,1), (4,2), (4,3), (4,4), (4,5), (4,6),
(5,1), (5,2), (5,3), (5,4), (5,5), (5,6), (6,1), (6,2), (6,3), (6,4), (6,5), (6,6)}
5
(i) There are 5 pairs of numbers that have a sum of 8, so 𝑃𝑃(𝐴𝐴) =
36
2 1
(ii) There are 2 pairs of numbers that are a 3 and a 5, so 𝑃𝑃(𝐵𝐵 ) = =
36 18

97
 ISC XI- Applied Mathematics

Example 6. Find the probability that, in a random arrangement of the letters of "SCHOOL", the two O’s
appear together.
Solution:
Total letters = 6
6! 720
Since there are two O’s, the total permutations = = = 360
2! 2
Treating "OO" as a single unit, remaining letters = 5
Total favourable arrangements = 5! = 120
120 1
P (O’s together) = =3
360

Example 7. Two cards are drawn at random from a well-shuffled deck of 52 cards. Find the probability that
both are aces.
Solution:
52! 52×51
Total ways to choose 2 cards from 52 = 𝐶𝐶(52,2) = = = 1326
[2!(52−2)!] 2

4! 4×3
Ways to choose 2 aces from 4 aces = 𝐶𝐶(4,2) = = =6
2!(4−2)! 2

6 1
P (Both Aces) = = 221
1326

Example 8. Two dice are thrown together. Find the probability of getting a sum of at least 10.
Solution:
Total outcomes = 6 × 6 = 36
Cases when sum ≥ 10 = {(4,6), (5,5), (6,4), (5,6), (6,5), (6,6)} ∴ 6 favourable cases
6 1
P (Sum ≥ 10) = =
36 6

Example 9. A committee of 3 is to be formed from a group of 4 men and 5 women. Find the probability that
the committee has at least one woman.
Solution:
9!
Total ways to select 3 members from 9 = 𝐶𝐶(9,3) = = 84
3!(9−3)!
4!
Ways to select only men (no women) = 𝐶𝐶(4,3) = =4
3!(4−3)!
4 80 20
P (At least 1 Woman) = 1 − 𝑃𝑃 (𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 𝑀𝑀𝑀𝑀𝑀𝑀) = 1 − = =
84 84 21

Example 10. A 4-digit PIN is created using the numbers 0 to 9. Find the probability that all four digits are
different.
Solution:
Total possible PINs =10×10×10×10=10000
Ways to select 4 different digits:
• First digit: 10 choices
• Second digit: 9 choices
• Third digit: 8 choices
• Fourth digit: 7 choices
Total ways to select 4 different digits = 10 × 9 × 8 × 7 = 5040
P (All digits different) = 5040/10000 = 0.504
98
 ISC XI- Applied Mathematics

Example 11. A box contains five blue balls, two green balls, and six yellow balls. What are the
(i) odds of drawing a blue ball?
(ii) odds against drawing a blue ball from the box?
Solution:
Let P(B) represent the event that a blue ball is drawn from the box.
5
Therefore, 𝑃𝑃(𝐵𝐵) =
13
5 5
5 13 5
(i) Odds of drawing a blue ball from the box = 13
5 = 13
8 =( )×( )=
�1− � � � 13 8 8
13 13
5 8
�1− � 8 13 8
(ii) Odds against drawing a blue ball from the box= 5
13
= 13
5 = � �� � = 5
� � 13 5
13 13

Addition Theorem of Probability for Two Events

Theorem: Let S be the sample space of a random experiment and events A and B are subsets of S, then
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵).
Corollary: If events A and B are mutually exclusive events, then do not share any outcomes, meaning 𝑃𝑃(𝐴𝐴 ∩
𝐵𝐵) = 0. Therefore, 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵). This is known as the Addition theorem for mutually
exclusive events.

Suppose in a class of 30 students, 8 students are in the band, 15 students play a sport, and 5 students are both
in the band and play a sport. Let A be the event that a student is in the band and let B be the event that a student
plays a sport.
P(A∪B) is the probability that a student is in the band or plays a sport or both.
With the help of the Venn diagram,

3 + 10 + 5 18 3
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = = =
30 30 5

You could also compute this probability using the Addition Rule:
Note that by using the Addition Rule, you avoid having to determine that 3 people
are in a band and don't play a sport and 10 people who play a sport but are not in
the band. The Addition Rule is easier when you have not created a Venn diagram.
Addition Theorem of Probability for Three Events
If A, B, C are three events, then
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵 ∪ 𝐶𝐶) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) + 𝑃𝑃(𝐶𝐶) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) − 𝑃𝑃(𝐵𝐵 ∩ 𝐶𝐶) − 𝑃𝑃(𝐴𝐴 ∩ 𝐶𝐶) + 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵 ∩ 𝐶𝐶)
This theorem extends the two-event addition rule and ensures that each outcome is counted exactly once.
Corollary:
If A, B, C are mutually exclusive 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) = 𝑃𝑃(𝐵𝐵 ∩ 𝐶𝐶) = 𝑃𝑃(𝐴𝐴 ∩ 𝐶𝐶) = 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵 ∩ 𝐶𝐶) = 0. Therefore,
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵 ∪ 𝐶𝐶) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) + 𝑃𝑃(𝐶𝐶)

99
 ISC XI- Applied Mathematics

Example 12. In a school of 100 students, the following data is collected:

• 40 students participate in drama (A).
• 30 students participate in sports (B).
• 20 students participate in music (C).
• 15 students are in both drama and sports.
• 10 students are in both sports and music.
• 5 students are in both drama and music.
• 3 students are in all three activities.
Calculate the probability that a randomly selected student participates in at least one of these activities.
Solution:
Using the Addition Theorem for three events,
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵 ∪ 𝐶𝐶) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) + 𝑃𝑃(𝐶𝐶) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) − 𝑃𝑃(𝐵𝐵 ∩ 𝐶𝐶) − 𝑃𝑃(𝐴𝐴 ∩ 𝐶𝐶) + 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵 ∩ 𝐶𝐶)
40 30 20 15 10 5 3
= + + − − − + = 0.63
100 100 100 100 100 100 100
Thus, 63% of the students participate in at least one of the three activities.

Special Cases of the Addition Theorem - Probability of any one of two events occurring exactly
For two events A and B, the probability that only one of them occurs but not both is given by

𝑃𝑃�(𝐴𝐴 ∩ 𝐵𝐵�) ∪ (𝐴𝐴̅ ∩ 𝐵𝐵)� = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 2𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵)

This formula ensures that we count only the cases in which either A or B occurs individually and not the
instances in which they occur together.
P(A) and P(B) include all cases where either event A or B happens.
𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) is the probability of both events occurring simultaneously. If we simply subtract 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) once (as
in the standard addition theorem), we still count cases where both events occur. To ensure that these cases are
completely excluded, we subtract 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) twice, ensuring that the overlap is fully removed.

Example 13. Nick enjoys dessert after dinner and on any given night the probability that Nick has a
cookie (A) is 10%, the probability that Nick has ice cream (B) is 50% and the probability that Nick has both
cookie and ice cream (A∩B) is 5%. Calculate the probability that Nick has exactly one of the two desserts.
Solution:
Using the formula, 𝑃𝑃((𝐴𝐴 ∩ 𝐵𝐵�) ∪ (𝐴𝐴̅ ∩ 𝐵𝐵)) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 2𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵)
= 0.10 + 0.50 − 2(0.05) = 0.10 + 0.50 − 0.10 = 0.50
There is a 50% chance that Nick has either only a cookie or only ice cream but not
both.
Remark 1:
𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵�) or P (A – B) is the probability of occurrence of A only. This means that
among the outcomes in event A, we exclude those that also belong to event B.

100
 ISC XI- Applied Mathematics

Remark 2:

𝑃𝑃(𝐴𝐴̅ ∩ 𝐵𝐵) or P (B − A) is the probability of occurrence of B only. Among the

outcomes in event B, we exclude those that also belong to event A.

Remark 3:
𝑃𝑃((𝐴𝐴 ∩ 𝐵𝐵�) ∪ (𝐴𝐴̅ ∩ 𝐵𝐵)) is the probability of exactly one of two events occurring.
This means either event A occurs without event B or event B occurs without
event A.

Remark 4:
If A and B are events such that A⊂B, then B splits into A and B. Therefore,
𝑃𝑃(𝐴𝐴) ≤ 𝑃𝑃(𝐵𝐵). This states that if A is completely contained within B, then the
probability of A occurring can never exceed the probability of B.

Probability - Bounds and Limitations

The probability of the intersection (both events occurring) should always be less than or equal to the
probability of each individual event. Similarly, the probability of the union (at least one event occurring)
should always be less than or equal to the sum of the probabilities of the individual events.
For any two events A and B, the probability follows these bounds:

𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) ≤ 𝑃𝑃(𝐴𝐴) ≤ 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) ≤ 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵)

Suppose the probability of rain today is 0.6 and the probability of high winds is 0.3. The probability of both
rain and high winds occurring together is 0.2.
𝑃𝑃(𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 ∩ 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤) = 0.2 ≤ 𝑃𝑃(𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟) = 0.6 ≤ 𝑃𝑃(𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 ∪ 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤) ≤ 0.6 + 0.3 = 0.9
Thus, the probability of either rain or high winds (or both) must be between 0.6 and 0.9.
3 1
Example 14. Given 𝑃𝑃(𝐴𝐴) = and 𝑃𝑃(𝐵𝐵) = , find P(A∪B) if A and B are mutually exclusive events.
5 5

Solution:
Since A and B are mutually exclusive, they do not overlap, meaning 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) = 0.
The addition theorem states 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵)
3 1 4
Substituting the given values, 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = + =
5 5 5

Example 15. Two mutually exclusive events A and B are given in an experiment. If 𝑃𝑃(𝐴𝐴̅) = 0.65 and
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 0.65, find P(B).
Solution:
𝑃𝑃(𝐴𝐴) = 1 − 𝑃𝑃(𝐴𝐴̅) = 1 − 0.65 = 0.35
Since A and B are mutually exclusive,
101
 ISC XI- Applied Mathematics

𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) ⇒ 0.65 = 0.35 + 𝑃𝑃(𝐵𝐵)

∴𝑃𝑃(𝐵𝐵) = 0.65 − 0.35 = 0.30

1 2 1
Example 16. For two non-mutually exclusive events,𝑃𝑃(𝐴𝐴) = , 𝑃𝑃(𝐵𝐵) = and 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = . Find the values
4 5 2
of 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) and 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵�).
Solution:
1 1 2
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) ⇒ = + − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵)
2 4 5
1 5 8 1 13 3
2
= 20 + 20 − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) ⇒ 2 = 20 − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) ⇒ 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) = 20

1 3 1
� ) = 𝑃𝑃(𝐴𝐴) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) = −
Now, using 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵 = 10
4 20

1 1 1
Example 17. If 𝑃𝑃(𝐸𝐸) = , 𝑃𝑃(𝐹𝐹) = and 𝑃𝑃(𝐸𝐸 ∩ 𝐹𝐹) = , find
4 2 8
(i) 𝑃𝑃(𝐸𝐸 ∪ 𝐹𝐹)
(ii) 𝑃𝑃(𝐸𝐸� ∩ 𝐹𝐹� ) (Probability that neither event occurs).
Solution:
1 1 1 5
(i) 𝑃𝑃(𝐸𝐸 ∪ 𝐹𝐹) = 𝑃𝑃(𝐸𝐸) + 𝑃𝑃(𝐹𝐹) − 𝑃𝑃(𝐸𝐸 ∩ 𝐹𝐹) = 4 + 2 − 8 = 8
5 3
(ii) 𝑃𝑃(𝐸𝐸� ∩ 𝐹𝐹� ) = 1 − 𝑃𝑃(𝐸𝐸 ∪ 𝐹𝐹) = 1 − 8 = 8

Example 18. The probability that at least one of the events A and B occurs is 0.6. If 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) =
0.2, find 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵).
Solution:
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) ⇒ 0.6 = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 0.2
∴ 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) = 0.6 + 0.2 = 0.8

Example 19. A number is chosen at random from 1 to 200. What is the probability that it is divisible by 4 or
6?
Solution:
Let A be the event that the number is divisible by 4 and B be the event that the number is divisible by 6.
For A, 200 = 4 + (𝑛𝑛 − 1)4 ⇒ 4(𝑛𝑛 − 1) = 196 ⇒ 𝑛𝑛 − 1 = 49 ⇒ 𝑛𝑛 = 50.
For B, 198 = 6 + (𝑛𝑛 − 1)6 ⇒ 6(𝑛𝑛 − 1) = 192 ⇒ 𝑛𝑛 − 1 = 32 ⇒ 𝑛𝑛 = 33.
For 𝐴𝐴 ∩ 𝐵𝐵, 192 = 12 + (𝑛𝑛 − 1)12 ⇒ 12(𝑛𝑛 − 1) = 180 ⇒ 𝑛𝑛 − 1 = 15 ⇒ 𝑛𝑛 = 16.
50 33 16
𝑃𝑃(𝐴𝐴) = , 𝑃𝑃(𝐵𝐵) = , 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) =
200 200 200 Recall, Arithmetic
50 33 16 67 progression to find the
𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴) + 𝑃𝑃(𝐵𝐵) − 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) = + − = number of terms.
200 200 200 200

102
 ISC XI- Applied Mathematics

Example 20. A survey found that:

• 70% of people use Facebook.
• 50% use Instagram.
• 30% use both Facebook and Instagram.
Find the probability that a randomly selected person uses at least one of these platforms.

Solution:
𝑃𝑃(𝐹𝐹 ∪ 𝐼𝐼) = 𝑃𝑃(𝐹𝐹) + 𝑃𝑃(𝐼𝐼) − 𝑃𝑃(𝐹𝐹 ∩ 𝐼𝐼) = 0.7 + 0.5 − 0.3 = 0.9
The probability that a random person uses either Facebook, Instagram, or both is 0.9 (90%).

Example 21. The Union Budget is going to be announced by the government this week. The probability that
it will be announced on a day is given below:

Day Monday Tuesday Wednesday Thursday Friday

Probability 1 3 1 1 1
7 7 7 7 7

Find the probability of the budget getting announced between Monday to Wednesday.
Solution:
1 3 1 5
P(Monday to Wednesday) = P(Monday) + P(Tuesday) + P(Wednesday) = + + =
7 7 7 7

Example 22. A and B are two candidates seeking admission to IIT. The probability that A getting selected is
0.5 and the probability that both A and B getting selected is 0.3. Prove that the probability of B being selected
is at most 0.8.
Solution
𝑃𝑃 (𝐴𝐴) = 0.5, 𝑃𝑃 (𝐴𝐴 ∩ 𝐵𝐵) = 0.3
We have P (A U B) ≤ 1 ⇒ P (A) + P(B) − P (A ∩ B) ≤ 1
∴ 0.5 + P (B) − 0.3 ≤ 1 ⇒ P (B) ≤ 1 − 0.2 ⇒ P (B) ≤ 0.8
Therefore, probability of B getting selected is at most 0.8.

REAL- LIFE APPLICATIONS

• Weather Forecasting - Predicting the chance of rain, storms, or temperature changes.

• Insurance - Calculating life, health, and vehicle insurance premiums based on risk.
• Games and Gambling - Designing fair games (cards, dice, roulette) and calculating odds.
• Medical Testing - Determining the probability of disease presence or test accuracy.
• Stock Market and Finance - Assessing risk and return, predicting market trends, and portfolio
decisions.
• Manufacturing and Quality Control - Estimating defect rates, and sampling products for inspection.
• Traffic Flow and Accident Analysis - Calculating accident risk at intersections or during certain
times.

103
 ISC XI- Applied Mathematics

• Elections and Polling - Predicting election results using random samples (exit polls).
• Sports - Calculating winning chances, player performance, and match outcomes.
• Genetics and Biology - Determining the probability of inheriting certain traits or diseases.
• Computer Algorithms (e.g., AI and Machine Learning) - Making predictions or decisions based on
data probabilities.
• Project Management - Assessing the probability of meeting deadlines and budget constraints.
• Cryptography and Cybersecurity - Estimating chances of data breaches or decoding encrypted data.

Exercises
1. Choose the correct answer.
1 1
(i) If A, B, C are mutually exclusive and exhaustive events, find P(B) if 𝑃𝑃(𝐶𝐶) = 2 𝑃𝑃(𝐴𝐴) = 𝑃𝑃(𝐵𝐵).
3

1 1
(a) (c)
6 3
1 1
(b) (d)
4 2

(ii) There are four hotels in a hill station. If three tourists check into hotels in a day, what is the
probability that each check into a different hotel?
1 3
(a) (c)
4 8
1 3
(b) (d)
3 4

(iii) One ticket is drawn from a bag containing 30 tickets numbered 1 to 30. Find the probability that it is
a multiple of 3 or 5.
1 7
(a) (c)
15 15
1 8
(b) (d)
3 15

(iv) Let A and B be two events such that 𝑃𝑃(𝐴𝐴 ∪ 𝐵𝐵) = 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵). Then
�) − P(A
Statement 1. P(A ∩ B � ∩ B) = 0.
Statement 2. P(A) + P(B) = 1.
(a) Only 1 (c) Both 1 and 2
(b) Only 2 (d) Neither 1 nor 2

(v) Assertion: The probability that a person visiting a zoo will see tigers is 0.72, the probability that he
will see the elephants is 0.84 and the probability that he will see both is 0.52.
Reason: 0 ≤ 𝑃𝑃(𝐴𝐴) ≤ 1
(a) Both A and R individually true and R is the correct explanation of A.
(b) Both A and R individually true but R is not the correct explanation of A.
(c) A is true but R is false.
(d) A is false but R is true.

104
 ISC XI- Applied Mathematics

2. An ice cream shop has vanilla, chocolate or strawberry ice cream and you can have either chocolate or
caramel sauce on top.
(i) Represent this information on a tree diagram.
Use the tree diagram to answer the following questions: What is the probability that a customer
chosen at random will order?
(ii) a chocolate ice cream?
(iii) an ice cream with caramel sauce?
(iv) a strawberry ice cream with caramel sauce?

3. You are buying a jumper from a shop. You have to choose a size, colour and style. You can choose
between small, medium or large for size; blue, pink, red or brown for colour; and round-neck or V-neck
for style.
(i) Draw a tree diagram to represent the sample space.
Use the tree diagram to answer the following questions:
(ii) How many different options do you have?
(iii) What is the probability that you pick a brown jumper?
(iv) What is the probability that you pick a jumper that has a V-neck?
(v) What is the probability that you pick a large, red, round-neck jumper?

4. A bag contains 4 white and 2 black balls. Two balls are drawn at random. Find the probability that they
are white.

5. A, B and C are bidding for a contract. It is believed that A has exactly half the chance of B and B in turn
4
is th. as likely as C to win the contract. What is the probability for each to win the contract?
5

6. Three dice are tossed. What is the probability that the numbers shown will all be different?

7. A bag contains 6 red marbles, 3 blue marbles, and 1 yellow marble.

(i) What are the odds in favour of choosing a red marble?
(ii) What are the odds against choosing a blue marble?

8. From a pack of cards if one card is drawn at random what is the probability that it is either a King or a
Queen?

9. If 𝑃𝑃(𝐴𝐴) = 0.5, 𝑃𝑃(𝐵𝐵) = 0 ⋅ 3, 𝑃𝑃(𝐴𝐴 ∩ 𝐵𝐵) = 0 ⋅ 2 obtain the probability that

(i) A occurs but not B.
(ii) At least one of A or B occurs.
(iii) Neither A nor B occurs

10. Four-digit numbers are formed by using the digits 1, 2, 3, 4 and 5 without repetition. Find the probability
that a number, chosen at random, is an odd number.

11. The letters of the word ‘SOCIETY’ are placed at random in a row. What is the probability that three
vowels come together?

12. An urn contains 9 balls, 2 of which are white, 3 blue and 4 black. 3 balls are drawn at random from the
urn. What is the chance that (i) the three balls will be of different colours (ii) 2 balls will be of the same
colour and the third a different colour (iii) three balls will be of the same colour?

105
 ISC XI- Applied Mathematics

13. There are 10 persons who are to be seated around a circular table. Find the probability that two particular
persons will always sit together.

14. Eight people in a start-up of 25 people are graduates. If 3 are picked out at random, what is the probability
that
(i) they are all graduates.
(ii) there is at least one graduate.

15. Given are the ages of six people in an old age home:62,90,78, 85,79, 68
If two people are selected as representatives, what is the probability that at least one will have age lower
than the average.

16. Two dice are thrown together. Find the probability that the total is
(i) five
(ii) two
(iii) two or five
(iv) at least five

17. In the Annual sports meet, among the 260 students in XI standard in the school, 90 participated in Kabaddi,
120 participated in Hockey, and 50 participated in Kabaddi and Hockey. A Student is selected at random.
Find the probability that the student participated in
(i) Either Kabaddi or Hockey,
(ii) Neither of the two tournaments,
(iii) Hockey only,
(iv) Kabaddi only,
(v) Exactly one of the tournaments.

18. Twenty-six people were surveyed about their choice of mobile phones. The survey finds that 14 people
have Apple iPhones, 10 have Samsungs and five have Motorola. Four have Apple iPhones and Samsungs,
three have Apple iPhones and Motorola and one has a Samsung and a Motorola. No one has all three types
of phones. Represent this information on a Venn diagram and use it to calculate the probability that a
person chosen at random has:
(i) An iphone
(ii) A Samsung and a Motorola
(iii) Two phones
(iv) No phone
(v) No iphone

19. Two candidates, Riya and Megha, recently sat for a national-level entrance examination.
Riya has been juggling her preparation with a part-time internship, limiting the number of hours she could
dedicate to study. Megha, although studying full-time, has had difficulty grasping some critical topics.
Based on their preparation patterns and previous mock test performances:
• The probability that Riya qualifies the exam is 0.05,
• The probability that Megha qualifies is 0.10, and
• The probability that both Riya and Megha qualify is 0.02.
Answer the following:
(i) What is the probability that at least one of them will qualify the examination?
(ii) What is the probability that at least one of them will not qualify?

106
 ISC XI- Applied Mathematics

(iii) What is the probability that neither Riya nor Megha qualifies the exam?
(iv) What is the probability that only one of them qualifies the exam?

20. Out of four team members — Arjun, Nisha, Faizan, and Leena — only one person will be selected for a
leadership training program next month. Therefore, the sample space is:
S = {Arjun selected, Nisha selected, Faizan selected, Leena selected}
You are given the following information:
• The chances of Arjun being selected are equal to Leena's.
• Nisha's chances are twice that of Arjun.
• Faizan's chances are four times that of Arjun.
Based on this, answer the following:
(i) What is the probability that Arjun gets selected?
(ii) What is the probability that Nisha gets selected?
(iii) What is the probability that Faizan gets selected?
(iv) What is the probability that Leena gets selected?

21. A sample space consists of 9 elementary outcomes E1, E2, ..., E9 whose probabilities are
P(E1) = P(E2) = 0.08,
P (E3) = P (E4) = 0.1
P (E6) = P (E7) = 0.2
P (E8) = P (E9) = 0.07
Suppose A = {E1, E5, E8}, B = {E2, E5, E8, E9}
(i) Calculate P (A), P (B), and P (A ∩ B).
(ii) Calculate P (A ∪ B).
(iii) Calculate P (𝐵𝐵�).

Answers
1
1 (i) (a)
6
3
(ii) (c)
8
7
(iii) (c)
15

(iv) (a) Only 1

(v) (b) Both A and R individually true and R is not the correct explanation of A
2. (i)

107
 ISC XI- Applied Mathematics

1
(ii) P (Chocolate) =
3
1
(iii) (iii) P (Caramel) =
2
1
(iv) (iv) P (Strawberry with Caramel) =
6

3. (i)

(ii) 24
1
(iii) P(Brown) =
4
1
(iv) P (V neck) =
2
1
(v) P (large, red, round-neck) =
24

2
4
5
2 4 5
5. P(A) = , P(B) = , P(C) =
11 11 11
5
6
9
7. (i) 3:2 (ii) 7:3
2
8.
13
9. (i) 0.3 (ii) 0.6 (iii) 0.4
10. 0.6
1
11.
7
2 55 5
12. (i) (ii) 84 (iii) 84
7
2
13.
9
14 81
14. (i) (ii) 115
575
3
15.
5
1 1 5 5
16. (i) (ii) 36 (iii) 36 (iv) 6
9
8 5 7 2 11
17. (i) (ii) 13 (iii) 26 (iv) 13 (v) 26
13
108
 ISC XI- Applied Mathematics

7 1 4 5 6
18. (i) (ii) 26 (iii) 13 (iv) 26 (v) 13
13
19. (i) 0.13 (ii) 0.98 (iii) 0.87 (iv) 0.11
1 1 1 1
20. (i) (ii) (iii)2 (iv)8
8 4
21. P(A) = 0.25, P(B) = 0.32, P (𝐴𝐴 ∩ 𝐵𝐵) = 0.17, P( 𝐴𝐴 ∪ 𝐵𝐵) = 0.4, 𝑃𝑃(𝐵𝐵�) = 0.68

HISTORICAL NOTE

• Early Beginnings: The earliest motivation for probability came from gambling and games of
chance. People wanted to understand how likely they were to win or lose in games involving dice,
cards, or coins.
• A physician and mathematician, Gerolamo Cardano, wrote a book titled "Liber de Ludo Aleae"
(Book on Games of Chance), which included rules of probability based on dice throws. He was one
of the first to define probability as a ratio of favourable to total outcomes.
• The modern theory of probability began with a famous correspondence between Blaise Pascal and
Pierre de Fermat in [Link] solved the "Problem of Points", which dealt with how to fairly divide
winnings in an unfinished game — laying the foundation of expected value and fair division. This
is considered the birth of modern probability theory.
• Christiaan Huygens published the first formal textbook on probability, "De Ratiociniis in Ludo
Aleae" (On Reasoning in Games of Chance), in 1657. He used mathematical methods to calculate
probabilities in dice games and introduced terminology like "expectation."
• Mathematicians like Jacob Bernoulli, Abraham de Moivre, and Pierre-Simon Laplace expanded the
subject by developing laws of large numbers, the normal distribution, and more complex probability
models. Set theory and formal logic, developed later, helped formalize the ideas of events, sample
spaces, and operations on events like union and intersection.
• With the development of set theory in the 19th century by Georg Cantor and others, events in
probability could be represented as sets, allowing for more systematic calculation.

109