)

3
651
01
STA
34
80
201/209/221/229
(0
Y

TUTORIAL NOTE
U
G
TS
A

(Statistics for Physical Sciences and

ST

Engineering / Statistics for Biological and

T

Agricultural Sciences)
A
TH

THAT STATS GUY

 STA 201/209/221/229 TUTORIAL NOTE

INTRODUCTION

This guide is designed to give you a clear and straightforward approach

to mastering Statistics, setting you ahead of your peers. However,
simplicity doesn't replace effort. To achieve excellence, you'll need to
engage deeply with the material provided.

)
3
I've worked hard to make this guide as accessible as possible.

6 51
You're embarking on a journey toward academic success, and I'm here

01
cheering you on every step of the way. Congratulations on taking this
first step—let's achieve greatness together!

34
That Stats Guy ( 2024 )

80
(0
Need a Personal Statistics Tutor? Text me on WhatsApp :
08034016513
Y
U
G

OR SCAN ME
TS
A
ST
T
A
TH

THIS MATERIAL IS NOT TO BE SOLD FOR

ANY REASON

1
 STA 201/209/221/229 TUTORIAL NOTE

COURSE OUTLINES
1. Scope of Statistical Methods
2. Measure of Location, Partition and Dispersion :
3. Measure of Symmetry and Peakedness : Skewness and Kurtosis
4. Elements of Probability
5. Probability Distributions : Bernoulli Distribution, Binomial
Distribution, Poisson Distribution.

)
3
6. Point and Interval Estimation: Confidence Intervals

51
7. Test of Hypothesis for Mean and Proportion of both Large and
Small Samples

6
01
34
80
(0
Y
U
G

THIS MATERIAL IS NOT TO BE SOLD FOR

TS

ANY REASON
A
ST
T
A
TH

2
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 1 : SCOPE FOR STATISTICAL METHOD

WHAT’S STATISTICS?
Statistics can be described as a branch of Sciences which deals with the
Collection, Organization, Sorting, Presentation and Analysis of Data. You
can take statistics as a branch of Science that help describe and make

)
conclusions about life situations that can be evaluated numerically.

3
51
Statistical methods are tools that help us understand and make sense

6
of data. They involve collecting, organizing, analyzing, and interpreting

01
information so that we can draw conclusions and make decisions based
on that information. Imagine having a lot of numbers and trying to

34
figure out what they mean—that’s where statistical methods come in.

80
They help us summarize the data, see patterns, and understand the
bigger picture.
(0
Statistics has its application in all disciplines from Aeronautics,
Anatomy, Quality Control, Zoology e.t.c
Y
U

PURPOSE OF STATISTICAL METHOD

G
TS

The main purpose of using statistical methods is to help us make sense

of large amounts of data in a clear and understandable way. Here’s why
A

they’re important:
ST

1. Summarizing Data: Statistical methods help us take large sets of

data and summarize them with simple numbers, like averages
T

(mean), or by showing us how spread out the data is (range,

A

standard deviation). This makes it easier to understand what the

TH

data is telling us without getting lost in all the details.

2. Making Predictions: Often, we can't look at every single piece of
data (like asking every person in a country their opinion). Instead,
we take a small sample and use statistical methods to make
educated guesses (inferences) about the whole group.

3
 STA 201/209/221/229 TUTORIAL NOTE

3. Testing Ideas: If we have a guess or hypothesis about something,

statistical methods allow us to test that idea using data. For
example, if a new drug is thought to work better than an old one,
statistics can help show whether this is really true based on the
data from experiments.
4. Finding Relationships: Statistical methods help us find out if
there’s a connection between two things, like whether studying

)
more leads to better grades. They show us how strong that

3
connection is and whether it's likely to be real or just by chance.

51
5. Making Decisions: Whether it’s in business, healthcare, or

6
government, decisions often rely on data. Statistical methods

01
provide the tools to analyze that data so that decisions are based

34
on solid evidence rather than guesswork.
6. Ensuring Accuracy: Statistical methods help ensure that the

80
conclusions we draw from data are accurate. They help us check
for errors, biases, or mistakes in the data so that our findings are
(0
trustworthy.
7. Being Fair and Ethical: Statistical methods guide us in using data
Y

in a way that’s fair and ethical. This means protecting people’s

U

privacy, being honest about the results, and not twisting the data
G

to fit a desired outcome.

TS
A
ST
T
A
TH

4
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 2 : MEASURE OF LOCATION, PARTITION AND

DISPERSION

MEASURES OF LOCATION

Measures of location, also known as measures of central tendency, are

)
statistical tools used to identify the central value within a data set. These

3
measures help us understand where the "center" of the data lies, giving

51
us a summary point around which the data points are distributed. The

6
most common measures of location are the mean, median, and mode,

01
each offering a different perspective on the data

34
THE MEAN

80
The mean, often referred to as the average, is the sum of all the values in
(0
a data set divided by the number of values. It’s a commonly used
measure because it considers all data points, making it a good indicator
Y

of the overall level of the data.

U

The Mean can be evaluated for both Grouped and Ungrouped Data.
G

Ungrouped data is raw data that has been collected but not yet
TS

organized into any specific format. It is presented as individual data

A

points and is often listed in the order in which it was collected. For
example, if you collect the ages of 10 people, you might just list them as
ST

21, 25, 19, 22, 20, 23, 24, 26, 21, 22.
T
A

Grouped data is data that has been organized into groups or classes.
Instead of listing every individual data point, the data is sorted into
TH

intervals, and the frequency (the number of times data points fall within
each interval) is recorded. This method is particularly useful when
dealing with large data sets. For example, ages might be grouped into
intervals like 18-20, 21-23, and 24-26, with the frequency of ages falling
within each range recorded.

5
 STA 201/209/221/229 TUTORIAL NOTE

MEAN OF UNGROUPED DATA

To calculate the mean of an ungrouped data, all you have to do is sum
all the data points and divide it by the number of observations. The
formula is given as

Σ𝑥𝑖
Mean = 𝑁

)
Example 1.11 : Calculate the mean for the set of data :

3
13,12,24,7,9,10,13,13,18,8

6 51
Solution:

01
Σ𝑥𝑖
Mean = 𝑁

34
13+12+24+7+9+10+13+13+18+8 127
= 10
= 10
= 12. 7

MEAN OF GROUPED DATA

80
(0
Evaluating the mean of a grouped Data is different from that of and
ungrouped data and the formula is given by
Y

Σ𝑓𝑖𝑥𝑖
U

𝑀𝑒𝑎𝑛 = Σ𝑓
G

𝑙𝑜𝑤𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑙𝑖𝑚𝑖𝑡 + 𝑢𝑝𝑝𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑙𝑖𝑚𝑖𝑡

Where, 𝑥 = 2
TS

Lower Class Limit : The lower class limit is the smallest value that can
A

belong to a particular class or interval. For example, if you have a class

ST

interval of 10-19, the lower class limit is 10.

T

Upper Class Limit : The upper class limit is the largest value that can
A

belong to a particular class or interval. Using the same example of the

TH

class interval 10-19, the upper class limit is 19

Example 1.12 : The following data represent the weight of products

manufactured in the Factory ( in Kg ). Evaluate the Mean.

Weight 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99

6
 STA 201/209/221/229 TUTORIAL NOTE

Freq 0 41 98 149 192 197 152 97 44 10

Solution

Class Interval f x fx

0-9 20 4.5 90

)
3
10-19 41 14.5 594.5

51
20-29 98 24.5 2401

6
01
30-39 149 34.5 5140.5

40-49 192 44.5

34
8544

50-59 197 54.5 10736.5

60-69 152 64.5

80 9804
(0
70-79 97 74.5 7226.5

80-89 44 84.5
Y

3718
U

90-99 10 94.5 945

G

Σ𝑓 = 1000 Σ𝑓𝑥 = 49200

TS

Σ𝑓𝑖𝑥𝑖
A

𝑀𝑒𝑎𝑛 = Σ𝑓
ST

49200
= 1000
= 49. 2
T
A

THE MEDIAN
TH

The median is the middle value in a data set when the values are
arranged in ascending or descending order. If the number of
observations is odd, the median is the middle value. If even, it's the
average of the two middle values.

MEDIAN OF UNGROUPED DATA

7
 STA 201/209/221/229 TUTORIAL NOTE

The Median of an ungrouped data is given by the Formula below after

the data has been arranged in either ascending or descending order

𝑁+1 𝑡ℎ
𝑀𝑒𝑑𝑖𝑎𝑛 = ( 2
) 𝑖𝑡𝑒𝑚

Example 1.21 : Evaluate the median for the set of data :

)
13,12,7,9,10,13,13,18,8

3
Solution

51
Arranging the data in ascending order we have ; 7,8,9,10,12,13,13,13,18

6
Here, N = 9 (Odd Value)

01
𝑁+1 𝑡ℎ
𝑀𝑒𝑑𝑖𝑎𝑛 = ( 2
) 𝑖𝑡𝑒𝑚

34
9+1 𝑡ℎ
=( 2
) 𝑖𝑡𝑒𝑚

80
10 𝑡ℎ
=( 2
) 𝑖𝑡𝑒𝑚
(0
= 5𝑡ℎ 𝑖𝑡𝑒𝑚
From the data above we can see that the fifth item is 12. Therefore, our
Y

median is 12
U

Example 1.22 : Evaluate the median for the set of data :

G

13,12,7,9,10,13,13,18,8,24
TS

Solution
Arranging the data in ascending order we have ; 7,8,9,10,12,13,13,13,18,24
A

Here, N = 10 (Even Value)

ST

𝑁+1 𝑡ℎ
𝑀𝑒𝑑𝑖𝑎𝑛 = ( 2
) 𝑖𝑡𝑒𝑚
T

10+1 𝑡ℎ
=( ) 𝑖𝑡𝑒𝑚
A

2
TH

11 𝑡ℎ
=( 2
) 𝑖𝑡𝑒𝑚

= 5. 5𝑡ℎ 𝑖𝑡𝑒𝑚
From the data above we can see that the 5. 5𝑡ℎ 𝑖𝑡𝑒𝑚 is between 12 and
12+13 25
13. Therefore, our median is 2
= 2
= 12. 5

8
 STA 201/209/221/229 TUTORIAL NOTE

MEDIAN OF GROUPED DATA

The Median of a grouped data is given by the formula below;

𝑁
2
− 𝐶𝑓𝑏𝑚
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝐿𝑖 + ( 𝑓𝑚
)𝑐
where , 𝐿𝑖 = 𝐿𝑜𝑤𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑏𝑜𝑢𝑛𝑑𝑎𝑟𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑀𝑒𝑑𝑖𝑎𝑛 𝐶𝑙𝑎𝑠𝑠
𝑁 = 𝑇𝑜𝑡𝑎𝑙 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑑𝑎𝑡𝑎

)
𝐶𝑓𝑏𝑚 = 𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑀𝑒𝑑𝑖𝑎𝑛 𝐶𝑙𝑎𝑠𝑠

3
51
𝑓𝑚 = 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑀𝑒𝑑𝑖𝑎𝑛 𝐶𝑙𝑎𝑠𝑠
𝑐 = 𝐶𝑙𝑎𝑠𝑠 𝑤𝑖𝑑𝑡ℎ

6
01
Lower Class Boundary ( LCB ): The lower class boundary is the smallest

34
value that can still be considered part of a particular class or interval
when considering the data's continuous nature. It is typically found by

80
subtracting 0.5 from the lower class limit. This adjustment accounts for
the possibility of data points being right on the edge of the class
(0
intervals. For a class interval with a lower class limit of 10, the lower
class boundary would be 9.5
Y
U

Higher Class Boundary ( HCB ): The upper class boundary is the largest
G

value that can still be considered part of a particular class or interval. It

is typically found by adding 0.5 to the upper class limit. This adjustment
TS

ensures that any value just below the next class's lower limit is included
in the current class. For a class interval with an upper class limit of 19,
A

the upper class boundary would be 19.5.

ST

Cumulative Frequency ( CF ) : Cumulative frequency is the running

T

total of frequencies up to the current class interval. It shows the number

A

of data points that fall within or below a certain class interval. By adding
TH

the frequency of each class to the sum of the frequencies of all previous
classes, cumulative frequency helps in understanding how data
accumulates across intervals. If the frequency of the first class is 5, the
second is 7, and the third is 3, the cumulative frequency for the third
class would be 5+7+3=15

9
 STA 201/209/221/229 TUTORIAL NOTE

Class Width : Class Width is the difference between the Lower Class
Boundary and Higher Class Boundary of a class. If the lower class
boundary is 9.5 and the upper class boundary is 19.5, the class width
would be 19.5−9.5=10

Example 1.23 : The following data represent the weight of products

manufactured in the Factory ( in Kg ). Evaluate the Median

)
3
51
Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99
ht

6
01
Freq 20 41 98 149 192 197 152 97 44 10

34
Solution:

Class
Interval
Frequency
80
LCB - HCB CF
(0
0-9 20 -0.5 - 9.5 20
Y

10-19 41 9.5 - 19.5 61

U
G

20-29 98 19.5 - 29.5 159

TS

30-39 149 29.5 - 39.5 308

40-49 192 39.5 - 49.5 500

A
ST

50-59 197 49.5 - 59.5 697

60-69 152 59.5 - 69.5 849

T
A

70-79 97 69.5 - 79.5 946

TH

80-89 44 79.5 - 89.5 990

90-99 10 89.5 - 99.5 1000

Σ𝑓 = 1000

10
 STA 201/209/221/229 TUTORIAL NOTE

𝑁 𝑡ℎ
The median class is the class where the ( 2
) 𝑖𝑡𝑒𝑚 is. Here, N = 1000
1000 𝑡ℎ
Median Class = ( 2
) 𝑖𝑡𝑒𝑚 = 500𝑡ℎ 𝑖𝑡𝑒𝑚
Checking the Cumulative Frequency ( CF ) we can see that the 500th
item is in the class 40 - 49
Median Class = 40 - 49

)
In order to calculate for the Median Value we make use of the formula

3
𝑁
2
− 𝐶𝑓𝑏𝑚
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝐿𝑖 + ( )𝑐

51
𝑓𝑚
𝐿𝑖 = 39. 5 𝐶𝑓𝑏𝑚 = 308 𝑓𝑚 = 192 𝑁 = 1000 𝑐 = 49. 5 − 39. 5 = 10

6
01
500 − 308
𝑀𝑒𝑑𝑖𝑎𝑛 = 39. 5 + ( 192
)10

34
192
= 39. 5 + ( 192
)10
= 39. 5 + (1)10

80
= 39. 5 + 10 = 49. 5
(0
THE MODE
The mode is the value that appears most frequently in a data set. A data
Y

set may have one mode (unimodal), more than one mode (bimodal or
U

multimodal), or no mode at all if all values occur with the same

G

frequency.
TS

MODE OF UNGROUPED DATA

A

The Mode of an ungrouped data can be found by observation i.e by

ST

checking the highest occurring value manually.

Example 1.31 : Evaluate the mode for the set of data : 13,12,7,9,10,13,13,18,8
Solution
T
A

By Observation, we can see that the highest occurring value is 13.

TH

Therefore, the modal value is 13

MODE OF GROUPED DATA

The Mode of a grouped data is given by the formula below;
1
𝐷
𝑀𝑜𝑑𝑒 = 𝐿𝑖 + ( 𝐷 +𝐷 )𝐶
1 2

where , 𝐿𝑖 = 𝐿𝑜𝑤𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑏𝑜𝑢𝑛𝑑𝑎𝑟𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑀𝑜𝑑𝑎𝑙 𝐶𝑙𝑎𝑠𝑠

11
 STA 201/209/221/229 TUTORIAL NOTE

𝐷1 = 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑡ℎ𝑒 𝑚𝑜𝑑𝑎𝑙 𝑐𝑙𝑎𝑠𝑠 𝑎𝑛𝑑 𝑡ℎ𝑒 𝑐𝑙𝑎𝑠𝑠 𝑏𝑒𝑓𝑜𝑟𝑒 𝑖𝑡

𝐷2 = 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑡ℎ𝑒 𝑚𝑜𝑑𝑎𝑙 𝑐𝑙𝑎𝑠𝑠 𝑎𝑛𝑑 𝑡ℎ𝑒 𝑐𝑙𝑎𝑠𝑠 𝑎𝑓𝑡𝑒𝑟 𝑖𝑡
𝐶 = 𝐶𝑙𝑎𝑠𝑠 𝑤𝑖𝑑𝑡ℎ

Example 1.32 : The following data represent the weight of products

manufactured in the Factory ( in Kg ). Evaluate the Mode

)
3
Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99

51
ht

6
Freq 20 41 98 149 192 197 152 97 44 10

01
34
Solution:

80
Class Frequency LCB - HCB
Interval
(0
0-9 20 -0.5 - 9.5
Y

10-19 41 9.5 - 19.5

U

20-29 98 19.5 - 29.5

G

30-39 149 29.5 - 39.5

TS

40-49 192 39.5 - 49.5

A

50-59 197 49.5 - 59.5

ST

60-69 152 59.5 - 69.5

T

70-79 97 69.5 - 79.5

A
TH

80-89 44 79.5 - 89.5

90-99 10 89.5 - 99.5

Σ𝑓 = 1000

12
 STA 201/209/221/229 TUTORIAL NOTE

By looking at this we can see that the class with the highest frequency
50-59 with a frequency of 197
Modal Class = 50-59
To calculate the Modal value we make use of the Formula
1
𝐷
𝑀𝑜𝑑𝑒 = 𝐿𝑖 + ( 𝐷 +𝐷 )𝐶
1 2

𝐿𝑖 = 49. 5 𝐷1 = 197 − 192 = 5 𝐷2 = 197 − 152 = 45 𝐶 = 10

)
3
5
𝑀𝑜𝑑𝑒 = 49. 5 + ( 5+45 )10

51
5
= 49. 5 + ( 50 )10

6
= 49. 5 + (0. 1)10

01
= 49. 5 + 1 = 59. 5

34
THE GEOMETRIC MEAN

80
The geometric mean is the nth root of the product of all values, typically
used for data that is multiplicative or for growth rates.
(0
𝑁
𝐺𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐 𝑀𝑒𝑎𝑛 = 𝑥1𝑥2... 𝑥𝑛
Y
U

Example 1.41 : Calculate the Geometric mean for the set of data :
G

13,12,24,7,9,10,13,13,18,8
TS

Solution:
A

10
𝐺𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐 𝑀𝑒𝑎𝑛 = (13)(12)(24)(7)(9)(10)(13)(13)(18)(8)
ST

10 10
= 5. 7401𝑥10
T

= 11. 91
A

THE HARMONIC MEAN

TH

The harmonic mean is the reciprocal of the average of the reciprocals of

the data values. It is particularly useful in averaging ratios or rates.
𝑁
𝐻𝑎𝑟𝑚𝑜𝑛𝑖𝑐 𝑀𝑒𝑎𝑛 = 1
Σ𝑛
𝑖

𝑁
𝐻𝑎𝑟𝑚𝑜𝑛𝑖𝑐 𝑀𝑒𝑎𝑛 = 1 1 1
𝑛1
+ 𝑛2
+...+ 𝑛
𝑛

13
 STA 201/209/221/229 TUTORIAL NOTE

Example 1.51 : Calculate the Harmonic mean for the set of data :
13,12,24,7,9,10,13,13,18,8
𝑁
𝐻𝑎𝑟𝑚𝑜𝑛𝑖𝑐 𝑀𝑒𝑎𝑛 = 1 1 1
𝑛1
+ 𝑛2
+...+ 𝑛
𝑛

10
𝐻𝑎𝑟𝑚𝑜𝑛𝑖𝑐 𝑀𝑒𝑎𝑛 = 1 1 1 1 1 1 1 1 1 1
13
+ 12 + 24 + 7 + 9 + 10 + 13 + 13 + 18 + 8
10
= = 11. 23

)
4861
5460

3
51
MEASURE OF PARTITION

6
Measures of partition, also known as quantiles, divide a data set into

01
equal parts, each representing a fraction of the data. The most common
measures of partition are quartiles, percentiles, and deciles. These

34
measures help to understand the distribution and spread of data,

80
indicating where a particular value lies within the overall distribution.
They can be applied to both grouped and ungrouped data, but the
(0
methods for calculation differ slightly.

QUARTILE
Y
U

Quartiles divide the data into four equal parts. The first quartile (Q1)
represents the 25th percentile, the second quartile (Q2) represents the
G

50th percentile or the median, and the third quartile (Q3) represents
TS

the 75th percentile.

A

QUARTILE FOR UNGROUPED DATA

ST

To calculate the Quartile for ungrouped data you'll need to arrange the
data in Ascending data
T

𝑛+1 𝑡ℎ
A

For the First Quartile (Q1) : ( 4

) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
TH

2(𝑛+1) 𝑡ℎ
For the Second Quartile (Q2) : ( 4
) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
3(𝑛+1) 𝑡ℎ
For the Third Quartile (Q3) : ( 4
) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
Example 1.61 : Evaluate Q1, Q2 and Q3 for the set of data : 13,12,7,9,10,13,8
Solution
Arranging the data in ascending order we have ; 7,8,9,10,12,13,13

14
 STA 201/209/221/229 TUTORIAL NOTE

𝑛+1 𝑡ℎ
𝑄1 = ( 4
) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
7+1 𝑡ℎ
𝑄1 = ( 4
) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
8 𝑡ℎ
𝑄1 = ( 4 ) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
𝑛𝑑
𝑄1 = 2 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛
The value in the second position is 8. Therefore Q1 is 8

)
Follow the same steps to evaluate Q2 and Q3

3
51
QUARTILE FOR GROUPED DATA

6
Quartile for Ungrouped data is similar to that of Grouped data. Here is

01
the formula to evaluate the nth quartile (1st, 2nd, 3rd..)

34
𝑛𝑁
− 𝐶𝑓𝑏𝑄𝑛

80
4
𝑄𝑛 = 𝐿𝑄 + ( 𝑓𝑄𝑛
)𝑐
𝑛

where , 𝐿𝑄 = 𝐿𝑜𝑤𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑏𝑜𝑢𝑛𝑑𝑎𝑟𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑄𝑢𝑎𝑟𝑡𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

(0
𝑛

𝑁 = 𝑇𝑜𝑡𝑎𝑙 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑑𝑎𝑡𝑎

Y

𝐶𝑓𝑏𝑄𝑛 = 𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑄𝑢𝑎𝑟𝑡𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

U

𝑓𝑄𝑛 = 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑄𝑢𝑎𝑟𝑡𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

G

𝑐 = 𝐶𝑙𝑎𝑠𝑠 𝑤𝑖𝑑𝑡ℎ
TS

For the first quartile ( n = 1 ) then the formula becomes

A

𝑁
4
− 𝐶𝑓𝑏𝑄1
𝑄1 = 𝐿𝑄 + ( )𝑐
ST

𝑓𝑄1
1

For the Second Quartile ( n = 2 ) and so on…

T
A

Example 1.62 : The following data represent the weight of products

TH

manufactured in the Factory ( in Kg ). Evaluate Q3

Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99
ht

Freq 20 41 98 149 192 197 152 97 44 10

15
 STA 201/209/221/229 TUTORIAL NOTE

Solution:

Class Frequency LCB - HCB CF

Interval

0-9 20 -0.5 - 9.5 20

10-19 41 9.5 - 19.5 61

)
3
20-29 98 19.5 - 29.5 159

51
30-39 149 29.5 - 39.5 308

6
01
40-49 192 39.5 - 49.5 500

34
50-59 197 49.5 - 59.5 697

60-69 152 59.5 - 69.5 849

70-79 97 80
69.5 - 79.5 946
(0
80-89 44 79.5 - 89.5 990
Y

90-99 10 89.5 - 99.5 1000

U

Σ𝑓 = 1000
G
TS

3𝑁
4
− 𝐶𝑓𝑏𝑄3
𝑄3 = 𝐿𝑄 + ( 𝑓𝑄3
)𝑐
A

3𝑁 𝑡ℎ
ST

To get the class of the 3rd Quartile we evaluate the ( 4

) 𝑖𝑡𝑒𝑚 =
3(1000) 𝑡ℎ
( ) 𝑖𝑡𝑒𝑚 = 750𝑡ℎ 𝑖𝑡𝑒𝑚
T

4
The 750th item will be in class 60-69. Therefore
A

𝐿𝑄 = 59. 5 𝐶𝑓𝑏𝑄3 = 697 𝑓𝑄3 = 152 𝑁 = 1000 𝑐 = 69. 5 − 59. 5 = 10

TH

3
750 − 697
𝑄3 = 59. 5 + ( 152
)10
53
𝑄3 = 59. 5 + ( 152 )10
265
𝑄3 = 59. 5 + ( 76
)
𝑄3 = 59. 5 + 3. 49

16
 STA 201/209/221/229 TUTORIAL NOTE

𝑄3 = 62. 99
Evaluate Q1 by yourself using these same step

PERCENTILE
Percentiles divide the data into hundred equal parts.

)
PERCENTILE FOR UNGROUPED DATA

3
To calculate the Percentile for ungrouped data you'll need to arrange

51
the data in Ascending data

6
𝑛+1 𝑡ℎ
For the First Percentile (P1) : ( ) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛

01
100
2(𝑛+1) 𝑡ℎ
For the Second Percentile (P2) : ( ) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛

34
100
3(𝑛+1) 𝑡ℎ
For the Third Percentile (P3) : ( ) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛 𝑎𝑛𝑑 𝑠𝑜 𝑜𝑛..

80
100

PERCENTILE FOR GROUPED DATA

(0
Percentile for Ungrouped data is similar to that of Grouped data. Here is
the formula to evaluate the nth quartile (1st, 2nd, 3rd…, 99th)
Y
U

𝑛𝑁
100
− 𝐶𝑓𝑏𝑃𝑛
G

𝑃𝑛 = 𝐿𝑃 + ( 𝑓𝑃𝑛
)𝑐
𝑛

where , 𝐿𝑃 = 𝐿𝑜𝑤𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑏𝑜𝑢𝑛𝑑𝑎𝑟𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑃𝑒𝑟𝑐𝑒𝑛𝑡𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

TS

𝑁 = 𝑇𝑜𝑡𝑎𝑙 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑑𝑎𝑡𝑎

A

𝐶𝑓𝑏𝑃𝑛 = 𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑝𝑒𝑟𝑐𝑒𝑛𝑡𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

ST

𝑓𝑃𝑛 = 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑝𝑒𝑟𝑐𝑒𝑛𝑡𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

T

𝑐 = 𝐶𝑙𝑎𝑠𝑠 𝑤𝑖𝑑𝑡ℎ
A
TH

For the first percentile ( n = 1 ) then the formula becomes

𝑁
100
− 𝐶𝑓𝑏𝑃1
𝑃1 = 𝐿𝑃 + ( 𝑓𝑃1
)𝑐
1

For the Second Percentile ( n = 2 ) and so on…

Getting the Percentile for Grouped data is the same method as quartile
the only difference is that you divide by 100 instead of 4.

17
 STA 201/209/221/229 TUTORIAL NOTE

DECILE
Deciles divide the data into ten equal parts.

DECILE FOR UNGROUPED DATA

To calculate the Decile for ungrouped data you'll need to arrange the
data in Ascending data
𝑛+1 𝑡ℎ

)
For the First Decile (D1) : ( 10
) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛

3
2(𝑛+1) 𝑡ℎ

51
For the Second Decile (D2) : ( 10
) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛

6
3(𝑛+1) 𝑡ℎ
For the Third Decile (D3) : ( ) 𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛 𝑎𝑛𝑑 𝑠𝑜 𝑜𝑛..

01
10

34
PERCENTILE FOR GROUPED DATA
Percentile for Ungrouped data is similar to that of Grouped data. Here is

80
the formula to evaluate the nth quartile (1st, 2nd, 3rd…, 9th)
(0
𝑛𝑁
10
− 𝐶𝑓𝑏𝐷𝑛
𝐷𝑛 = 𝐿𝐷 + ( 𝑓𝐷𝑛
)𝑐
𝑛
Y

where , 𝐿𝐷 = 𝐿𝑜𝑤𝑒𝑟 𝑐𝑙𝑎𝑠𝑠 𝑏𝑜𝑢𝑛𝑑𝑎𝑟𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑑𝑒𝑐𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

U

𝑁 = 𝑇𝑜𝑡𝑎𝑙 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑑𝑎𝑡𝑎

G

𝐶𝑓𝑏𝐷𝑛 = 𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑑𝑒𝑐𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

TS

𝑓𝐷𝑛 = 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑡ℎ 𝑑𝑒𝑐𝑖𝑙𝑒 𝐶𝑙𝑎𝑠𝑠

A

𝑐 = 𝐶𝑙𝑎𝑠𝑠 𝑤𝑖𝑑𝑡ℎ
ST

For the first percentile ( n = 1 ) then the formula becomes

𝑁
T

10
− 𝐶𝑓𝑏𝐷1
𝐷1 = 𝐿𝐷 + ( )𝑐
A

𝑓𝐷1
1

For the Second Decile ( n = 2 ) and so on…

TH

Getting the Decile for Grouped data is the same method as quartile, the
only difference is that you divide by 10 instead of 4.

MEASURE OF VARIATION OR DISPERSION

Measures of dispersion describe the spread or variability of a data set,
indicating how much the data points differ from the central tendency

18
 STA 201/209/221/229 TUTORIAL NOTE

(mean, median, mode). Understanding dispersion helps in assessing

the reliability and consistency of the data.

THE RANGE
It is the difference between the maximum and minimum values in a
data set.

)
𝑅𝑎𝑛𝑔𝑒 = 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑣𝑎𝑙𝑢𝑒 − 𝑀𝑖𝑛𝑖𝑚𝑢𝑚 𝑉𝑎𝑙𝑢𝑒

3
Example 1.71 :Calculate the Range for the set of data :

51
13,12,24,7,9,10,13,13,18,8

6
01
𝑀𝑖𝑛𝑖𝑚𝑢𝑚 𝑉𝑎𝑙𝑢𝑒 = 8 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑉𝑎𝑙𝑢𝑒 = 24

34
𝑅𝑎𝑛𝑔𝑒 = 24 − 8 = 16

80
VARIANCE
Variance measures the average squared deviation of each data point
(0
from the mean. It quantifies how much the data points spread out
around the mean.
Y
U

VARIANCE FOR UNGROUPED DATA

G

The formula for variance for any ungrouped set of data is given by
2
2
TS

Σ(𝑥−𝑥)
𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒(σ ) = 𝑁
Example 1.81 :Calculate the Variance for the set of data :
A

13,12,24,7,9,10,13,13,18,8
ST

Σ𝑥𝑖 13+12+24+7+9+10+13+13+18+8 127

𝑀𝑒𝑎𝑛(𝑥) = 𝑁
= 10
= 10
= 12. 7
T
A

𝑥 2
TH

𝑥 −𝑥 (𝑥 − 𝑥)

13 0.3 0.09

12 -0.7 0.49

24 11.3 127.69

7 -5.7 32.49

19
 STA 201/209/221/229 TUTORIAL NOTE

9 -3.7 13.69

10 -2.7 7.29

13 0.3 0.09

13 0.3 0.09

18

)
5.3 28.09

3
8 -4.7 22.09

51
2

6
Σ(𝑥 − 𝑥) = 232. 1

01
2
2 Σ(𝑥−𝑥) 232.1
𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒(σ ) = = = 23. 21

34
𝑁 10

80
VARIANCE FOR GROUPED DATA
The variance of a grouped data is given by the formula
(0
2
2 Σ𝑓(𝑥−𝑥)
𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒(σ ) = Σ𝑓
Example 1.82 : The following data represent the weight of products
Y

manufactured in the Factory ( in Kg ). Evaluate the Variance

U
G

Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99
TS

ht

Freq 20 41 98 149 192 197 152 97 44 10

A
ST

Solution:
T

Class x
A

𝑓𝑥
Interva
TH

2 2
l 𝑓 (𝑥 − 𝑥) (𝑥 − 𝑥) 𝑓(𝑥 − 𝑥)
0-9 20 4.5 90 -44.7 1998.09 39961.8
10-19 41 14.5 594.5 -34.7 1204.09 49367.69
20-29 98 24.5 2401 -24.7 610.09 59788.82
30-39 149 34.5 5140.5 -14.7 216.09 32197.41

20
 STA 201/209/221/229 TUTORIAL NOTE

40-49 192 44.5 8544 -4.7 22.09 4241.28

50-59 197 54.5 10736.5 5.3 28.09 5533.73
60-69 152 64.5 9804 15.3 234.09 35581.68
70-79 97 74.5 7226.5 25.3 640.09 62088.73
80-89 44 84.5 3718 35.3 1246.09 54827.96

)
90-99 10 94.5 945 45.3 2052.09 20520.9

3
Σ𝑓𝑥 = 492 2

51
Σ𝑓 = 1000 Σ𝑓(𝑥 − 𝑥) == 36411

6
01
Σ𝑓𝑥 49200
Mean = Σ𝑓
= 1000
= 49, 2
2

34
2 Σ𝑓(𝑥−𝑥) 364110
𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒(σ ) = Σ𝑓
= 1000
= 364. 11

STANDARD DEVIATION
80
Standard deviation is the square root of variance and provides a
(0
measure of dispersion in the same units as the original data. The
standard deviation can be found by finding the square root of the
Y

variance i.e 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒

U
G

STANDARD DEVIATION FOR UNGROUPED DATA

TS

The formula for standard deviation for any ungrouped set of data is
given by
A

2
ST

Σ(𝑥−𝑥)
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛(σ) = 𝑁
Example 1.91 :Calculate the Standard Deviation for the set of data :
T

13,12,24,7,9,10,13,13,18,8
A

From Example 1.81 we found the variance to be 12.7 therefore

TH

𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = 12. 7 = 3. 56

STANDARD DEVIATION FOR GROUPED DATA

The standard deviation of a grouped data is given by the formula
2
Σ𝑓(𝑥−𝑥)
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛(σ) = Σ𝑓

21
 STA 201/209/221/229 TUTORIAL NOTE

Example 1.92 : The following data represent the weight of products

manufactured in the Factory ( in Kg ). Evaluate the standard deviation

Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99
ht

)
Freq 20 41 98 149 192 197 152 97 44 10

3
51
Solution:

6
From Example 1.82 we found the variance to be 364.11 therefore

01
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = 364. 11 = 19. 08

34
COEFFICIENT OF VARIATION

80
The coefficient of variation is the ratio of the standard deviation to the
mean, expressed as a percentage. It measures the relative variability of
(0
the data
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
𝐶𝑂𝑉 = ( 𝑀𝑒𝑎𝑛
)𝑥100
Y
U

INTERQUARTILE RANGE (IQR)

G

The IQR measures the range of the middle 50% of the data, calculated
TS

as the difference between the third quartile (Q3) and the first quartile
(Q1).
A

𝐼𝑄𝑅 = 𝑄3 − 𝑄1
ST

SEMI INTERQUARTILE RANGE OR QUARTILE DEVIATION

T

Quartile deviation is half of the interquartile range and provides a

A

measure of the spread of the middle 50% of the data.

TH

𝑄3−𝑄1
𝑄𝑢𝑎𝑟𝑡𝑖𝑙𝑒 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = 2

MEAN DEVIATION OR MEAN ABSOLUTE DEVIATION

The mean absolute deviation is the average of the absolute differences
between each data point and the mean.

22
 STA 201/209/221/229 TUTORIAL NOTE

MEAN DEVIATION FOR UNGROUPED DATA

The formula for Mean deviation for any ungrouped set of data is given
by
Σ|𝑥−𝑥|
𝑀𝑒𝑎𝑛 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = 𝑁
Example :Calculate the Mean Deviation for the set of data :
13,12,24,7,9,10,13,13,18,8

)
Σ𝑥𝑖 13+12+24+7+9+10+13+13+18+8 127
𝑀𝑒𝑎𝑛(𝑥) = = = = 12. 7

3
𝑁 10 10

6 51
𝑥 𝑥 −𝑥 |𝑥 − 𝑥|

01
13 0.3 0.3

34
12 -0.7 0.7

24 11.3
80
11.3
(0
7 -5.7 5.7

9 -3.7 3.7
Y
U

10 -2.7 2.7
G

13 0.3 0.3
TS

13 0.3 0.3
A

18 5.3 5.3
ST

8 -4.7 4.7
T

Σ|𝑥 − 𝑥| = 35
A

Σ|𝑥−𝑥| 35
TH

𝑀𝑒𝑎𝑛 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = 𝑁
= 10
= 3. 5

MEAN DEVIATION FOR GROUPED DATA

The mean deviation of a grouped data is given by the formula
Σ𝑓|𝑥−𝑥|
𝑀𝑒𝑎𝑛 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = Σ𝑓

23
 STA 201/209/221/229 TUTORIAL NOTE

Example 1.92 : The following data represent the weight of products

manufactured in the Factory ( in Kg ). Evaluate the mean deviation

Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99
ht

Freq 20 41 98 149 192 197 152 97 44 10

)
3
Solution:

6 51
Class

01
Interval 𝑓 𝑥 𝑓𝑥 𝑥 − 𝑥 |𝑥 − 𝑥| 𝑓|𝑥 − 𝑥|

34
0-9 20 4.5 90 -44.7 44.7 894
10-19 41 14.5 594.5 -34.7 34.7 1422.7
20-29 98 24.5 2401
80
-24.7 24.7 2420.6
(0
30-39 149 34.5 5140.5 -14.7 14.7 2190.3
40-49 192 44.5 8544 -4.7 4.7 902.4
Y

50-59 197 54.5 10736.5 5.3 5.3 1044.1

U

60-69 152 64.5 9804 15.3 15.3 2325.6

G

70-79 97 74.5 7226.5 25.3 25.3 2454.1

TS

80-89 44 84.5 3718 35.3 35.3 1553.2

A

90-99 10 94.5 945 45.3 45.3 453

ST

Σ𝑓 = 1000 Σ𝑓𝑥 = 49200 Σ𝑓|𝑥 − 𝑥| = 15660

Σ𝑓𝑥 49200
T

Mean = Σ𝑓
= 1000
= 49, 2
A

Σ𝑓|𝑥−𝑥| 15660
𝑀𝑒𝑎𝑛 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = = = 15. 66
TH

Σ𝑓 1000

I only did my best as at this time, this should not limit your reading to this
guide only.

24
 STA 201/209/221/229 TUTORIAL NOTE

Want to join me and your peers as we go on a challenge of solving 500+

likely to come out in your forthcoming exams? Text me on WhatsApp at
08034016513 with the code “500 MCQ” and wait for a reply. The earlier, the
better. Take action now!!

Note: This is different from the Tutorials

)
3
6 51
01
34
THIS MATERIAL IS NOT TO BE SOLD FOR 80
(0
ANY REASON
Y
U
G
TS
A
ST
T
A
TH

25
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 3 : SKEWNESS AND KURTOSIS

MEASURE OF SYMMETRY
In statistics, the symmetry of a data distribution refers to how evenly the
data is spread around the central point. A perfectly symmetrical
distribution is one where the left side of the distribution mirrors the

)
right side. The concept of symmetry is essential for understanding the

3
shape of a data distribution and for making assumptions about the

51
data when applying statistical methods. Measures of symmetry

6
quantify the degree to which a distribution deviates from perfect

01
symmetry.

34
SKEWNESS
Skewness is the most common measure of symmetry in a data

80
distribution. It indicates the direction and degree of asymmetry. If a
distribution is symmetrical, the skewness is zero. If it is not, skewness
(0
tells us whether the data is skewed to the left (negatively skewed) or to
the right (positively skewed).
Y
U

TYPES OF SKEWNESS
G

Positive Skewness: The distribution has a longer tail on the right side.
TS

Most data points are clustered on the left, with a few larger values
pulling the mean to the right.
A

Example: Income distributions often exhibit positive skewness because

ST

a small number of individuals have much higher incomes than the

majority.
T
A

Negative Skewness: The distribution has a longer tail on the left side.
TH

Most data points are clustered on the right, with a few smaller values
pulling the mean to the left.

Example: Test scores where most students perform well, but a few have
low scores.

If Skewness = 0 , the distribution is perfectly symmetrical

26
 STA 201/209/221/229 TUTORIAL NOTE

If Skewness > 0 the distribution is Positively Skewed

If Skewness < 0 the distribution is Negatively Skewed

)
3
6 51
01
34
80
For a Negatively skewed distribution Mean < Median < Mode
(0
For a Positively Skewed Distribution Mean > Median >Mode
Y

For a Normal Distribution (No Skewness) Mean = Median = Mode

U

METHODS OF EVALUATING SKEWNESS

G

There are various methods to calculate skewness but here we’ll be

TS

treating just two out of the most common methods

A

Karl-Pearson Coefficient of Skewness:

ST

Karl Pearson's Coefficient of Skewness is a measure that indicates the

degree and direction of skewness in a data set. It compares the mean,
T

median, and mode of the distribution to determine whether the data is

A

skewed to the left (negatively skewed) or to the right (positively

TH

skewed). Below is the Karl-Pearson's formula

𝑀𝑒𝑎𝑛−𝑀𝑜𝑑𝑒
𝑆𝑘𝑒𝑤𝑛𝑒𝑠𝑠 = 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
OR
3(𝑀𝑒𝑎𝑛−𝑀𝑒𝑑𝑖𝑎𝑛)
𝑆𝑘𝑒𝑤𝑛𝑒𝑠𝑠 = 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛

27
 STA 201/209/221/229 TUTORIAL NOTE

From the formula above we can say that

𝑀𝑒𝑎𝑛 − 𝑀𝑜𝑑𝑒 ≈ 3(𝑀𝑒𝑎𝑛 − 𝑀𝑒𝑑𝑖𝑎𝑛)

METHODS OF MOMENT

3
𝑀3 Σ𝑓(𝑥−𝑥)
𝑆𝑘𝑒𝑤𝑛𝑒𝑠𝑠(ψ3) = 3 = 3

)
2
𝑀22 [Σ𝑓(𝑥−𝑥) ]
2

3
Example 2.0 : The following data represent the weight of products

51
manufactured in the Factory ( in Kg ). Evaluate the Skewness using

6
method of Moments

01
Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99

34
ht

80
Freq 20 41 98 149 192
(0 197 152 97 44 10

Solution:
Y

Interv
2 2 3 3
al f x fx
U

𝑥 −𝑥 (𝑥 − 𝑥) 𝑓(𝑥 − 𝑥) (𝑥 − 𝑥) 𝑓(𝑥 − 𝑥)
G

0-9 20 4.5 90 -44.7 1998.09 39961.8 -89314.623 -1786292.46

10-19 41 14.5 594.5 -34.7 1204.09 49367.69 -41781.923 -1713058.843
TS

20-29 98 24.5 2401 -24.7 610.09 59788.82 -15069.223 -1476783.854

A

30-39 149 34.5 5140.5 -14.7 216.09 32197.41 -3176.523 -473301.927

ST

40-49 192 44.5 8544 -4.7 22.09 4241.28 -103.823 -19934.016

50-59 197 54.5 10736.5 5.3 28.09 5533.73 148.877 29328.769
T

60-69 152 64.5 9804 15.3 234.09 35581.68 3581.577 544399.704

A

70-79 97 74.5 7226.5 25.3 640.09 62088.73 16194.277 1570844.869

TH

80-89 44 84.5 3718 35.3 1246.09 54827.96 43986.977 1935426.988

90-99 10 94.5 945 45.3 2052.09 20520.9 92959.677 929596.77
Σ 1000 49200 364110 -459774

3
𝑀3 Σ𝑓(𝑥−𝑥) −459774 −459774
𝑆𝑘𝑒𝑤𝑛𝑒𝑠𝑠(ψ3) = 3 = 3 = 3 = 219709537.6
=− 0. 00209
2 2 2 2
𝑀2 [Σ𝑓(𝑥−𝑥) ] (364110)

28
 STA 201/209/221/229 TUTORIAL NOTE

KURTOSIS
While primarily a measure of the "peakedness" of a distribution, kurtosis
also provides insights into the symmetry of data. It complements
skewness by focusing on the tails of the distribution rather than the
symmetry around the center.

)
Types of Kurtosis:

3
51
Leptokurtic (Kurtosis > 3): Distributions with heavy tails and a sharp

6
peak, indicating that data values are concentrated around the mean

01
with a higher likelihood of extreme values.

34
Mesokurtic (Kurtosis = 3): A distribution with tails similar to a normal
distribution, such as the standard bell curve.

80
Platykurtic (Kurtosis < 3): Distributions with light tails and a flatter peak,
(0
indicating that data values are more spread out and less likely to be
extreme.
Y
U
G
TS
A
ST
T

𝑀4
A

𝐾𝑢𝑟𝑡𝑜𝑠𝑖𝑠(φ4) = 2
𝑀2
TH

4 2
Σ𝑓(𝑥−𝑥) Σ𝑓(𝑥−𝑥)
𝑤ℎ𝑒𝑟𝑒 𝑀4 = Σ𝑓
𝑎𝑛𝑑 𝑀2 = Σ𝑓

Example 2.1 : The following data represent the weight of products

manufactured in the Factory ( in Kg ). Evaluate the Kurtosis

29
 STA 201/209/221/229 TUTORIAL NOTE

Weig 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99
ht

Freq 20 41 98 149 192 197 152 97 44 10

Solution:

)
Interv
2 2 4 4

3
al f x fx 𝑥 −𝑥 (𝑥 − 𝑥) 𝑓(𝑥 − 𝑥) (𝑥 − 𝑥) 𝑓(𝑥 − 𝑥)

51
0-9 20 4.5 90 -44.7 1998.09 39961.8 3992363.648 79847272.96

6
10-19 41 14.5 594.5 -34.7 1204.09 49367.69 1449832.728 59443141.85

01
20-29 98 24.5 2401 -24.7 610.09 59788.82 372209.8081 36476561.19

30-39 149 34.5 5140.5 -14.7 216.09 32197.41

34
46694.8881 6957538.327

40-49 192 44.5 8544 -4.7 22.09 4241.28 487.9681 93689.8752

80
50-59 197 54.5 10736.5 5.3 28.09 5533.73 789.0481 155442.4757

60-69 152 64.5 9804 15.3 234.09 35581.68 54798.1281 8329315.471

(0
70-79 97 74.5 7226.5 25.3 640.09 62088.73 409715.2081 39742375.19

80-89 44 84.5 3718 35.3 1246.09 54827.96 1552740.288 68320572.68

Y

90-99 10 94.5 945 45.3 2052.09 20520.9 4211073.368 42110733.68

U

Σ 1000 49200 364110 341476643.7

G
TS

4
Σ𝑓(𝑥−𝑥) 341476643.7
𝑀4 = Σ𝑓
= 1000
= 341476. 6437
A

2
Σ𝑓(𝑥−𝑥) 364110
𝑀2 = = = 364. 11
ST

Σ𝑓 1000
𝑀4 341476.6437
𝐾𝑢𝑟𝑡𝑜𝑠𝑖𝑠(φ4) = 2 = 2 = 2. 58 < 3(𝑃𝑙𝑎𝑡𝑦𝑘𝑢𝑟𝑡𝑖𝑐)
T

𝑀2 (364.11)
A
TH

I only did my best as at this time, this should not limit your reading to this
guide only.

Want to join me and your peers as we go on a challenge of solving 500+

likely to come out in your forthcoming exams? Text me on WhatsApp at
08034016513 with the code “500 MCQ” and wait for a reply. The earlier, the
better. Take action now!!
Note: This is different from the Tutorials

30
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 4 : PROBABILITY
Probability is often described as the Mathematics of Chance or the
likeliness of an event to occur. We shall develop foundation knowledge
of Probability in this Lesson

Any result of an experiment in probability is known as an outcome. If

we cannot predict beforehand the outcome of an experiment, the

)
3
experiment is called a Random Experiment.

51
The set of all possible outcomes of any random experiment will be
called a sample space and it will be denoted as S. A subset of the

6
sample space which may be a collection of Outcomes is called an Event

01
Space, and it will be denoted by A.

34
The Probability of an Event A denoted by P(A) is defined as

𝑃(𝐴) =
𝑛(𝐴)
80
(0
𝑛(𝑆)

GENERAL PROBABILITY RULES

Y

If A is an event defined on a Sample Space

U

1. 𝑃(𝐴) ≥ 0
G

2. 0 ≤ 𝑃(𝐴) ≤ 1
TS

𝑐 𝑐
3. 1 − 𝑃(𝐴 ) = 𝑃(𝐴) 𝑤ℎ𝑒𝑟𝑒 𝐴 = 𝐶𝑜𝑚𝑝𝑙𝑒𝑚𝑒𝑛𝑡 𝑜𝑓 𝑒𝑣𝑒𝑛𝑡 𝐴
Let’s treat a few examples
A
ST

Example 3.0
1. In a single throw of a fair coin, find the probability that: (a) a head
T

appears
A

2. In a single throw of two fair coins, find the probability that: (a) two
TH

heads appear (b) one head and one tail appears

3. Two fair dice are tossed, find the probability that the total sum
score that shows up is: (a) 8 (b) a prime number (c) greater than or
equal to 9
4. In a single throw of three fair coins find the probability that one
head two tails appear

31
 STA 201/209/221/229 TUTORIAL NOTE

Solutions:
1. 𝐿𝑒𝑡 𝐴 𝑏𝑒 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑎 ℎ𝑒𝑎𝑑 𝑜𝑐𝑐𝑢𝑟𝑠
𝑆 = {𝐻, 𝑇} 𝑛(𝑆) = 2 𝑛(𝐴) = 1
𝑛(𝐴) 1
𝑃(𝐴) = 𝑛(𝑆)
= 2
= 0. 5
2. 𝐿𝑒𝑡 𝐵 𝑏𝑒 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑡𝑤𝑜 ℎ𝑒𝑎𝑑𝑠 𝑜𝑐𝑐𝑢𝑟𝑠

)
𝑆 = {𝐻𝐻, 𝑇𝐻, 𝐻𝑇. 𝑇𝑇} 𝑛(𝑆) = 4 𝑛(𝐵) = 1

3
51
𝑛(𝐵) 1
𝑃(𝐴) = 𝑛(𝑆)
= 4
= 0. 25

6
01
𝐿𝑒𝑡 𝐶 𝑏𝑒 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑜𝑛𝑒 ℎ𝑒𝑎𝑑 𝑜𝑛𝑒 𝑡𝑎𝑖𝑙 𝑜𝑐𝑐𝑢𝑟𝑠

34
𝑆 = {𝐻𝐻, 𝑇𝐻, 𝐻𝑇. 𝑇𝑇} 𝑛(𝑆) = 4 𝑛(𝐶) = 2
𝑛(𝐶) 2
𝑃(𝐴) = = = 0. 5

80
𝑛(𝑆) 4
3. Let’s construct a table for the sums
(0
+ 1 2 3 4 5 6
Y

1 2 3 4 5 6 7
U

2 3 4 5 6 7 8
G

3 4 5 6 7 8 9
TS

4 5 6 7 8 9 10
A
ST

5 6 7 8 9 10 11

6 7 8 9 10 11 12
T
A

𝑛(𝑠) = 36
TH

(a) 𝐿𝑒𝑡 𝐷 𝑏𝑒 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑎 𝑠𝑢𝑚 𝑜𝑓 8 𝑜𝑐𝑐𝑢𝑟𝑠

𝑛(𝐴) = 5
𝑛(𝐷) 5
𝑃(𝐴) = 𝑛(𝑆)
= 36
= 0. 139
(b)𝐿𝑒𝑡 𝐸 𝑏𝑒 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑎 𝑝𝑟𝑖𝑚𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑐𝑐𝑢𝑟𝑠
𝑛(𝐴) = 15

32
 STA 201/209/221/229 TUTORIAL NOTE

𝑛(𝐸) 15
𝑃(𝐴) = 𝑛(𝑆)
= 36
= 0. 417
4. 𝐿𝑒𝑡 𝐹 𝑏𝑒 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑎 ℎ𝑒𝑎𝑑 𝑎𝑛𝑑 𝑡𝑤𝑜 𝑡𝑎𝑖𝑙𝑠 𝑜𝑐𝑐𝑢𝑟𝑠
𝑆 = {𝐻𝐻𝐻, 𝐻𝐻𝑇, 𝐻𝑇𝐻, 𝐻𝑇𝑇, 𝑇𝐻𝐻, 𝑇𝐻𝑇, 𝑇𝑇𝐻, 𝑇𝑇𝑇} 𝑛(𝑆) = 8 𝑛(𝐹) = 3
𝑛(𝐹) 3
𝑃(𝐴) = 𝑛(𝑆)
= 8
= 0. 375

COMBINATION OF EVENTS

)
Two or more events which are subsets of a sample space can be

3
combined using a set notation ∩ 𝑜𝑟 ∪. For example, if 𝐸 ⊂ 𝑆 and 𝐹 ⊂ 𝑆

51
(a) 𝐸 ∩ 𝐹 𝑖𝑠 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑡ℎ𝑎𝑡 𝑒𝑖𝑡ℎ𝑒𝑟 𝐸 𝑜𝑟 𝐹 𝑜𝑟 𝑏𝑜𝑡ℎ 𝑜𝑐𝑐𝑢𝑟

6
(b)𝐸 ∪ 𝐹 𝑖𝑠 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 𝑏𝑜𝑡ℎ 𝐸 𝑎𝑛𝑑 𝐹 𝑜𝑐𝑐𝑢𝑟

01
MUTUALLY EXCLUSIVE EVENTS

34
Two events E and F are said to be mutually exclusive if they cannot

80
occur simultaneously. We say that the two sets are disjoint.
𝑖. 𝑒 𝐸 ∩ 𝐹 = ϕ
(0
Addition of two events (say E and F) is given by
Y

𝑃𝑟(𝐸 ∪ 𝐹) = 𝑃𝑟(𝐸) + 𝑃𝑟(𝐹) − 𝑃𝑟(𝐸 ∩ 𝐹)

U

𝑃𝑟(𝐸 ∩ 𝐹) = 𝑃𝑟(𝐸) × 𝑃𝑟(𝐹) (Multiplication of Events)

For Mutually Exclusive Events,
G

𝑃𝑟(𝐸 ∪ 𝐹) = 𝑃𝑟(𝐸) + 𝑃𝑟(𝐹) since 𝑃𝑟( 𝐸 ∩ 𝐹) = 0

TS

INDEPENDENT EVENTS
A

Two events E and F are said to be mutually exclusive if they can occur
ST

simultaneously.
For Independent Events,
T

𝑃𝑟(𝐸 ∪ 𝐹) = 𝑃𝑟(𝐸) + 𝑃𝑟(𝐹) − 𝑃𝑟(𝐸 ∩ 𝐹)

A
TH

CONDITIONAL PROBABILITY
Given a sample space S, Let E be a non-empty proper subset of S. The
Probability of an event F occurs after E has occurred denoted by 𝑃𝑟(𝐹|𝐸)
is defined as

𝑃𝑟(𝐹∩𝐸)
𝑃𝑟(𝐹|𝐸) = 𝑃𝑟(𝐸)

33
 STA 201/209/221/229 TUTORIAL NOTE

More exercises with solutions are in the 500 MCQ PDF. Want more? Pay
for more and set yourself ahead of 80% of your peers

Want to join me and your peers as we go on a challenge of solving 500+

likely to come out in your forthcoming exams? Text me on WhatsApp at

)
08034016513 with the code “500 MCQ” and wait for a reply. The earlier, the

3
better. Take action now!!

6 51
Note: This is different from the Tutorials

01
34
80
(0
Y

THIS MATERIAL IS NOT TO BE SOLD FOR

U
G

ANY REASON
TS
A
ST
T
A
TH

34
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 5 : PROBABILITY DISTRIBUTIONS

Before moving to Probability Distribution you need to understand what
we mean by Random Variables.

A random variable is a numerical outcome of a random phenomenon

or experiment. It assigns a specific number to each possible outcome of
an event. Random variables are essential in probability and statistics as

)
they bridge the gap between real-world randomness and mathematical

3
analysis.

6 51
There are two main types of random variables:

01
1. Discrete Random Variables

34
2. Continuous Random Variables

80
Discrete Random Variables
(0
A discrete random variable takes on a finite or countably infinite set of
values. These values are distinct and separate, meaning there are gaps
between possible outcomes. Discrete random variables are often
Y

associated with processes where the results are counted, like the
U

number of heads in a coin flip, or the number of students in a

G

classroom.
TS

Examples of Discrete Random Variables:

A

1. Number of heads when flipping a coin three times: It can take the
ST

values 0, 1, 2, or 3.
2. Number of students in a class: It can take values such as 20, 21, 22,
T

etc.
A

3. Roll of a die: A six-sided die produces outcomes like 1, 2, 3, 4, 5, and

TH

6.

Continuous Random Variables

A continuous random variable can take on any value within a given

range, often corresponding to measurements. The possible outcomes
are not countable because they form a continuum of values, and there

35
 STA 201/209/221/229 TUTORIAL NOTE

are no gaps between outcomes. Continuous random variables are often

associated with processes that involve measuring, such as time,
temperature, or height.

Examples of Continuous Random Variables:

1. Height of a person: It could be any value within a range, such as 160.3

cm, 170.45 cm, etc.

)
2. Time to complete a task: It could be 5.5 minutes, 6.7 minutes, or

3
51
any other value in between.
3. Temperature: It could be measured as 25.3°C, 25.35°C, or any other

6
value in a continuous range.

01
PROBABILITY MASS FUNCTION

34
For discrete random variables, the Probability Mass Function (PMF)

80
describes the probability of each possible value that the random
variable can take. It is denoted as:
(0
𝑃(𝑋 = 𝑥)
Y
U

Where 𝑃(𝑋 = 𝑥) gives the probability that the random variable X takes
the value x.
G
TS

For example, if X is the number rolled on a fair six-sided die, then:

A

1
𝑃𝑟(𝑋 = 1) = 𝑃𝑟(𝑋 = 2) = 𝑃𝑟(𝑋 = 3) = 𝑃𝑟(𝑋 = 4) = 𝑃𝑟(𝑋 = 5) = 𝑃𝑟(𝑋 = 6) = 6
ST

Each outcome is equally likely, and the sum of all probabilities is 1.

T

Examples of discrete distributions are Binomial distribution, Poisson

A

distribution, Bernoulli distribution etc

TH

Properties of a PMF

𝑃𝑟(𝑋 = 𝑥) ≥ 0 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 ∈ 𝑋

0 ≤ 𝑃𝑟(𝑋 = 𝑥) ≤ 1 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 ∈ 𝑋

36
 STA 201/209/221/229 TUTORIAL NOTE

𝑛
Σ𝑖=1𝑃𝑟(𝑋 = 𝑥) = 1

PROBABILITY DENSITY FUNCTION

For continuous random variables, the Probability Density Function

(PDF) describes the likelihood of the random variable taking values
within a certain range. Unlike discrete random variables, the probability

)
of a continuous random variable taking any exact value is zero. Instead,

3
probabilities are calculated over intervals.

6 51
The PDF is denoted as 𝑓(𝑥), where the probability that the random

01
variable 𝑋 lies within a particular interval [𝑎, 𝑏] is given by:

34
𝑏
𝑃𝑟(𝑎 ≤ 𝑋 ≤ 𝑏) = ∫ 𝑓(𝑥)𝑑𝑥

80
𝑎

The Normal Distribution is a continuous Distribution

(0
Properties of a PDF
Y

1. 𝑓(𝑥) ≥ 0 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 ∈ [𝑎, 𝑏]

U

2. 0 ≤ 𝑓(𝑥) ≤ 1 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 ∈ [𝑎, 𝑏]

G

∞
3. ∫ 𝑓(𝑥)𝑑𝑥 = 1 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 ∈ [𝑎, 𝑏]
TS

−∞
A

BERNOULLI DISTRIBUTION
ST

The Bernoulli distribution is a discrete probability distribution

representing two possible outcomes: success (1) and failure (0). It is
T

named after the Swiss mathematician Jacob Bernoulli. The distribution

A

is parameterized by ppp, which represents the probability of success,

TH

where 0 ≤ 𝑝 ≤ 1. The probability mass function (PMF) for a Bernoulli

distribution is:
𝑃(𝑋 = 𝑥) = 𝑝 𝑤ℎ𝑒𝑛 𝑥 = 1
𝑃(𝑋 = 𝑥) = 1 − 𝑝 𝑤ℎ𝑒𝑛 𝑥 = 0

Note:

37
 STA 201/209/221/229 TUTORIAL NOTE

● It is the simplest form of a distribution

2
● The mean is µ = 𝑝 and the variance is σ = 𝑝(1 − 𝑝) 𝑜𝑟 𝑝𝑞
● Used to model a single trial with two outcomes, such as flipping a
coin.

BINOMIAL DISTRIBUTION

)
3
The Binomial distribution extends the Bernoulli distribution to multiple

51
independent trials. It represents the number of successes in a fixed

6
number of n trials, where each trial is independent, and the probability

01
of success p remains constant. The PMF of the binomial distribution is
given by

34
𝑥 𝑛−𝑥
𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑋𝑃 (1 − 𝑝)

Where: 80
(0
● 𝑋 is the number of successes.
● 𝑛 is the number of trials.
Y

● 𝑥 is the number of successes.

U

● 𝑝 is the probability of success

G

𝑛!
● 𝑛𝐶𝑋 = (𝑛−𝑥)!𝑥!
TS
A

Properties:
ST

2
● The mean is µ = 𝑛𝑝 and the variance is σ = 𝑛𝑝(1 − 𝑝) 𝑜𝑟 𝑛𝑝𝑞
● Used when there are a fixed number of independent trials with
T

the same probability of success, like rolling a die multiple times

A

and counting how many times a specific number appears

TH

Example:
Past experience has shown that 7% of all luncheon vouchers are in error.
If a random sample of 5 vouchers is selected, what is the probability
that;
a. Exactly two will have error
b. At least two will have error

38
 STA 201/209/221/229 TUTORIAL NOTE

Solution: 𝑛 = 5 𝑝 = 0. 07 𝑞 = 1 − 0. 07 = 0. 93
𝑥 𝑖𝑠 𝑡ℎ𝑒 𝑖𝑠 𝑡ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑙𝑢𝑛𝑐ℎ𝑒𝑜𝑛𝑠 ℎ𝑎𝑣𝑖𝑛𝑔 𝑎𝑛 𝑒𝑟𝑟𝑜𝑟
a. 𝑥 = 2

2 5−2
𝑃(𝑋 = 2) = 5𝐶2(0. 07) (0. 93)

2 3
𝑃(𝑋 = 2) = 10 × (0. 07) (0. 93) = 0. 0394

)
3
b. 𝑥 ≥ 2

651
𝑃(𝑋 ≥ 2) = 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5)

01
2 3
𝑃(𝑋 = 2) = 10 × (0. 07) (0. 93) = 0. 0394

34
3 2
𝑃(𝑋 = 3) = 5𝐶3 × (0. 07) (0. 93) = 0. 002967

4
80 1
𝑃(𝑋 = 4) = 5𝐶4 × (0. 07) (0. 93) = 0. 0001116
(0
5 0
𝑃(𝑋 = 5) = 5𝐶5 × (0. 07) (0. 93) = 0. 00001681
Y
U

𝑃(𝑋 ≥ 2) = 0. 0394 + 0. 002967 + 0. 0001116 + 0. 00001681 = 0. 0425

G
TS

POISSON DISTRIBUTION
A
ST

The Poisson distribution is a discrete probability distribution that

models the number of times an event occurs within a fixed interval of
T

time or space, given that these events happen independently and the
A

average rate of occurrence (denoted as λ) is constant. The PMF of the

TH

Poisson distribution is:

𝑥 −λ
λ𝑒
𝑃(𝑋 = 𝑥) = 𝑥!

Where:

● 𝑋 is the number of events.

39
 STA 201/209/221/229 TUTORIAL NOTE

● λ is the average rate of occurrence.

● 𝑒 is Euler’s number 𝑒 ≃ 2. 718

Properties:

● Poisson experiment results in outcomes that can be classified as

success or failures.

)
● The average number of successes that occur in a specified region

3
is known as λ

51
● The probability that a success will occur in an extremely small
region is virtually zero

6
01
Example:

34
The average number of defective 100-watts electric bulbs in every
production batch is given as 4. What is the probability that the current
production batch will contain;
80
(0
a. 6 defectives
b. At most 1 defective
Y

c. Between 2 and 5 defectives

U
G

Solution:
TS

λ=4 𝑥 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑑𝑒𝑓𝑒𝑐𝑡𝑖𝑣𝑒 100 − 𝑤𝑎𝑡𝑡𝑠 𝑏𝑢𝑙𝑏

A

a. 𝑥 = 6
ST

6 −4
4𝑒 4096×(0.0183156)
𝑃(𝑋 = 6) = 6!
= 720
= 0. 1042
T
A

b. 𝑥 ≤ 1
TH

𝑃(𝑋 ≤ 1) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1)

0 −4
4𝑒 1×(0.0183156)
𝑃(𝑋 = 0) = 0!
= 1
= 0. 0183156

1 −4
4𝑒 4×(0.0183156)
𝑃(𝑋 = 1) = 1!
= 1
= 0. 0732626

40
 STA 201/209/221/229 TUTORIAL NOTE

𝑃(𝑋 ≤ 1) = 0. 0183156 + 0. 0732626 = 0. 0916

c. 2 < 𝑥 < 5

𝑃(2 < 𝑋 < 5) = 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4)

3 −4
4𝑒 64×(0.0183156)
𝑃(𝑋 = 3) = 3!
= 6
= 0. 19537

)
3
4 −4
4𝑒 256×(0.0183156)
𝑃(𝑋 = 4) = = = 0. 19537

51
4! 24

6
𝑃(2 < 𝑋 < 5) = 0. 19537 + 0. 19537 = 0. 3907

01
34
Want to join me and your peers as we go on a challenge of solving 500+
likely to come out in your forthcoming exams? Text me on WhatsApp at

80
08034016513 with the code “500 MCQ” and wait for a reply. The earlier, the
better. Take action now!!
(0
Note: This is different from the Tutorials
Y
U
G
TS
A
ST

THIS MATERIAL IS NOT TO BE SOLD FOR

T
A
TH

ANY REASON

41
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 6 : ESTIMATION
Estimation theory is a framework that uses a combination of effect
sizes, confidence intervals and precision to plan experiments, analyzes
data and interpret results originated from samples as a means of
getting the true estimate of population parameter
.
There are two forms of estimation theory namely:

)
3
1. Point Estimation

51
2. Interval Estimation

6
Some important Definitions

01
1. Sample Statistic: A sample is a subset of the population chosen

34
for study. A sample statistic is a measure that describes a
characteristic of this sample, which we use to estimate the

80
corresponding population parameter. Since a sample is more
feasible to obtain than the entire population, sample statistics are
(0
practical for data analysis. Examples of sample statistics include:
a. Sample mean (𝑥): the average value of a characteristic in the
Y

sample.
U

2
b. Sample variance (𝑠 ): a measure of data dispersion in the
G

sample.
TS

c. Sample proportion (𝑝): the fraction of the sample with a

particular characteristic.
A

Sample statistics vary from sample to sample, so they are considered

ST

variables rather than fixed values. However, with proper sampling and
analysis methods, they can serve as good estimators for population
T

parameters.
A
TH

2. Population Parameter: A population refers to the entire set of

individuals, items, or data points we are interested in studying. A
population parameter is a value that describes a characteristic of
the entire population. Since populations are often too large to
measure in full, parameters are usually theoretical values that we
aim to estimate. Examples of population parameters include

42
 STA 201/209/221/229 TUTORIAL NOTE

a. Population mean (µ): the average value of a characteristic in

the population.
2
b. Population variance (σ ): a measure of how data points are
spread out in the population.
c. Population proportion (𝑃): the fraction of the population
with a particular characteristic.
Population parameters are typically fixed but unknown values, as

)
3
measuring an entire population can be impractical or impossible.

51
Here’s a summary of populations characteristics (parameter) and their

6
corresponding sample statistic

01
34
Characteristics Parameter Statistic

80
Mean µ 𝑥
(0
Standard Deviation σ 𝑠

Variance σ
2
𝑠
2
Y
U

Proportion 𝑃 𝑝
G
TS

POINT ESTIMATION
This is the procedure that allow us to calculate a single sample statistic
A
ST

(such as 𝑥, 𝑝 𝑜𝑟 𝑠) from a selected sample to provide a best estimate of

the true value of the corresponding population parameters.
T
A

INTERVAL ESTIMATION
TH

It is the procedure of constructing intervals (i.e. Lower bound and Upper

bound) in terms of point estimate and margin of error in which the
population parameter being estimated is expected to lie within. The
σ
margin of error for a normally distributed population is given as 𝑍 α
1− 2 𝑛
σ
or 𝑍 α depending on the type of statistical table in use. Thus, the
2 𝑛

43
 STA 201/209/221/229 TUTORIAL NOTE

Confidence Interval estimate of a population parameter is obtained by

the formula
σ
𝑃𝑜𝑖𝑛𝑡 𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒 ± 𝑍 α
1− 2 𝑛
σ
𝑃𝑜𝑖𝑛𝑡 𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒 ± 𝑍 α
2 𝑛

For instance, a 95% C.I estimate implies that, if all possible samples of

)
3
the same size were drawn, then 95% of them would include the true

51
population mean somewhere within the interval around their sample
mean and only 5% of them would not. Generally, we expressed

6
confidence interval as 100(1 − 𝛼)% and for the determination of level of

01
significance (𝛼), the value of Student-t distribution at infinity (∞) is

34
always equal to that of Normal distribution.

80
CONFIDENCE INTERVAL FOR MEAN OF LARGE SAMPLE AND SMALL
SAMPLES
(0
We’ll treat this in two cases which are small and Large Samples
Y

Case 1: When σ is known and n≥30

U

The (1 − α)100% confidence interval for µ is given by

G

σ
𝑋 ± 𝑍α
𝑛
TS

Case 2: When σ is unknown and n<30

The (1 − α)100% confidence interval for µ is given by
A

𝑆
ST

𝑋 ± 𝑡α
2
, (𝑛−1) 𝑛

2
Σ(𝑥−𝑥)
T

where S is the sample standard deviation given by 𝑛−1

A

Where n-1 is the degree of freedom

TH

Example (Case 1): According to a voters register in Lagos state, it was

found that the mean and standard deviation of ages of a random
sample of 40 registered voters in a particular ward were 25 years and 5
years respectively. Construct a 95% confidence interval for the true
mean age of the entire registered voters.

44
 STA 201/209/221/229 TUTORIAL NOTE

Solution: 𝑛 = 40 𝑋 = 25 𝑦𝑒𝑎𝑟𝑠 σ = 5 𝑦𝑒𝑎𝑟𝑠 (1 − α)100% = 95%

Since n > 30 we’ll be using case 1
(1 − α)100% = 95%
(1 − α)100 = 95
(1 − α) = 0. 95
α = 1 − 0. 95
α = 0. 05

)
3
51
σ
Margin of Error = 𝑍 α
2 𝑛

6
5
= 𝑍 0.05

01
2 40
5
= 𝑍 0.05

34
2 40

= 𝑍0.025 × 0. 7905694
= 1. 96 × 0. 7905694
= 1.54952 80
(0
Confidence Interval = 𝑋 ± 𝑀𝑎𝑟𝑔𝑖𝑛 𝑜𝑓 𝐸𝑟𝑟𝑜𝑟
= 25 ± 1. 54952
Y

Interval: 25 − 1. 54952 < µ < 25 + 1. 54952

U

23. 4505 < µ < 26. 5495

G
TS

Example (Case 2): The following represent the scores of 10 randomly

selected students in Statistics methods exams: 59, 40, 38, 67, 78, 45, 56,
A

60, 45, 80. Construct a 95% confidence interval for the average score of
ST

the entire students who registered for the exams.

Solution: 𝑛 = 10 , (1 − α)100% = 95% 𝑖. 𝑒 α = 0. 05
T

Σ𝑋 59+40+38+67+78+45+56+60+45+80 568
𝑋= = = = 56. 8
A

𝑛 10 10
TH

2
Σ(𝑥−𝑥)
𝑆= 𝑛−1

𝑥 2
𝑥 −𝑥 (𝑥 − 𝑥)

59 2.2 4.84

45
 STA 201/209/221/229 TUTORIAL NOTE

40 -16.8 282.24

38 -18.8 353.44

67 10.2 104.04

78 21.2 449.44

45 -11.8

)
139.24

3
56 -0.8 0.64

51
60 3.2 10.24

6
01
48 -8.8 77.44

34
80 23.2 538.24

80
(0 1959.8

1959.8
𝑆= 9
= 14. 8
14.8
Margin of Error = 𝑡 0.05
Y

2
, (10−1) 10
U

= 𝑡0.025, (9) × 4. 68017

G

= 2. 26 × 4. 68017
TS

= 10. 577
Confidence Interval = 𝑋 ± 𝑀𝑎𝑟𝑔𝑖𝑛 𝑜𝑓 𝐸𝑟𝑟𝑜𝑟
A

= 56. 8 ± 10. 577

ST

Interval: 56. 8 − 10. 577 < µ < 56. 8 + 10. 577

46. 223 < µ < 67. 377
T
A

CONFIDENCE INTERVAL FOR PROPORTION

TH

The (1 − α)100% confidence interval for 𝑃 is given by

𝑝(1−𝑝)
𝑃 ± 𝑍α 𝑛
2

𝑥
where 𝑃 = 𝑛
Example: Out of a random sample of 800 people who casted their votes
at Lagos polling post, it was discovered that only 400 people voted for

46
 STA 201/209/221/229 TUTORIAL NOTE

APC. Construct a 95% confidence interval for the actual proportion of

the entire voters who casted their votes for APC at the polling station.
Solution: 𝑥 = 400, 𝑛 = 800, α = 0. 05
𝑥 400
𝑃= 𝑛
= 800
= 0. 5

𝑝(1−𝑝)
Margin of Error =𝑍 α 𝑛
2

)
0.5(1−0.5)
=𝑍 0.05

3
800

51
2

0.5(0.5)
=𝑍0.025

6
800

01
0.25
=𝑍0.025 800

34
=1. 96 × 0. 0177
=0. 0346
Confidence Interval = 𝑃 ± 𝑀𝑎𝑟𝑔𝑖𝑛 𝑜𝑓 𝐸𝑟𝑟𝑜𝑟
= 0. 5 ± 0. 0346 80
(0
Interval: 0. 5 − 0. 0346 < 𝑃 < 0. 5 + 0. 0346
0. 4654 < 𝑃 < 0. 5346
Y
U
G

Want to join me and your peers as we go on a challenge of solving 500+

likely to come out in your forthcoming exams? Text me on WhatsApp at
TS

08034016513 with the code “500 MCQ” and wait for a reply. The earlier, the
better. Take action now!!
A
ST

Note: This is different from the Tutorials

T
A
TH

THIS MATERIAL IS NOT TO BE SOLD FOR

ANY REASON

47
 STA 201/209/221/229 TUTORIAL NOTE

LESSON 7 : HYPOTHESIS TESTING

Hypothesis Testing is the methodology in which a decision maker draws
meaningful conclusions about population characteristics ( parameters ) by
analyzing the difference between value of the sample statistic and the
corresponding hypothesized parameter value.

SOME IMPORTANT DEFINITIONS

)
3
Level of significance: This is the probability of rejecting the null hypothesis

51
when it is true. It’s denoted by α

6
Null Hypothesis (𝐻0): It is the hypothesis being tested and it is usually referred

01
to as the hypothesis of no significance or no change.

34
Alternative Hypothesis (𝐻1): It is the hypothesis of practical interest which is

80
expected to disprove the null hypothesis. It is the hypothesis of change or
(0
hypothesis that shows difference.

One-tailed test: It is also known as Directional hypothesis. A test is one-tailed

Y
U

if the alternative hypothesis states that the population parameter is greater

than or less than the one being tested such that it leads to the following
G

hypothetical statements
TS

𝐻0: µ = µ0
A

Right Tailed Test

ST

𝐻0: µ > µ0
𝐻0: µ = µ0
T

𝐻0: µ < µ0 Left Tailed Test

A
TH

Two-tailed test: It is also known as Non-Directional hypothesis. A test is

two-tailed if the alternative hypothesis states that the population parameter is
different from the one being tested such that it leads to the following
hypothetical statements
𝐻0: µ = µ0
𝐻0: µ ≠ µ0

48
 STA 201/209/221/229 TUTORIAL NOTE

Critical Region: Is that region such that the null hypothesis is rejected if the
test statistic computed from a particular sample takes on a value in the region.
The critical region for one-tail hypothesis occupies only one side of the normal
curve tails while that of two-tail lies on the two tails. This can be illustrated in
the figures below

)
3
6 51
01
34
80
(0
Y
U

Acceptance Region: Is that region such that the null hypothesis is not
G

rejected if the test statistic computed from a particular sample takes on a

value in the region. It is the complement of the critical region.
TS

Type I Error: Error of rejecting null hypothesis when it is true (α) it's also known
A

as a false positive.
ST

Type II Error: Error of failing to reject null hypothesis when (β)it is false it's also
known as a false negative
T

Power of a Test: Probability of rejecting 𝐻0 when it is false (1 − β)

A
TH

Note, (1 − α ) refers to as confidence level, measures the probability level of not

rejecting a true null hypothesis.

Size of a test: Probability of falsely rejecting the null hypothesis. It is the same
as the Type I Error and is also denoted by α
Critical Value: is the value of the sample statistic that separates the regions of
acceptance and rejection.

49
 STA 201/209/221/229 TUTORIAL NOTE

STEPS IN PERFORMING AN HYPOTHESIS TEST

This is often referred to as the 8 step procedures in hypothesis testing
1. From the problem context, Identify the parameter of interest
2. Identify the Null Hypothesis 𝐻0
3. Identify the Alternative Hypothesis 𝐻1
4. Identify the Level of Significance α

)
5. Write out the Test Statistic

3
6. Write your Decision Rule

51
7. Perform all necessary calculations

6
8. State your Conclusions

01
HYPOTHESIS TEST FOR MEAN OF LARGE AND SMALL SAMPLES

34
Case 1 (Large Samples) : When n≥30

80
NB: n is known as the Sample Size and N is known as the population size
(0
The hypothesis test can be one tailed or two tailed and can be stated as
follows
Y

𝐻0: µ = µ0 𝑣𝑠 𝐻1: µ = µ0 𝑜𝑟 𝐻1: µ > µ0 𝑜𝑟 𝐻1: µ < µ0

U

at α level of significance
G

The test statistic is given by

TS

𝑋−µ0 𝑋−µ0
𝑍 = = (σ𝑋 = σ/ 𝑛)
A

σ𝑋 σ/ 𝑛
ST

σ𝑋 is known as standard error of the mean.

T

Case 2 (Small Samples) : When n<30

A
TH

The hypothesis test can be one tailed or two tailed and can be stated as
follows
𝐻0: µ = µ0 𝑣𝑠 𝐻1: µ ≠ µ0 𝑜𝑟 𝐻1: µ > µ0 𝑜𝑟 𝐻1: µ < µ0
at α level of significance
The test statistic is given by

50
 STA 201/209/221/229 TUTORIAL NOTE

𝑋−µ0
𝑡 =
𝑠/ 𝑛

Example 1 : According to a voters register in Lagos state, it was found that the
mean and standard deviation of ages of a random sample of 40 registered
voters in a particular ward were 25 years and 5 years respectively. Test the

)
hypothesis that the average age of the entire registered voters is significantly

3
different from 27 years at 5% level of significance.

51
Solution:

6
Step 1: From the problem context, Identify the parameter of interest

01
Parameter of Interest: Mean ages of registered voters

34
Step 2: Identify the Null Hypothesis 𝐻0
𝐻0: µ = 27 𝑦𝑒𝑎𝑟𝑠

80
Step 3: Identify the Alternative Hypothesis 𝐻1
(0
𝐻1: µ ≠ 27 𝑦𝑒𝑎𝑟𝑠
Step 4: Identify the Level of Significance
Y

α = 0. 05
U

Step 5: Write out the Test Statistic

G

since we are dealing with a large sample i.e n>30 we’ll be using Case 1 stated
above
TS

𝑋−µ0
𝑍𝑐𝑎𝑙𝑐 =
A

σ/ 𝑛
ST

Step 6: Write your Decision Rule

Since it is a two tailed test our decision rule is stated as follows
T

Decision Rule: Reject 𝐻0 if 𝑍𝑐𝑎𝑙𝑐 <− 𝑍α/2 or 𝑍𝑐𝑎𝑙𝑐 > 𝑍α/2

A

Step 7: Perform all necessary calculations

TH

𝑛 = 40 𝑋 = 25 𝑦𝑒𝑎𝑟𝑠 σ = 5 𝑦𝑒𝑎𝑟𝑠 α = 0. 05 µ = 27 𝑦𝑒𝑎𝑟𝑠

𝑋−µ0
𝑍𝑐𝑎𝑙𝑐 =
σ/ 𝑛
25−27
𝑍𝑐𝑎𝑙𝑐 =
5/ 40

51
 STA 201/209/221/229 TUTORIAL NOTE

𝑍𝑐𝑎𝑙𝑐 =− 2. 53
− 𝑍α/2 =− 𝑍0.05/2 = − 𝑍0.025 =− 1. 96
Step 8: State your Conclusions
Conclusion: Since 𝑍𝑐𝑎𝑙𝑐 <− 𝑍α/2 we reject 𝐻0 and therefore conclude that the
mean ages of registered voters may not be 27 years at 5% level of significance.

)
3
Example 2: The following represent the scores of 10 randomly selected

51
students in Statistics methods exams: 59, 40, 38, 67, 78, 45, 56, 60, 45, 80.
Test hypothesis against the claim that average scores of the entire

6
01
students in Statistics methods exams is different from 72 at 5% level of
significance.

34
Solution:
𝑛 = 10 𝑋 = 56. 8 𝑠 = 14. 8 µ = 72 α = 0. 05

80
Step 1: From the problem context, Identify the parameter of interest
Parameter of Interest: Mean Scores of Students in Statistical Methods Exam
(0
Step 2: Identify the Null Hypothesis 𝐻0
Y

𝐻0: µ = 72
U

Step 3: Identify the Alternative Hypothesis 𝐻1

G

𝐻1: µ ≠ 72
TS

Step 4: Identify the Level of Significance

α = 0. 05
A

Step 5: Write out the Test Statistic

ST

since we are dealing with a small sample i.e n<30 we’ll be using Case 2 stated
above
T

𝑋−µ0
A

𝑡𝑐𝑎𝑙𝑐 =
TH

𝑠/ 𝑛
Step 6: Write your Decision Rule
Since it is a two tailed test our decision rule is stated as follows
Decision Rule: Reject 𝐻0 if 𝑡𝑐𝑎𝑙𝑐 <− 𝑡 α or 𝑡𝑐𝑎𝑙𝑐 > 𝑡 α
2
, (𝑛−1) 2
, (𝑛−1)

Step 7: Perform all necessary calculations

𝑛 = 10 𝑋 = 56. 8 𝑠 = 14. 8 µ = 72 α = 0. 05

52
 STA 201/209/221/229 TUTORIAL NOTE

𝑋−µ0
𝑡𝑐𝑎𝑙𝑐 =
𝑠/ 𝑛
56.8−72
𝑡𝑐𝑎𝑙𝑐 =
14.8/ 10
𝑡𝑐𝑎𝑙𝑐 =− 3. 25

)
− 𝑡α =− 𝑡 0.05 =− 𝑡0.025, 9 =− 2. 26
, (𝑛−1) , (10−1)

3
2 2

Step 8: State your Conclusions

51
Conclusion: Since 𝑡𝑐𝑎𝑙𝑐 <− 𝑡 α we reject 𝐻0 and therefore conclude that

6
2
, (𝑛−1)

01
the Mean Scores of Students in Statistical Methods Exam may not be 72 at 5%
level of significance.

34
HYPOTHESIS TEST FOR PROPORTION
The null and Alternative Hypothesis are stated as follows
𝐻0: 𝑃 = 𝑃0 𝑣𝑠 𝐻1: 𝑃 ≠ 𝑃0 𝑜𝑟 𝐻1: 𝑃 > 𝑃0 𝑜𝑟 80 𝐻1: 𝑃 < 𝑃0
(0
at α level of significance
The test statistic is given by
Y
U

𝑃−𝑃0 𝑃−𝑃0
𝑍 = =
G

σ𝑃
𝑃(1−𝑃)/𝑛
TS

Example: Out of a random sample of 800 people who casted their votes at
Lagos polling post, it was discovered that only 400 people voted for APC, test
A

hypothesis against the claim that the proportion of registered voters who
ST

cast their votes for APC is significantly different from 0.30.

Solution:
T
A

𝑛 = 800 𝑥 = 400 α = 0. 05 𝑃0 = 0. 30 𝑃 = 𝑥/𝑛 = 400/800 = 0. 5

TH

Step 1: From the problem context, Identify the parameter of interest

Parameter of Interest: Proportion of voters who voted for APC
Step 2: Identify the Null Hypothesis 𝐻0
𝐻0: 𝑃 = 0. 3
Step 3: Identify the Alternative Hypothesis 𝐻1
𝐻1: 𝑃≠ 0. 3

53
 STA 201/209/221/229 TUTORIAL NOTE

Step 4: Identify the Level of Significance

α = 0. 05
Step 5: Write out the Test Statistic
𝑃−𝑃0
𝑍𝑐𝑎𝑙𝑐 =
𝑃(1−𝑃)/𝑛
Step 6: Write your Decision Rule

)
Since it is a two tailed test our decision rule is stated as follows

3
Decision Rule: Reject 𝐻0 if 𝑍𝑐𝑎𝑙𝑐 <− 𝑍α/2 or 𝑍𝑐𝑎𝑙𝑐 > 𝑍α/2

51
Step 7: Perform all necessary calculations

6
01
𝑛 = 800 α = 0. 05 𝑃0 = 0. 30 𝑃 = 0. 5

34
0.5−0.3
𝑍𝑐𝑎𝑙𝑐 =
80
0.5(1−0.5)/800
0.2
𝑍𝑐𝑎𝑙𝑐 =
(0
0.5(0.5)/800
0.2
𝑍𝑐𝑎𝑙𝑐 =
Y

0.25/800
U

0.2
𝑍𝑐𝑎𝑙𝑐 =
G

0.0176777
TS

𝑍𝑐𝑎𝑙𝑐 = 11. 31
A

𝑍α/2 = 𝑍0.05/2 = 𝑍0.025 = 1. 96

ST

Conclusion: Since 𝑍𝑐𝑎𝑙𝑐 > 𝑍α/2 we reject 𝐻0 and therefore conclude that the
Proportion of voters who voted for APC may not be 0.3 at 5% level of
T

significance.
A
TH

THIS MATERIAL IS NOT TO BE SOLD FOR

ANY REASON

54