Probability and Statistics

UNIT 1

Ebenezer Asabre

Ho Technical University, Department of Mathematics and Statistics

Probability and Statistics for Engineers

June, 2023.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 1 / 53
 Course Outline

Here are the areas we will be looking at:

1 Introduction to Statistics: The Role/Application of Statistics in
Engineering.
2 Introduction to Probability theory.
3 Random Variables: Discrete & Continuous.
4 Special Discrete Distributions.
5 Special Continuous Distributions.
6 Joint Probability Distributions.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 2 / 53
 UNIT 1: INTRODUCTION TO STATISTICS

Definition
Statistics is the science of conducting studies to collect, organize,
summarize, analyze,and draw conclusions from data.

The Role of Statistics in Engineering

Statistics is of great importance to engineers, providing a powerful tool for
understanding the data collected from experiments and practical activities.
Engineers rely heavily upon statistics in several ways, from using statistical
models for problem-solving to helping make decisions based on probability.
Some practical applications of statistics to engineering is given below.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 3 / 53
 INTRODUCTION TO STATISTICS

Branches of Statistics
Depending on how data are used, statistics can be put into two areas:
1 Descriptive Statistics
2 Inferential Statistics

Descriptive Statistics
Descriptive statistics consists of the collection, organization,
summarization and presentation of data. It presents data in tabular and
graphical forms as well as using nummerical measures such as percentages,
mean, standard deviation, etc in an informative way.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 4 / 53
 INTRODUCTION TO STATISTICS

Inferential Statistics
Inferential or inductive statistics is concerned with drawing conclusions
regarding all information of interest on the basis of samples taken.
For example the Department of Electrical/Electronic Engineering may be
contracted by the ECG to produce energy-saving light bulbs. In assessing
the quality of the bulbs, sample of the bulbs after production could be sent
to individuals to use and based on their responds, the whole production
can be commercialised.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 5 / 53
 INTRODUCTION TO STATISTICS
Application of Statistics in Engineering
1. Design of experiments (DOE): This is a statistical method to
optimize designs, processes, and products. The process involves using
various experiments to test different parameters and measure the
results. Statistical models are used to analyze the data collected
during these experiments, providing insights into which combinations
of factor levels or variables lead to optimal product performance. The
use of DOE method can significantly optimize production processes,
improve the quality of products, reduce costs, and enhance safety and
reliability.
2. Reliability: This measures how well an engineered system can perform
its intended functions under specified conditions for a specific period.
The analysis involves collecting data on operating conditions and
failure patterns and studying past failures and case studies. Statistical
methods such as hazard rate analysis can be used improve reliability
and increase productivity.
Ebenezer Asabre (Ho Technical University, Department
Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 6 / 53
 INTRODUCTION TO STATISTICS
Application of Statistics in Engineering
3. Quality Control: In engineering, quality control is crucial and
fundamental to ensure that all products and services meet the
required standards in terms of quality and safety. This process
involves collecting data, analyzing it, and making informed decisions
on quality characteristics, such as tolerance limits, process variability,
and process capability. Statistical process control methods are most
commonly used for quality control and involve monitoring and
controlling variables during manufacturing to ensure that products
meet required standards.
4. Risk analysis: This is the process of assessing the likelihood and
severity of potential risks to make informed decisions. Through
probability theory, engineers can to identify, analyze, and quantify
potential risks. It enables engineers to understand the risk associated
with particular activities or situations, allowing them to take more
effective steps to lessen them.
Ebenezer Asabre (Ho Technical University, Department
Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 7 / 53
 INTRODUCTION TO STATISTICS

Application of Statistics in Engineering

5. Predictive modeling: This is an application of statistics to forecast
future outcomes from existing data. It helps engineers make
better-informed decisions and plan future events more effectively.
Predictive models can forecast market trends, anticipate customer
behavior, improve forecasting accuracy and efficiency, and optimize
processes such as inventory management. The accuracy of predictions
depends on both the quality of data and the model used.
6. Process improvement: This involves the identification and elimination
of non-value-added activities in engineering processes. Engineers use
statistical methods to identify areas of improvement and optimize
process efficiency.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 8 / 53
 INTRODUCTION TO STATISTICS

Data & Variable

A variable is any quantity that changes from individual to individual or
object to object.

Data is/are the values (measurements or observations) which variables can

assume. In other words they are raw facts and figures from which
conclusions can be drawn.
A collection of data values forms a data set. Each value in the data set is
called a data value or a datum.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for Engineers)
2023. 9 / 53
 INTRODUCTION TO STATISTICS

Types of Data/Variables
Data can be put in two types depending on its nature:
Qualitative data/variable
Quantitative data/variable

Qualitative Variable
These are quantities that can be placed into distinct categories, according
to some characteristics or attributes. Example are gender, grade of student
in Probability & Statistics course, religious/political affiliation, etc.

Quantitative Variable
This represents data that are numerical and can be ordered or ranked.
Examples are weight, height, temperature, etc. Quantitative variables can
be further be distinguished into discrete and continuous types.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)10 / 53
 INTRODUCTION TO STATISTICS

Types of Quantitative Variable

Discrete variables assume values that can be counted.
Examples of discrete variables are the number of children in a family, the
number of students in a Department, the number of power meters in
hostel, etc.

Continuous variables can assume an infinite number of values between any

two specific values. They are obtained by measuring. They often include
fractions and decimals.
Example: Weight of a metal bar can be described as a continuous
quantitative variable because between 5kg & 6kg are infinite number of
measurements.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)11 / 53
 INTRODUCTION TO STATISTICS

Important characteristics of data

1. Center: the average value that indicates where the middle of the
data set is located.
2. Variation: a measure of how data points vary among each other.
3. Distribution: the nature or shape of the data set (i.e bell shaped,
uniform, or skewed)
4. Outlier: sample values that lies far apart from the majority of other
sample points. They are usually far below or above the center.
5. Time: changing characteristics of data overtime.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)12 / 53
 INTRODUCTION TO STATISTICS

Definition of Statistical Terms

1. Population: A population consist of the entire subjects under study.
2. Sample:A sample is a group of subjects selected from a population.
3. Parameter: A measurement describing some characteristics of a
population. Examples are population mean and population variance.
4. Statistic: A measurement describing some characteristics of a
sample. Examples are sample mean and sample variance.
5. Elementary Units: Each individual item in a population.
6. Accuracy: Refers to the closeness of measurement to the ”actual” or
”real” value of the physical quantity.
7. Constant: A characteristic whose observation do not change from
response to response.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)13 / 53
 Exercise 1
Categorize the following items into qualitative or quantitative variables.
1 The thickness (in cm) of a rod.
2 Sex of a student.
3 Use of electric heater (Yes/No).
4 Colour of a light bulb.
5 Average monthly electricity bill.

Exercise 2
Identify each variable as being either discrete or continuous variable giving
a reason for your answer.
1 The number of blue balls selected randomly from a box.
2 The average speed of a moving vehicle.
3 Weight loss after a series of aerobic exercises.
4 The number shown after a die is tossed once.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)14 / 53
 INTRODUCTION TO STATISTICS

Scale of Measurement
Scale of measurement refers to ways in which variables are defined and
categorized. Each scale of measurement has certain properties which in
turn determines the appropriateness for use of certain statistical analyses.
The four scales of measurement are nominal, ordinal, interval and ratio.

Nominal level of measurement

Classifies data into mutually exclusive (non-overlapping) categories in
which no order or ranking can be imposed on the data.
Example: gender (male or female), political party affiliation (NPP, NDC,
CPP, etc), religious affiliation (Christianity, African Tradition, Islam, etc.),
marital status (single, married, divorced, widowed, separated, etc).

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)15 / 53
 INTRODUCTION TO STATISTICS

Ordinal level of measurement

Ordinal level of measurement classifies data into categories that can be
ranked; however, precise differences between the ranks do not exist.
Example 1: From student evaluations, guest speakers might be ranked as
superior, average, or poor.
Example 2: Grades of students in an examination (A+, A, B+, B, etc).

Interval level of measurement

Interval level of measurement ranks data and precise differences between
units of measure do exist; however, there is no meaningful zero.
For example, many standardized psychological tests yield values measured
on an interval scale. IQ is an example of such a variable. There is a
meaningful difference of 1 point between an IQ of 109 and an IQ of 110.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)16 / 53
 INTRODUCTION TO STATISTICS

Ratio level of measurement

This level of measurement possesses all the characteristics of interval
measurement, and there exists a true zero. In addition, true ratios exist
when the same variable is measured on two different members of the
population.
Examples of ratio scales are those used to measure height, weight, area,
and number of phone calls received. Ratio scales have differences between
units (1 inch, 1 pound, etc.) and there exist a true zero.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)17 / 53
 INTRODUCTION TO STATISTICS

Sampling Techniques
Random Sampling: This is the sampling technique where all members
(belonging to the population) have equal chance of being selected (into
the sample).

Systematic Sampling
Researchers obtain systematic samples by numbering each subject of the
population and then selecting every kth subject. For example, suppose
there are 100 subjects in the population and a sample of 20 subjects are
needed. Our k = 100
20 = 5.
This means after selecting our first sample randomly (from 1 and 5), every
5th subject would be selected. If subject 3 is the first subject selected;
then the sample would consist of the subjects whose numbers are 3rd, 8th,
13th, etc.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)18 / 53
 Stratified Sampling
Stratified samples are obtained by dividing the population into groups
(called strata) according to some characteristic that is important to the
study, then sampling from each group. Samples within the strata should
be randomly selected. For example, the government of Ghana may want to
conduct a survey on citizens’ view on a policy to be implemented, the
entire population can be divided into groups (16 regions) and randomly
select districts within each region to conduct the survey.

Cluster Sampling
Here the population is divided into groups called clusters by some means
such as geographic area. Then the researcher randomly selects some of
these clusters and uses all members of the selected clusters as the subjects
of the samples.

Other sampling techniques, are convenience sampling, sequential sampling,

double sampling, and multi-stage sampling.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)19 / 53
 INTRODUCTION TO STATISTICS

Frequency Distribution
The frequency distribution table indicates the occurrence of the
observations or values in data obtained.

Example 1
The data below are the number of light bulbs per household sampled from
a community some time ago.
0 1 4 4 3 1 2 3 1 2
2 4 3 0 2 5 0 2 2 1
3 2 1 1 3 2 3 4 5 2
1 0 5 4 2 0 3 5 1 2
4 3 0 2 2 1 1 2 2 4

Construct a frequency distribution table for the above data showing

relative and cumulative frequencies.
Ebenezer Asabre (Ho Technical University, Department
Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)20 / 53
 Table 1:
No. of bulbs (x) No. of Fam (f) Relative Freq. Cum. Freq.
0 6 0.12 6
1 10 0.20 16
2 15 0.30 31
3 8 0.16 39
4 7 0.14 46
5 4 0.08 50
Total 50 1.00

When the range of the data is large, the data must be grouped into classes
that are more than one unit in width, in what is called a grouped
frequency distribution.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)21 / 53
 INTRODUCTION TO STATISTICS

Definitions:
1 Class Intervals: These are continuous intervals selected in such as
that they are mutually exclusive.
2 Class Boundaries: These are numbers that are the upper and lower
class if there are no gaps between the consecutive classes.
3 Class Width/Class size: The difference between the lower and the
upper class boundaries
4 Class Mark/Class Midpoint: This is the midpoint of the class
interval.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)22 / 53
 INTRODUCTION TO STATISTICS

Example 2 : Grouped Data

The table below shows the weight (in ounces) of malignant tumour
removed from the abdomen of 65 patients.
18 63 42 27 30 57 28 32 12 28
23 24 25 44 51 36 12 45 25 28
28 42 36 51 74 25 43 65 12 32
38 42 27 31 50 38 21 16 24 59
23 22 43 27 49 38 23 19 49 30
49 12 22 31 49 47 43 80 63 35
55 41 54 11 38

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)23 / 53
 Solution
Since the data is quite large and no classes is given, we use Sturges
Approximation Rule to obtain number of classes to be used:
=⇒ K = 1 + 3.322 log10 n = 1 + 3.322 log10 65 = 7.0225 ≈ 7
We also calculate the Range (R) and Class width (C) as
R = Max − Min = 80 − 11 = 69
Range 69
C= K = 7 = 9.857 ≈ 10

Class boundaries = Lb − Ub
Lb = lower limit − 0.5
Ub = Upper limit + 0.5
=⇒ For the class limits 11 − 20 the boundaries are 10.5 − 20.5

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)24 / 53
 INTRODUCTION TO STATISTICS

Class interval Class boundaries Frequency

11 - 20 10.5 - 20.5 6
21 - 30 20.5 - 30.5 20
31 - 40 30.5 - 40.5 11
Table 2: 41 - 50 40.5 - 50.5 16
51 - 60 50.5 - 60.5 6
61 - 70 60.5 - 70.5 4
71 - 80 70.5 - 80.5 2
Total: 65

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)25 / 53
 INTRODUCTION TO STATISTICS

Exercise 3
Twenty-five army inductees at the 66 Artillery Military Barracks - Ho, were
given a blood test to determine their blood type.
The data set obtained was:
A B B AB O
O O B AB B
B B O A O
A O O O B
B AB O A B

Construct a frequency distribution for the data.

NB: Since the data is categorical, discrete classes can be used and
percentage frequency of each blood type calculated.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)26 / 53
 MEASURES OF CENTRAL TENDENCY

INTRODUCTION
An average is a value which is typical and a representation of a set of
data. There are different approaches of measuring central tendency.
Example of such approaches are arithmetic mean, geometric mean,
harmonic mean and quadratic mean.

Arithmetic Mean
The arithmetic mean is the sum of the values, divided by the total number
of values. The symbol represents the sample mean.
It is denoted by:
X̄ = X1 +X2 +Xn3 +...+Xn = n1 ni=1 Xi
P

For population, the Greek letter µ is used instead of X̄

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)27 / 53
 MEASURES OF CENTRAL TENDENCY

Example 3
The data below represents the mid semester exam (marked out of 20)
scores of some 10 students in a Prob. & Stats class.
15, 13, 14, 18, 18, 14, 16, 19, 11, 17.
Find the mean score.

Mean of Frequency Distribution

Pn
i i fX
For an ungrouped data, the mean is given by: X̄ = Pi=1 n
i=1 fi
where Xi are the individual observations and fi are their corresponding
frequencies for i = 1, 2, 3, ..., n
For grouped data, class midpoints of intervals are used in calculating
the mean. Thus; P
fX
X̄ = P fmpt

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)28 / 53
 MEASURES OF CENTRAL TENDENCY

Example 4
The length of 40 electric cables are given to the nearest centimetres as:
Length(cm) 4-8 9-13 14-18 19-23 24-28 29-33
Frequency 2 4 7 14 8 5

Class interval Midpoint (Xmpt ) frequency fXmpt

4-8 6 2 12
9-13 11 4 44
14-18 16 7 112
Table 3:
19-23 21 14 294
24-28 26 8 208
29-33 31 5 155
P P
Total f = 40 fXmpt = 825
P
fXmpt 825
X̄ = P
f = 40 = 20.625cm

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)29 / 53
 MEASURES OF CENTRAL TENDENCY

Assumed Mean
This is also called working or guessed mean. To use this method we have
some assumed mean to work with. Normally it is recommended when the
data large.

Asummed mean for ungrouped data

P
di
X¯A = A + n

Asummed mean for grouped data

P
X¯A = A + Pfi di
fi
P
Where A is the assumed mean, di = xi − A and f = n the number of
observations.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)30 / 53
 MEASURES OF CENTRAL TENDENCY

Example 5
Given an assumed mean of 60, find the mean of the data below.
60, 65, 54, 56, 70, 65.

Solution
Since di = xi − A
d1 = x1 − A = 60 − 60 = 0
d2 = x2 − A = 65 − 60 = 5
P
So di = 0, 5, −6, −4, 10, 5 ∴ di = 10
P
di
=⇒ X¯A = A + n = X̄ = 60 + 10
6 = 61.667

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)31 / 53
 MEASURES OF CENTRAL TENDENCY

Weighted Mean
In some cases, data points carries different weights and we have to
consider their weights in calculating the overall mean.

Given weights wi of data points xi , i = 1, 2, ..., n the weighted mean is

given given by: P
wi xi
X̄w = P
wi
Areas of application of the weighted mean is calculating CGPA or CWA of
students. This is because courses offered normally have different weights
(credit hours).

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)32 / 53
 Example 6
Calculate the GPA of a student who had 80% in Communication skills,
90% in Basic Statistics, 75% in Introductory ICT and 73% in Basic
Chinese Language.
Assume the first two courses are 3 credits each and the last two are 2
credit hours each. Use the grading system:
80 − 100 =⇒ A+, 75 − 79 =⇒ A, 70 − 74 =⇒ B+, etc
With grade points: A+ = 5.0, A = 4.5, B+ = 4.0, etc

Solution
P
wi xi (3 × 5) + (3 × 4.5) + (2 × 4.5) + (2 × 4)
GPA = X̄w = P =
wi 3+3+2+2

45.5
= = 4.55
10

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)33 / 53
 Exercise 4
Suppose we need a mean of 5 test scores (58, 79, 84, 56, 81), but the test
score counts 30% ,40%, 50%, 40%, 30% for first, second, third, forth and
fifth respectively of the final grade.
Calculate the weighted mean score.

Harmonic Mean
The harmonic mean of n observation X1 , X2 , ..., Xn is given by:
For an ungrouped data:
n
X¯h = P 1
Xi

For a grouped data:

P
fi
X¯h = P f
i
Xi

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)34 / 53
 MEASURES OF CENTRAL TENDENCY

Applications of Harmonic Mean

1 For averaging speed
2 For averaging cost

Example 7
A student rides a motor bike from campus to Ho Sports Stadium at an
average speed of 60km/h and returns to campus along the same route at
an average speed of 40km/h. Find the average speed of the entire journey.

Solution
n 2
X̄h = P 1 = 1 1
= 48km/h
Xi 60 + 40

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)35 / 53
 MEASURES OF CENTRAL TENDENCY

Exercise 5
A lecturer drives the first 210km at an average speed of 90km/h and the
remaining 146km at an average speed of 110km/h from Kumasi to Ho.
What is the average speed of the entire trip.

Geometric Mean
The k th root of the product of k observations is the geometric mean. The
geometric mean is applicable when finding means of percentages, ratios,
indexes or growth rates.
For ungrouped data:
p
X¯g = k X1 × X2 × X3 × · · · × Xk
q
For grouped data: X¯g = k X1f1 × X2f2 × X3 × · · · × Xkfk

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)36 / 53
 MEASURES OF CENTRAL TENDENCY

Example 8
The population of a community increased by 20% from the year 2019 to
2020 and 36% from 2020 to 2021. Find the average rate of increase
during the entire period (i.e 2019-2021).
This is a geometric mean with k = 2.
√
=⇒ X¯g = 1.20 + 1.36 = 1.277
This means that on the average the population increased by 27.7%.
In other words, the population increased 1.277 times during the period
under consideration (i.e 2019-2021).

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)37 / 53
 Median
The median of a data set is the measure of center, that is the middle value
when the data is arranged in increasing or decreasing order of magnitude.

For ungrouped data:

(n+1)th
Median = 2 observation

For grouped data:

( N2 − Cfm−1 )
Median = Lb + ×h
fm
Where,
Lb = lower class boundary of the median class
N = no. of observations in the data, fm = freq. of the median class
Cfm−1 = cumm freq. of the class immediately before the median class,
h = class width of the median class

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)38 / 53
 MEASURES OF CENTRAL TENDENCY
Example 9
The number of rooms in seven selected hotels in Ghana are 713, 300, 618,
595, 311, 401, and 292. Find the median.

Example 10
The weight of 40 metal bars in kg is presented in the table below. Find
the median.
Wieght (kg) Frequency Cumulative Frequency
118-126 3 3
127-135 5 8
136-144 9 17
145-153 12 29
154-162 5 34
163-171 4 38
172-180 2 40
Ebenezer Asabre (Ho Technical University, Department
Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)39 / 53
 Solution
We have the following parameters
40
N = 40 i.e 2 = 20

We now look for the position where 20 is accumulated in the cumulative

frequency column. (i.e 29).

This means that the median class is 145-153 and hence,

Lb = 144.5, fm = 12, Cfm−1 = 17 and h = 9

( N2 −Cfm−1 ) (20−17)
=⇒ Median = Lb + fm × h = 144.5 + 12 × 9 = 146.8cm

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)40 / 53
 MEASURES OF CENTRAL TENDENCY

MODE
The mode is the observation with the highest frequency.
Data set that has only one value that occurs with the greatest frequency is
said to be unimodal. If a data set has two values that occur with the
same greatest frequency, both values are considered to be the mode and
the data set is said to be bimodal. If a data set has more than two values
that occur with the same greatest frequency, each value is used as the
mode, and the data set is said to be multimodal.

Findind the mode of grouped data

The mode of grouped data set is given by:
d1
Mode = Lb + ( )×h
d1 + d2

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)41 / 53
 MEASURES OF CENTRAL TENDENCY

Where,
Lb = lower class boundary of the modal class
d1 = the difference between the frequency of the modal class and the one
before it.
d2 = the difference between the frequency of the modal class and the one
after it.
h = class with of the modal class.
Exercise 5
Find the mode of the following data:
Class interval 20-22 23-25 26-28 29-31 32-34
Frequency 3 6 12 9 2

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)42 / 53
 MEASURES OF CENTRAL TENDENCY

MIDRANGE
The midrange (MR) of a data set is average of the of the lowest and
highest values in the data set.
lowest value + highest value
MR =
2

Exercise 6
The data below shows the points per game of Michael Jordan in the
1996/1997 NBA regular season (in 10 selected games). Calculate the
midrange of the data.
37, 39, 33, 45, 41, 35, 28, 38, 43, 54.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)43 / 53
 QUARTILES
These are descriptive measures that split the ordered data into four
quarters. That is the lower (first) quartile (Q1 ) which divides data into
1/4, (i.e 25% of values), second quartile (median), Q2 (i.e 50% of values)
which divides data into 1/2 and the third (Upper) quartile Q3 which
divides data into 3/4.(i.e 75% of values)

For ungrouped data:

n+1 2(n + 1) 3(n + 1)

Q1 = th Q2 = th Q3 = th
4 4 4

For grouped data:

qN
4 − Cfq−1
Qq = Lq + ( )×h
fq

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)44 / 53
 Where,
q = 1, 2, 3, so Qq = q th quartile
Lq = lower boundary of the class interval containing the q th quartile
N = total frequency
Cfq−1 = cumulative frequency for the class intervals before the class
containing the q th quartile
fq = frequency of the class interval containing the q th quartile
h = width of the class containing the q th quartile

Example 11
Find the lower and second quartiles for the following data.
Class interval Frequency cumulative frequency
20-22 3 3
23-25 6 9
26-28 12 21
29-31 9 30
32-34 2 32

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)45 / 53
 Solution
To find Q1 , first find the first quartile class
=⇒ 1N 32
4 = 4 =8
This means that we we want to find the point Q1 that leaves 8
observations in the data below it.
Hence, from the table the quartile class is 23-25.
N
−Cfq−1
=⇒ Q1 = L1 + 4
f1 ×h
(8−3)
= 22.5 + 6 × 3 = 25

Inter Quartile Range

By definition,
IQR = Q3 − Q1
NB: Calculate the inter quartile range for Example 11.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)46 / 53
 MEASURE OF VARIATION

Variation or dispersion is the degree to which numerical data tend to

spread about an average value.

Range
By definition,
Range = Maximum value - Minimum value

Mean Deviation
The mean deviation of a set of n observations X1 , X2 , ..., Xn with mean X̄
is given by: Pn
|Xi − X̄ |
Mean deviation = i=1
n

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)47 / 53
 MEASURE OF VARIATION
Example 12
Find the mean deviation of the observations: 12, 10, 8, 12, 11, 10, 9, 11, 7
P
Solution: X̄ = nX = 90 9 = 10
P
|Xi − X̄ | = |12 − 10| + |10 − 10| + |8 − 10| + |12 − 10| + |11 − 10| +
|9 − 10| + |11 − 10| + |7 − 10| = 12
12
∴ Mean deviation = 9 = 1.333

VARIANCE & STANDARD DEVIATION

The variance is the average of the squares of the distance each value is
from the mean.
Standard deviation on the other hand is the positive root of variance. For
a sample, Var and Std is given by:
Pn 2
qP
i=1 (Xi −X̄ ) (Xi −X̄ )2
s2 = n−1 and s= n−1

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)48 / 53
 For population, Var and Std is given by:
PN 2
qP
i=1 (Xi −µ) (X −µ)2
σ2 = N and σ= N

COEFFICIENT OF VARIATION
The coefficient of variation (CV) for a sample is given by:
s
CV = × 100%
x̄
For population:
σ
CV = × 100%
µ
The coefficient of variation enables us to compare the variation of values
taken from different populations.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)49 / 53
 Example 13
The table below shows the sample height and weight data for 40 NBA
players, find the coefficient of variation.
Mean(x̄) Standard deviation (s)
Height 198.6 cm 9.8 cm
Weight 216 lbs 60.45 lbs

Solution
9.8
Height: CV = 198.6 × 100% = 4.93%
Weight: CV = 60.45
216 × 100% = 27.99%

From the above we can clearly see that there is less variation in height
than weight. This makes some kind of sense because generally weights
among men vary more than height.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)50 / 53
 MEASURE OF VARIATION

Exercise 7
1 Find the variance and standard deviation of the data given below:
Class 11-20 21-30 31-40 41-50 51-60 61-70
f 3 7 17 12 19 5
2 The mean weight of 150 students is 60kg. The mean weight of boys
is 70kg with standard deviation of 10kg.
For the girls, the mean weight is 55kg and standard deviation of 15kg.
Find the number of boys.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)51 / 53
 Exercise 8
In order to choose two measuring instruments X and Y, each was used to
measure the diameter of 20 peswass coin. Instrument X gave the following
measurements in cm.
3.35 3.54 3.35 3.52 3.54 3.57 3.54 3.56 3.54 3.53 3.55
3.55 3.52 3.53 3.54 3.53 3.55 3.56 3.55 3.55

i. Using an assumed mean of 3.54cm, calculate the mean, standard

deviation and Coefficient of Variation of the measurements.
ii. If instrument Y gave the same mean as X but its standard deviation
was 0.0010754 cm which of the two instruments is better and why?
iii. what will be the mean and standard deviation if the above
measurements is decreased by 0.25 cm?

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)52 / 53
 End of Unit 1: Introduction to Statistics
Lecturer’s Details:
Name: Ebenezer Asabre
: +233 547 916717
Email: [email protected]
Kindly send all suggestions via mail.

Ebenezer Asabre (Ho Technical University, Department

Probability
of Mathematics
and Statistics
and Statistics
UNIT 1 Probability and Statistics
June,
for2023.
Engineers)53 / 53