Department of Statistics

STA1506
Basic Statistical Computing

Module Guide
 ii

“The torch is a powerful instrument in darkness – where research starts.

An effective torch tells the investigator where openings exists. Good readers and book
finders discover openings in the field of knowledge search. Scholarship and reading
belong together. A scholar familiar with his/her research is a well-read individual”

CONTENTS
ORIENTATION iii

STUDY UNIT 1 Data management techniques

1.1 Introduction 1
1.2 Objectives of the unit 2
1.3 What is statistics? 2
1.4 Main branches of statistics 3
1.5 Nature of data 5
1.6 Population and sample 16
1.7 Data and information 22
1.8 Data management 24
Exercise 1.1 33
1.9 Learning outcomes 35
References 37

STUDY UNIT 2 Basics of statistical softwares

2.1 Introduction 38
2.2 Objectives of the unit 39
2.3 What is data analysis? 39
2.4 What is Microsoft Excel? 40
2.5 Adding the Analysis ToolPak function 47
2.6 Entering data in Microsoft Excel 47
2.7 Tables and graphical presentations in Excel 49
2.8 Basic calculations in Excel 80
2.9 Statistics analysis in Microsoft Excel using Analysis ToolPak 97
2.10 Use of a software 102
Exercise 2.1 104
2.11 Learning outcomes 107

STUDY UNIT 3 Report writing

3.1 Introduction 108
3.2 Objectives of the unit 109
3.3 Report writing 109
 iii STA1506/1

3.4 Criteria of producing a professional report 110

3.5 Interpretation of results 111
3.6 Managerial reports 148
3.7 Conclusion 107
3.8 Learning outcomes 107
Exercise 3.1 160
References 163
 iv

ORIENTATION
Introduction
Welcome to STA1506! The three modules STA1505, STA1506 and STA1507 form the three modules
in statistics for the Higher Certificate in Mathematics and Statistics. The name Basic Statistical
Computing was chosen because this module introduces you to basic data analysis using computers.
This means that you must have access to a suitable computer for a component of practical work.
This module is an online module which means there will be no printed material to be issued to
you. (Please read carefully through the section "Role of computers in research" following below.)

The purpose of this module is to give learners entry level practical and professional skills at using
a statistical computer program to do data management and data exploration and at writing reports.
The qualifying student is familiar with data cleaning and coding, data entry and capturing, as well as
exploring data and producing reports by summarising the key points of the data.

The contents of this module are at the entry level of data management and data exploration, for the
future professional statistician in the academic, government or industry sectors. This module forms
part of the Higher Certificate in Mathematics and Statistics at the College of Science, Engineering
and Technology (CSET). These competencies contribute to the development of the field of Statistics
in Southern Africa, and globally. The student is required to have access to a suitable computer with
appropriate software and Internet connectivity

Learning outcomes
At the end of each study unit we will list the learning outcomes for that unit but there are also very
specific overall outcomes for this module which we list below. Throughout your study of this module
you must come back to this page, sit back and reflect upon these outcomes, think them through,
digest them and feel confident in the end that you have mastered them.

Managing data.
Performing basic statistical analysis.
Exploring data.
Producing reports by summarising data in a given study.
 v STA1506/1

The prescribed textbook(s)

There is no prescribed book for the module. This study guide contains all the study notes, and in
addition we may refer you on the web site to additional online information.

The module guide

The purpose of this guide is to give you direction on issues of data management, performing basic
statistical analysis in Microsoft Excel, exploration of data and ability to summarise your work in the
form of a report.

Study units and workload

We realise that you might feel overwhelmed by the idea of analysing data and writing a report. Do
not be afraid. We will be with you a very step of the way. In this case we are just giving you direction
to start on your module. However you need to be able to research on your own and use online
materials to assist you in this course. The module introduces you to the skills as a researcher and
we hope that by the end of the semester you will have basic research skills.

Role of computers in research

The emphasis in this module is well beyond the arithmetics of calculating statistics and the focus is
on your ability to be able to perform basic statistical analysis using a software and use the information
obtained to write a report. This is achieved by you understanding how to use a software to perform
basic calculations in statistics. Statistical software or packages will give you the feeling that you are
really practising statistics.

It is a good idea that you initially go through the laborious manual computations to enhance your
understanding of the principles and mathematics involved in Statistics (STA1505). However, you
must be able to manage the Excel computations because using computers reflect the real world
outside. The additional advantage of using a computer is that you can do calculations for larger and
more realistic data sets.
It is impossible and impractical to do assessment of computer skills on computers in the examination
but it does not preclude us from providing you with printed output which you have to interpret.
We will give you definite instructions on where and how to use a computer for your calculations in
assignments. You must be able to use both a computer program and a statistical calculator as tools
for your calculations. However, the emphasis in this module will always be on the interpretation and
how to articulate the results.

Access to a suitable computer

You need access to a computer with Microsoft Excel 2016. Note that as a UNISA student, you have
free access to Microsoft 365 software which includes, among other programmes, both Microsoft
Word and Excel.
 vi

Something about the author(s)

This module guide was compiled by Ms Suwisa Muchengetwa and Prof Eeva Rapoo.
As this is an applied statistics course, it needs continuous improvement since we are living in
a dynamic world. A graduate of statistics needs to know about analysing data using statistical
packages. It is a dream the authors shared to equip modern students interested in the world around
them with the know-how to use a statistical package and be able to write a report. We hope you will
enjoy the module!
 1 STA1506/1

STUDY UNIT 1
Data management techniques
1.1 Introduction
As we embark on this journey of our studies, some of you have asked yourself the question, "why
do you have to study statistics?" Statistics is the subject that is a part of almost every programme
offered by universities, colleges and various private institutions. It also forms part of your everyday
life. When you are enjoying a game of football in front of your TV seeing Kaizer Chiefs playing with
Orlando Pirates or Liverpool playing with Manchester City or Real Madrid playing with Barcelona
or Manchester United playing with Juventus among many, then you will appreciate the beauty of
statistics. Usually at half time, the commentators will be showing the statistics of how the game
has been played so far. That’s the beauty of statistics. Even in the church, the Pastors will try to
give you statistics of how the church is progressing in terms of its activities. Then how can we go
living without such a subject. In the world in 2020 as the coronavirus was attacking the human
species every concerned human being was looking at the statistics asking themselves the question
"Is the pandemic coming to an end?" Instead of having a steep curve of deaths rising up, everyone
was anxiously waiting to see the graph levelling and then going down approaching zero. The virus
spiralled out of control due to the fact that medical practitioners did not have enough ammunition of
facts about the virus. The world started to release facts and figures about the virus to prevent the
human race from coming into extinction. Statistics is essential in giving us the facts of life so that we
are able to plan and make decisions.

This first study unit is designed to provide you with some background knowledge on how you manage
your data processing and are able to turn data into meaningful information. Statistics help planners,
decision makers and institutions to make decisions in the face of uncertainty. In order to make good
decisions one needs good quality information that must be timely presented, accurate, relevant,
adequate and readily available. Statistics deals with ways of collecting informative data, interpreting
these data and drawing conclusions about a phenomenon under study. The scope of statistics
naturally extends to all processes of acquiring knowledge that involve fact finding through collection
and examination of data.

The purpose of this unit on data is to discuss what statistics is, the types of data and their
measurement scales and the process of dealing with data until it is turned into information.
 2

1.2 Objectives of the unit

By the end of this unit, you should be able to:

understand the meaning of the word “Statistics”.

distinguish between the two branches of statistics.

define and distinguish between qualitative data and quantitative data.

define and describe the types of measurement scales.

define and distinguish between a population and a sample and the terminology associated with
them.

define and distinguish between data and information.

outline the data management process.

1.3 What is statistics?

The word ‘statistics’ was derived from the Latin word ‘status’ meaning ‘state’. The word statistics is
closely linked to the administrative issues of the state such as facts and figures regarding health,
education, population, housing, agriculture, financial resources and so on. The term statistics is
used in the singular and plural sense and thus it has two meanings. In the singular sense it has a
capital letter "S" and is used to denote the field or discipline of study whose purpose is to gather,
analyse and interpret data for decision making purposes. In the plural sense it has a small "s" where
it is used to denote figures or numbers.

Definition 1.1
Statistics is the field of study that involves the gathering or collecting, processing,
analysing and interpretation of data for decision making purposes.

It is a way of turning data into useful information. It is an art and science one uses to gather,
process, analyse and interpret data for one to be able to make meaningful decisions. One can say
that statistics involve five stages which are:

1. collection of data,

2. organisation of data,

3. presentation of data,
 3 STA1506/1

4. analysis of data, and

5. interpretation of data.

Statistics assist one in answering the questions like: What is the impact of the monetary policy on
inflation? What is the relationship between leadership style and the success of an organisation?
How does talent management strategies enhance organisational performance? All these questions
need to be answered and hence the study of statistics.

1.4 Main branches of statistics

Statistics has two main branches which are descriptive statistics, which involves summarisation of
data into a readable form, and inferential statistics, which involves drawing conclusions based on the
data processed.

Descriptive statistics
The main purpose of descriptive statistics is to be able to organise data into a format that allows
one to get a feel of the data set, without making any conclusions about the data set. An example
is suppose you are given profits of 500 small and medium enterprises (SME) in the last quarter. By
just looking at the 500 figures one can not make head or tail of the data. However if you find the
mean, median, mode, variance and standard deviation, or draw a chart of the data, you can be able
to explain how the SMEs are performing in South Africa.

Definition 1.2
Descriptive statistics is a branch of statistics that involves the collection,
presentation and characterisation of data set in a convenient and informative way,
that allows one to properly describe the various features of the data set.

Descriptive statistics are methods that involve organising, picturing, and summarisation of data from
samples or populations (Brase & Brase, 2015:10). According to Keller (2018), descriptive statistics
can be grouped into numerical techniques and graphical techniques. Numerical techniques involves
summarisation of data. For example in STA1505, you learnt about measures of central tendency like
mean, median and mode which are used to describe the center of a set of data. You also learnt
about measures of position that allows you to determine the position of an item or object or individual
with respect to the rest of the group like percentiles. You also learnt about measures of dispersion
or variability such as variance and standard deviation that allows you to determine how far apart
observations are.
 4

How does one know that a car needs to be serviced or in a production factory a machine needs to
go for service? If you are used to driving a car and you observe that it is consuming a lot of fuel then
you know that there is something wrong; it needs to be examined by a mechanic. Why is that? It is
because the consumption of fuel is deviating from the norm. You already know the fuel consumption
of your car and since it is now varying from the usual, you understand it needs a specialist in cars to
look at it.

Graphical techniques are methods that allow practitioners to present data in ways that makes it easy
for the reader to extract useful information (Keller, 2018:2). Graphical techniques can be in the form
of charts, diagrams or plots. These are pie charts, dot plots, bar charts, stem and leaf plots, box
plots, histograms, frequency polygons, ogive curves, scatterplots, time series plots, etc.

Although descriptive statistics is important in characterising and presenting data, it only allows one to
get the feel of the data but one can not make decisions with such information. Descriptive statistics
is often used as a step carried out before inferential statistics can be applied. The development
of statistical inference has been seen at the center of decision making that has led to the greater
application of statistical methods.

Inferential statistics
Statistical inference is seen as a branch in statistics where one uses sample information to draw
conclusions about the underlying population. Inferential statistics tend to answer questions like:

1. Is the salary for men and women in similar positions the same?

2. Is there a relationship between employee engagement, job satisfaction and job performance?

3. Is the average assignment one mark in STA1506 70%?

4. Does the lifetime of a 100 watt bulb exceed 3000 hours?

5. Has the average age at which women in South Africa are starting to give birth risen to 25 years?

6. Is the proportion of people abused by their partners 20%?

Definition 1.3
Inferential statistics is a branch of statistics that involves making judgement about
a population using information derived from a sample.

Thus the goal of inferential statistics is to use information obtained from the sample to make
conclusions on the underlying population.
 5 STA1506/1

Activity 1.1
1. What is your understanding of the word "statistics"?
2. Distinguish between descriptive and inferential statistics.

1.5 Nature of data

In order for one to be able to analyse or make meaning out of data, one needs to understand the
nature of the data. This assists one in assessing the quality of the data and being able to choose a
technique best suited to analyse the data.

Data are raw facts or figures. It is a collection of observations related to a given set of variables,
for example daily amount of rainfall recorded in the month of February 2020.

A data set is a set of collected data. A data set contains information about some group of
individuals, objects or items. The information is organised into variables.

Individuals are the objects described by a set of data.

A variable is any characteristic of an individual or any characteristic that varies from one
subject to the other. Thus a variable is a characteristic or condition that changes or has
different values for different individuals (Gravetter & Wallnau, 2017:4).

Data is influenced by three factors which are:

1. data type,

2. data source, and

3. method of collecting data.

The method of collecting data will not be discussed in this module. The purpose of this module is to
assume that data has already been collected and one would like to analyse it and come up with a
statistical report.
 6

Types of data
Data can either be classified as qualitative (categorical) data or quantitative data.

Definition 1.4
Qualitative data is data that cannot be measured numerically
but can fall into one or more non-numeric categories.

It is attribute data that can be separated into different categories distinguished by some non-numeric
characteristics. A qualitative variable is a variable that can not be measured, that is, its non-numeric.

Examples of qualitative or categorical variables are:

gender of an individual e.g., male or female.

levels of management (senior management, middle management, lower management).

types of blood groups (A, B, AB, O).

species of snakes (anaconda, king cobra, etc.).

make of cars (Toyota, Mazda, Ford, Mercedes Benz, Jaguar, Hyundai etc.)

Definition 1.5
Quantitative data is data that can be measured numerically

A quantitative variable is one that can be expressed as numeric quantities.

Examples of quantitative variables are:

Weight of a new born baby.

The amount of money in your pocket.

Number of aeroplanes owned by South African Airways.

There are two types of quantitative data, that is, discrete data and continuous data.

Definition 1.6
Discrete data is data that takes specific fixed values
 7 STA1506/1

A random variable that takes specific fixed values is called a discrete random variable.

Examples of discrete random variables are:

Number of books in the library.

Number of people who contracted the Covid-19 virus.

Number of degrees conferred by the Vice Chancellor of a University.

Number of small and medium enterprises in Gauteng.

Definition 1.7
Continuous data is data that takes values within a range or interval.

A continuous variable is a variable that is capable of assuming any of the values in a given range.

Examples of continuous random variables are:

Age of a person.

Height of a tree.

Time taken to write an examination.

Distance travelled by an ambulance.

Salary of a person.

In summary
Individual Characteristic Type of characteristic
A person Height in meters continuous variable
Weight in kgs continuous variable
Colour of eyes categorical variable
Type of blood group categorical variable
Marks in computing discrete variable
Sex categorical variable
Bread Length in cm continuous variable
Stale or not stale categorical variable
Thickness in cm continuous variable
Senelisiwe’s firm Size of firm discrete variable
Amount of profit continuous variable
Number of contracts discrete variable
 8

Measurement scale
Measurement is assigning a value or score to an observation.

Definition 1.8
Measurement is the assignment of numerals to objects or events
according to set rules for the quantity.

They are four types of measurement scales which are:

nominal scale,

ordinal scale,

interval scale, and.

ratio scale.

Nominal scale
This is the weakest measurement scale that is associated with categorical data. It is used when
objects or items are distinguished from each other by naming. No meaningful calculations can be
obtained from the data. For example when males and females are coded as 1 and 2 respectively.
If there are five males and five females, an average of 1.5 does not mean a thing. Can we say
someone is 0.5 more than a man and 0.5 less than a woman? The codes are just arbitrary.

Definition 1.9
Nominal scale is a type of measurement scale where you assign
objects or items into two or more categories of equal importance

It consists of names, labels or categories. It just classifies categories where ordering is not explicit
or implicit.

Examples of nominal data are:

Gender of a person.

Types of blood groups.

Survey response of true, false or do not know.

 9 STA1506/1

Colour of a dress.

Defectiveness of a machine as defective or not defective.

Species of a snake.

Ordinal scale
This is the second weakest measurement scale that is associated with categorical data that can
be arranged in order although differences between values are meaningless. The ordinal scale
provides information about relative comparisons but the degrees of differences are not usable for
or are meaningless.

Definition 1.10
Ordinal scale is a type of measurement scale where data can be arranged in
some order, but differences between data values either cannot be determined
or are meaningless.

The categories form a rank or continuum (Jackson, 2014). The Likert scale is a good example of an
ordinal scale.
Very dissatisfied Dissatisfied Somewhat Satisfied Very satisfied
satisfied
(1) (2) (3) (4) (5)
Satisfaction with salary

This is a five-point Likert scale that ranges from 1 (very dissatisfied) to 5 (very satisfied). In this case
we can not say a score of four (Satisfied) is twice a score of two (Dissatisfied). Ordinal scale allows
one to tell which is more or less of the characteristics being assessed but it is not possible to tell how
much more or less of the characteristic one object has than another.

Examples of ordinal data are:

Rating of hotels e.g., five star, four star etc.

Levels of agreement (strongly agree, agree, neutral, disagree, strongly disagree).

Size of tee-shirts e.g., small, medium, large etc..

Ranks of academic staff in a university e.g. full professor, associate professor, senior lecturer,
lecturer, junior lecturer etc.

Grades in a course as A, B, C, D and E.

The rate at which typists type as fast, average and slow.

 10

Interval scale
Interval scale is associated with numerical data and quantitative random variables. In an interval
scale the data can be arranged in a certain order and addition and differences are meaningful.
Interval scale is a type of measurement scale in which scores measure actual amounts, but zero
does not mean zero is present, so negative numbers are possible (Heiman, 2015:16). The interval
scale is the level of measurement which is like the ordinal level, with the additional property that we
can determine meaningful amounts of differences between the data.

Definition 1.11
A measurement has an interval scale if the numbers also tell us that one individual
“differs by a certain amount” of the property from another individual.

The uniqueness is that it has no absolute origin, that is, it has no inherent (natural) zero starting
point (where none of the quantity is present) . A zero in an interval scale does not necessarily mean
absence of the trait being measured For example 0o C does not necessarily mean that there is "no
heat" or that the temperature is absent. Secondly in an interval scale addition and subtraction are
possible but not multiplication and division. It is wrong for one to say 20o C is twice 10o C:

Time at which an event is taking place is another example of an interval scale. 12 o’clock midnight
is 0000 hours. It does not necessarily mean time is absent. Comparing eight o’clock in the morning
and four o’clock in the afternoon, one can not say 1600 hours is twice 0800 hours but rather you can
comfortably say that the difference is 8 hours.

The characteristics of interval data are:

ordered, constant scale but no natural zero.

differences make senses, but ratios do not.

Examples of interval data are:

Fahrenheit and Celsius temperature scales.

Time at a particular moment.

Scholastic Aptitude test like Graduate Record Examination (GRE).

Note: An interval scale is stronger than an ordinal scale. For time measurements, time at which an
event is taking place is an interval scale but how many hours one take to write an examination is a
ratio scale.
 11 STA1506/1

Ratio scale
This is the strongest measurement scale and it is associated with quantitative data. It is applicable to
data that can be arranged in order. In a ratio scale addition, subtraction, division and multiplications
are meaningful, that is, both differences between data values and ratios of data values are
meaningful. A zero in a ratio scale means absence of the trait or attribute being measured, that
is, it has a true origin. The characteristics of ratio data are therefore order, constant scale, natural
zero.

Definition 1.12
A measurement has a ratio scale if in addition the number tell us that
one individual has so "many times as much" of the property as does
another individual.

When measuring physical quantities such as weight, it means that 20 kgs is twice as heavy as 10
kgs and 0 kgs would mean that the object is weightless.

Examples of ratio data are:

Weight of a person.

Height of a tree.

Amount of rainfall.

Salary of a person.

Number of aeroplanes in South Africa .

Number of pages in a book.

Note:
(1) Qualitative data can either be nominal or ordinal whilst quantitative data can be interval or
ratio.

(2) Recent books are now combining interval and ratio and are calling the scale of measurement
interval (e.g., Keller, 2018)
 12

In summary:

The summary information on the measurement scale is

Type of Characteristics Examples Descriptive Level of

Measurement of variables statistics information
content
Nominal Categories only. Gender, types Counts, Lower
Data cannot be of blood proportions,
arranged in an groups, colours modes
ordering scheme.
Ordinal Ordered categories Likert scales, Counts, Intermediate
with intervals that levels of proportions,
cannot be quantified. management, modes
Differences cannot be size of tee-shirt,
determined or they degree of pain
are meaningless.
Interval Ranked spectrum Temperature, Counts, Higher
with quantifiable Scholastic proportions,
intervals. Differences Aptitude tests, modes,
between values can be time at which medians
determined, but there an event is
may be no inherent taking place
starting point. Ratios
are meaningless.
Ratio Like interval Height, age, Counts Highest
ranked spectrum weight, number proportions,
with quantifiable of hours taken to modes,
intervals but with write an medians,
an inherent starting, examination, means, variances,
point. Ratios are number of defects standard
meaningful. deviations
 13 STA1506/1

Activity 1.2
1. Classify the following variables as qualitative or quantitative.
If quantitative, state whether discrete or continuous.

(a) Number of books in the library.

(b) Service level of a hotel.
(c) Circumference of a tree.
(d) Ages of swimmers at the world championships.
(e) Marital status of employees.
(f) Time at which an exam starts.
(g) Brand of tea you prefer.
(h) Temperature of water.
(i) Time taken to finish a church service.
(j) Types of child abuse (physical, sexual, emotional or verbal).
(k) Monthly premiums payable on an insurance policy.
(l) Flying time between Cape Town and Johannesburg.
(m) Mass in kgs of bag of carrots.
(n) Number of outlets owned by Checkers departmental store.

2. Using the above variables, classify them as nominal, ordinal,

interval or ratio.

Data sources
Statisticians, planners, analysts etc., obtain data from different sources. Data sources can be
classified as (a) internal or external and as (b) primary or secondary data sources.

Internal data sources

Internal data sources are those data generated during normal working hours of a company or
organisation or firm.

Examples of internal data sources are:

Human resource data such as employee personal files, salaries and wages schedule in an
organisation.

Marks of students at a university.

Number of people suffering from coronavirus at a hospital.

 14

Financial data such as sales vouchers, credit notes, accounts receivable, asset register.

Production data such as production cost records, stock sheets, downtime records.

External data sources

External data sources is data that is available outside an organisation. This is data from public
institutions, private companies, state enterprises or government bodies. The cost and reliability
of external data depends on the source. Normally public or government institutions are less
expensive than private institutions. Some of the external data can be accessed through the
internet. For example data on the population distribution in South Africa can be accessed from
Statistics South Africa ([Link]), financial data and performance of companies from
the Johannesburg Stock Exchange (JSE) ([Link]) and data from the South African Audience
Research Foundation (SAARF)([Link]) and other marketing surveys for all media products
surveys (AMPS) and accident crash statistics at Arrive Alive ([Link]).

Primary data sources

Primary data are raw data that is captured for the first time at source and with a specific purpose
in mind. It is information collected at the first time to meet certain objectives and data collection is
tailored made to answer the objectives..

An organisation or company needs primary data which meet its own specific needs. The data can be
collected through their statistics department or agencies may be employed for this purpose. Agencies
are especially useful for collecting primary data for firms on market conditions for specific products
like market surveys. Primary data can also be internal or external. Internal primary data is data that
can be obtained directly from an internal business process such as sale invoices, machine speed
settings and employee leave days Primary external data is data obtained through surveys such
as market surveys, human resource surveys like salary surveys, personnel surveys and economic
surveys.

Advantages of primary data are:

One has greater control over data accuracy since the amount of error can be controlled or
minimised.

Data is problem specific since it can be tailored made to suit specific needs.

The researcher can formulate generalisations about his/her set of respondents.

Primary data reveals problems for which timely remedial measures may be instituted.

Note: Collecting your own data means you have full autonomy over the questions asked and the
variables that you can have access to.
 15 STA1506/1

Disadvantages of primary data sources are:

Takes time to collect, that is, it is time consuming.

It is expensive to collect.

May have a smaller sample size.

Secondary data sources

Secondary data its data that has been collected for other purposes other than the problem at
hand. It is processed data, that is, data that is already in existence. Secondary data can either
be internal or external secondary data. Internally sourced secondary data is past sales reports of a
company, "aged” market research figures, past salary scales of a company, previous past rates of
a university. Externally sourced data is data produced by external data sources like economic time
trends, employment statistics by Stats SA, or advertising expenditure trends in South Africa or by
sector from SAARF.

Advantages of secondary data are:

Data is readily available especially if it is on the internet.

Time to gather data is generally short.

Data is generally less expensive to obtain than primary data.

Disadvantages of secondary sources are:

Data may not be relevant to the problem at hand.

Data may be outdated (i.e., not current).

It may not be subject to further manipulation.

Reliability of secondary data is not always assured.

Combining various sources could lead to errors of collation and introduce bias.

In most cases primary data has more advantages compared to secondary data. However, secondary
data is also useful because it has already been collected and processed.
 16

Activity 1.3
1. Explain the difference, if any, between primary and secondary data.
Which is more reliable and why?

2. Explain the difference if any between

(a) primary data and internal data.
(b) secondary data and external data.

1.6 Population and sample

There is a saying which says that “You do not have to eat the whole cow to know that the meat is
tough.” A small piece of meat is enough to know whether the whole cow is tasty. This is the logic
behind sampling. It is sometimes difficult to gather all the data on a random variable. For example if
one wants to find the age at which women start giving birth in South Africa – every day new mothers
are coming around at the same time that women who have given birth are also dying.

Sometimes the population is inaccessible. For example one may want to study the reason why
aeroplanes crash. Think of the Malaysia Airlines Flight 370 that disappeared on 8 March 2014,
after departing from Kuala Lumpur for Beijing, with 227 passengers and 12 crew members on board.
Malaysia’s former Prime Minister, Najib Razak, stated that the aircraft’s flight ended somewhere in the
Indian Ocean, but no further explanation had been given The incident was still under investigation
in 2020, six years after its disappearance.

Often because of the size and costs involved in working with a population, we in practice use a
subset of a population called a sample.

In order to understand how statistical methods can be applied we need to understand about a
population and a sample and the terminology associated with the terms. Figure 1.1 shows the
population and the sample.
 17 STA1506/1

Figure 1.1: Population and sample

Source: Ott and Longnecker (2016:6)

Definition 1.13
A population is all the elements under investigation which we wish to
make some inferences on. This is the entire collection of all the elements
we are interested in.

For example if one is interested in determining at which age do women who enter university as
single ladies get married for the first time, then all the women who are in their first marriage who
were married after enrolling for a university programme are the population.

Examples of populations are:

All companies in South Africa.

All students in South Africa..a

All churches in South Africa.

All small and medium enterprises registered in South Africa..

The population should be clearly defined. A population is sometimes referred to as the target
population, underlying population or universe. The target population is the complete set of
elements or individuals that is the actual focus of a research investigation from which a sample
can be drawn.
 18

Definition 1.14
A population element is the subject on which the measurement is taken.
It is the unit of study or analysis.

A population element can be a company, an individual, an object or anything. It is an individual

member of the population. If a study involves looking at the financial data and performance of
companies at the Johannesburg Stock Exchange (JSE), then a company is the unit of analysis. If
one is looking at the weight of new born babies who are born to mothers who were taking drugs,
then a new born baby is the unit of analysis.

Definition 1.15
A census is the collection of data from all the elements or members
of the population.

When chooses to collect data from every one in the population without sampling, then a census has
been taken. It is an investigation of all the individuals that make up the population (Quinlan, Babin,
Carr, Griffin & Zikmund: 173). However in the business world, as well as many other fields, time is
an important factor. You need to make quick decisions in a short space of time and taking the whole
population will be time consuming. Thus, sometimes situations change fast and hence the need for
sampling in order to make accurate decisions. Imaging asking the voters who they are going to vote
for during election time. If you ask everyone you will not get the results before the elections are done.
Hence you take a group of people and ask them, then make your own conclusions from that group
to determine the favorite candidate. We sample in order to save cost and time.

Definition 1.16
A sample is a subset of the population under investigation or the
segment of the population selected for research (Bryman & Bell, 2015:538)

In research we aim to study a population. We want to make inferences or conclusions about the
population. The population need to be clearly defined and any exclusion and inclusion criteria stated.
The most important part is that the sample to be selected should be a representative sample, that is
one that resembles the characteristics of the population as closely as possible.
 19 STA1506/1

Definition 1.17
A representative sample is a sample selected in such a way that it closely
reproduces or represents features of interest in the population (Neuman, 2014:246-247)

The list of the elements under investigation is the sampling frame.

Definition 1.18
A sampling frame is the list of the entire population from which items
can be selected to form a sample.

It is the list of elements in the population from which the sample is actually drawn (Cooper &
Schindler, 2014:665).

The relationship between the sample and population is depicted in Figure 1.2

Figure 1.2: Relationship between a population and a

sample
Source: Gravetter and Wallnau (2017:4)

The aim of sampling is to draw conclusions about the population not the sample itself. Assume that
the World Health Organisation (WHO) wants to find the average age at which women start giving
birth in South Africa. As I said before this is a population where new mothers are coming around
while some mothers are dying. If one takes a representative sample across South Africa one might
 20

find out that the average age is 22 years for that sample. Then the question is, what conclusions
should we make to tell to WHO about the situation in South Africa?

In research we need to make distinctions about calculation based on the sample and calculations
based on the entire population. Population and sample information is summarised by making certain
calculations called summary measures. The names of the summary measures depend on whether
the calculations are from a sample or population.

Definition 1.19
A parameter is a numerical measure that describes the characteristics of a population.

It is a descriptive measure of a population or a numerical attribute of the population or a number that

describes the aspect of the scores in the population (Heiman, 2015). A parameter is a fixed number,
but in practice we do not know its value.

Definition 1.20
A statistic is a numerical measure that describe the characteristics of a sample.

It is a descriptive measure of a sample or a numerical attribute of the sample or a calculation from

the sample. The value of a statistic is known when we have taken a sample but it can change from
sample to sample. We often use a statistic to estimate an unknown parameter. The most common
parameters and statistics are shown below.

Summary measure Parameter Statistic

Total number of cases N n
Mean (mu) x
Standard deviation (sigma) s
Variance 2 s2
Correlation (rho) r
Proportion (pi) p

For a population the mean and the variance are denoted by and 2 respectively, while in a sample
they are denoted by x and s2 , respectively.

Why sample?
The main reasons for sampling are cost, timeliness, infinite population, inaccessibility of some of the
population, destructive testing and accuracy.
 21 STA1506/1

1. Cost: Directly observing only a portion of the population requires fewer resources than a census.
It is generally less costly to gather sample data than to use the whole population.

2. Timeliness: In statistics time is important and decisions have time limits. Gathering census data
is time consuming and hence by the time you get the data it will not be useful while sample data
can be gathered more timeously.

3. Infinite population: Some populations are very large or infinite. The size of the population will
probably make it physically impossible to conduct a census. In most cases the population is
infinite. An infinite population makes it virtually impossible to evaluate.

4. Inaccessibility of populations: Some populations contain elementary units so difficult to

observe that they are in a sense inaccessible or in certain cases the elementary units’ accessibility
may be limited for economic reasons alone that is, observation deemed too costly to be made, in a
real sense, are also inaccessible. If one is studying the population of aeroplanes that crashed then
some have not been found and some has sunk in the deep ocean such that they are inaccessible
and cannot be directly studied to examine the physical causes of their failure. In this case sampling
must be used to provide the desired information about the population.

5. Destructive Testing: Sometimes sampling involves destroying a unit. For example when testing
the defectiveness of a fuse, it entails destroying the fuse. Taking a census would mean complete
destruction of all the fuses and in this case a sample of fuses must be sampled. Thus, a census
is not appropriate for such data gathering. Destructive testing often happens in quality control like
milk testing etc.

6. Greater accuracy of results: Deming (1990:26) was quoted as saying "sampling possesses the
possibility of better interviewing (or testing), more thorough investigation of missing, wrong, or
suspicious information, better supervision, and better processing than is possible with complete
coverage". Dealing with a large data set can result in errors. It is better to manage a sample
than a census and the errors can be controlled more effectively. A sloppily conducted census can
provide less reliable information than a carefully obtained sample. Thus, more accurate details
can often be sourced from a small sample than the whole population.
 22

Activity 1.4
1. Distinguish between the following terms:
(a) population and sample.
(b) parameter and statistic.
(c) sampling frame and representative sample.

2. Give six reasons why we sample instead of

taking a census.

1.7 Data and information

Data and information are terms which have been linked together for decades. You can not talk of data
without bringing in information and vice versa. Data is seen as observed values of a random variable
(Keller, 2014). When data has been processed, that is, it has been grouped in order to give it meaning
and make it interpretable then it has been turned into information. Zins (2007) documented in his
paper titled "Conceptual approaches for defining data, information and knowledge" 130 definitions of
data, information and knowledge from 45 scholars across the world. Below, only ten definitions have
been chosen from ten scholars on the definition of data and information. The definitions are shown
in the table below.
 23 STA1506/1

Definitions for data and information

Scholar Data Information
Information is data that is communicated
1 Data are a string of symbols.
has meaning, has an effect, has a goal.
Information is data which is collected
Data are raw material
2 together with commentary, context and
of information, typically numeric.
analysis so as to be meaningful to others.
Information is facts, figures, and other forms
Data are sets of characters, symbols,
of meaningful representations that when
numbers, and audio/visual bits that are
3 encountered by or presented to a human
represented and/or encountered in raw
being are used to enhance his/her
forms.
understanding of a subject or related topics.
Data are unprocessed, unrelated raw Information is data or knowledge processed
4
facts or artifacts. into relations (between data and recipient).
Data are representations of facts and Information is data organised to produce
5
raw material of information. meaning.
Data are discrete items of information
Information is facts and ideas communicated
that one would call facts on some
6 (or made available for communication).
subject or other, not necessarily set
within a fully worked out framework.
Data are facts and statistics that can Information is data that has been
7 be quantified, measured, counted, categorised, counted and thus give
and stored. meaning, relevance or purpose.
Information is a set of symbols that represent
Data are alphabetic or numeric signs,
knowledge. Information is what context
which without context do not have
creates/gives to data. It is cognitive. Normally
8 any meaning.
it is understood as a new and additional
element in collecting data and information
for planned action.
Information is organised data (answering the
Data are a set of symbols representing a
9 following basic questions: What? Who?
perception of raw facts.
When? Where?)
Data are the basic individual items of
Information is that which is conveyed, and
numeric or other information, garnered
possibly amenable to analysis and
10 through observation; but in themselves,
interpretation, through data and the context
without context, they are devoid of
in which the data are assembled.
information.

In summary, data are observed values of a random variable (Keller, 2014), that is, data is the set of
individual values associated with a variable. When data is processed it is turned into information.
 24

1.8 Data management

Data management is the process of preparing data before analysis of the data. It is the assembling
of data before analysis. This is when you prepare your data for its transformation into information.
According to Hair, Black, Babin and Anderson (2019) with the arrival of the large and diverse datasets
from Big data researchers may be spending a lot of time on data management rather than analysis.
The authors indicated that sometimes it is called data wrangling whose principal task is data fusion,
which is becoming more commonplace as researchers in all areas attempt to combine data from
multiple sources.

Definition 1.21
Data management is an administrative process by which the required data
is acquired, validated, stored, protected and processed by which its
accessibility, reliability, and timeliness is ensured to satisfy the needs of the data
users. ([Link]

Data management is the correct, safe and secure management of data while data are being
gathered, stored and analysed (Quinlan et al., 2019: 399). According to Schoenbach (2002) data
management aims to acquire data and prepare for analysis; maintain quality control and data security
and support inquiries, review, reconstruction, and archiving. Thus data management starts the
minute data is being collected until just before analysis start. Organisations and enterprises are
utilising Big Data more than before for better business decisions and gaining deep insights into
customer behavior, trends, and opportunities for creating extraordinary customer experiences (Hair
et al, 2019; Galetto, 2016).

Data management is treated as an ethical issue in any project and the security of data is
a fundamental aspect because data can be lost, misplaced or even stolen (Quinlan et al.,
2019). The authors further indicated that the safety and security of data should be the priority
of every researcher. Thus data management is the process of securing and protecting data.
Data management is a broad practice that encompasses a number of data disciplines, including
data warehousing, data integration, data quality, data governance, content management, event
processing, database administration, and so on (Russom, 2013).

Data management process

The data management process is a system that shows the steps or stages in data management.
A lot has been said by researchers on the data management process. Some have come up with a
process that involves seven steps and some five steps. However, when looking at the literature on
data management, most processes involves selection of software, data editing, data coding, data
entry, data cleaning, data storage and data security.
 25 STA1506/1

The data management flow chart is shown in Figure 1.4.

Figure 1.4: Data management process

The stages are discussed in the next subsections.

 26

Selection of software
Choosing which software to use is often a simple decision. Nowadays, the participants in research
can use online survey software to respond to survey questions. There are a lot of websites on the
internet designed for capturing the participants responses. Some of them are SoGoSurvey, Survey
Monkey, Typeform, Google Forms, Client Heartbeat, Zoho Survey, Survey Gizmo, Survey Plane etc.
In most cases the researcher develops the questionnaire using the websites and then sends a link
to the participants. Some of the organisations have formulated their own software for administering
surveys. When the process is done most of this online survey software have a way for exporting the
data into a spreadsheet like Microsoft Excel. One of the main drawbacks of this method is that the
participants should have basic computer skills and also have internet access.

However, in the case where the data was collected as a hard copy, spreadsheet software is the
easiest and clearest way to create the database. Microsoft Excel is often used where the spreadsheet
is just a table of data or matrix where rows correspond to cases or observations and columns
correspond to variables. Microsoft Excel data can be exported into various statistical packages.

In the medical field they use a software called Epi Info. Epi Info is a statistical software for
epidemiology developed by Centers for Disease Control and Prevention in Atlanta, Georgia. It has
been in existence for over 20 years and is currently available for Microsoft Windows, Android and
iOS, along with a web and cloud version. Epi Info is designed to do questionnaire design, data entry
and validation, data analysis including mapping and graphing, and creation of reports. Epi Info allows
the researcher to indicate the minimum and maximum responses to a question. For example if the
variable level of management has three categories, senior, middle and lower management which
have been given the codes 1 to 3, then when designing the data entry template the researcher can
indicate that levels of management only allows codes 1 to 3. If a data entry clerk tries to a enter a
four, the system will not allow and there by reducing errors in data entry. Secondly two people can
enter the same questionnaire and then you can compare their responses. The package will assist
you in determining the areas where they differ and thus assisting in minimising data entry errors.

Data editing
Before starting data entry, data needs to be edited. Sometimes respondents tend to indicate
two responses or in certain cases ignore skip questions or branching questions. For example
respondents may be asked to indicate whether they own a house or not. Then a follow up question
comes which says that if you own a house, how many bedrooms does it have. In such cases you
find respondents who have said they do not own a house going further to indicate the number
of bedrooms the house has. In another case a participant is asked how old they are and one
can indicate 40 years and then asked the later question of how many years they have lived at a
certain place, and the respondent indicate 50 years. It means the respondent has not given correct
information on one of the questions.
 27 STA1506/1

Data editing is the stage where one starts the cleaning of the data before doing the data entry. It is
the process where one examines the raw data to detect errors and omissions and to correct them, in
an endeavour to ensure legibility, completeness, consistency and accuracy of the data. The recorded
data must be legible so that it can be coded later.

Definition 1.22
Data editing is the process of checking the completeness, consistency,
and legibility of data making the data ready for coding and transfer
to storage. (Zikmund, Babin, Carr & Griffin 2013)

It is the process for correcting and modifying collected data before the data is coded to ensure
completeness, accuracy and uniformity of the data. Therefore data editing is the process for checking
and adjusting data for omissions, consistency and legibility before one embarks on data coding.

Data coding
When analysing quantitative data, we need a statistical software. There are many software packages
designed for quantitative analysis, some of them are Statistical Package for the Social Sciences
(SPSS), SAS 9.4, STATA, R, Minitab and Microsoft Excel. Microsoft Excel has an Add-on package
on statistical analysis of data called XLSTAT which is the statistical software for Excel. Most of these
software are case sensitive, which means that capital letters of the same word and small letters of
the same word would be treated as two different things. For example if the word male is written as
"Male", "MALE", "MaLe" etc., it is treated as a different word. This problem can be minimised by
assigning a code value to males, for instance the number 1 standing for male and a 2 standing for
female. This also assist in increasing the speed at which data is entered. Data coding is the process
whereby raw data are transformed into a standardised form that is suitable for machine processing
and analysis (Rubin & Babbie, 2016: 499). When using such packages, it becomes a necessity to
assign codes to the responses.

Definition 1.23
Data coding is the process of labelling data using a code that symbolises
or summarises the meaning of that data. (Saunders, Lewis & Thornhill, 2016:712)

In quantitative analysis coding is just assigning a value to a category. It is the process of assigning
numerical values to the data for purposes of analysis (Hair, Celsi, Money, Samouel & Page,
2015:480).
 28

However, when dealing with qualitative data analysis it is more intense where researchers talk of
three stages of coding which are open coding, axial coding and selective coding. According to
Saunders, Lewis and Thornhill (2016:194), open coding is reorganisation of data into categories;
axial coding is recognising relationships between categories while selective coding is integration of
categories to produce a theory. Charmaz (2006) simplifies these three stages to two principal stages
which are initial coding and focused coding. Our main focus in this course will be the coding we do
when analysing quantitative data.

The main focus of data coding in quantitative data analysis is to assign numeric values to various
responses. For instance, levels of agreement can be coded as 1 (strongly disagree), 2 (disagree),
3 (neutral), 4 (agree) and 5 (strongly agree). The objective of data coding is to code collected data
so that they can be summarised and analysed and the coding tend to vary with the type of data
collected and the objective of the analysis (Jones & Hidiroglou, 2013). Numerical data can also be
coded. For example in your questionnaire you might have just asked the respondents to indicate
their age and in your analysis you want to determine whether the issues differ by age category. Then
age can be coded as follows: Below 30 years as 1, 30 - 39 years as 2, 40 - 49 years as 3 and 50
years and above as 4. When coding there is need to produce a coding list or key or code book. A
coding key is a guide to all of the codes used in coding data, that is, the codes used to input the data
into a computer software program. You should know that coding is different from scoring. Scoring
is when you assign a value to your responses; you can talk about a high score or a low score, and
one can obtain an average score. When you code responses you are just assigning numerical labels
which you are just using to differentiate the various categories and it may be meaningless to find an
average to such codes.

Data input/data entry

When data has been coded the next stage is to input the data into the computer. This stage is
sometimes referred to as data input, or data entry or data capturing. The main purpose of data
capturing is to transfer the responses of respondents or participants into an electronic file. They
are several ways in which data can be captured. Some universities, colleges and schools assess
students using multiple choice type of questions. These types of examinations are optically scanned
and the results are automatically captured on the system. Thus method used to capture data
depends on data collection methods. As indicated earlier in some cases the respondents are sent
a link where they will be answering the questions using online internet platforms like SurveyMonkey.
The minute the respondents finishes answering the question and click the submit option, the
responses have already been entered in the system. Other examples are use of computer assisted
telephoning interviewing (CATI) and computer assisted personal interviewing (CAPI) systems where
the data is recorded directly in computer readable form using optical or magnetic character readers,
optical or magnetic mark readers, or micro-computers during fieldwork (Schleicher & Saito, 2005).
 29 STA1506/1

However, in some cases the data is on hard copy especially surveys collected using face to face
interviews and they need to be captured into the computer.

Definition 1.24
Data entry is the process of converting information gathered by secondary
or primary methods to a medium for viewing and manipulation; usually done by
keyboarding or optical scanning (Cooper and Schindler, 2014:655)

According to Jones and Hidiroglou (2013), the data capturing process when the method of data
collection is a survey involves three stages as shown in Figure 1.5.

Figure 1.5: The three stages of data capturing

Source: Jones & Hidiroglou (2013:461)
 30

Data entry or capturing is subject to potential errors which may rise due to wrong keying of a code,
or mistakes in hearing, writing or electronic recognition (Jones & Hidiroglou, 2013). Due to this there
is need to come up with quality control checks like data cleaning of the data.

Data cleaning
Data editing is the first process in data cleaning where data is edited to determine any irregularities.
When data has been entered some errors might rise due to the data entry clerk or analyst keying
in the wrong value. For example instead of entering a six one enters a five. Secondly data may be
inconsistent in that certain responses do not correspond, for example you would not expect a man
to indicate that he gave birth to two children; rather you expect the man to indicate that he fathered
two children. Phases of data cleaning requires an in depth understanding of all types and sources of
errors possible during data collection and entry processes (acaps, 2016:1).

Normally programmes like Epi Info can validate data entry. A questionnaire can be entered twice by
two different data entry clerks and the questionnaires can be validated to check where they differ.
The data is then cleaned by checking with the hard copy the option which is correct. This is being
done by some commercial firms who validate the data entry process by entering a questionnaire
twice. In some cases a researcher has collected secondary data and within the data certain values
are outliers, that is, values which are very different from all the other values. In multivariate analysis
techniques, outliers tend to affect the analysis results. In this case special methods are used to
detect those outliers and deal with them, but this is beyond the objectives of this course. You will
learn more about this if you decide to pursue a career in Statistics.

Some of the errors come up when the responses to that particular question ends up at 5 but there
are captured values greater than five. Common errors that occur are when someone enters a one
or two twice instead of just a one or two. These are corrected by going to the original questionnaire
and tracking where the error is. .

Definition 1.25
Data cleaning is the process that consists primarily in implementing
error prevention strategies before they occur and this involves repeated
cycles of screening, diagnosing, treatment and documentation
of this process (acaps, 2016).

According to acaps (2016), screening involves systematically looking for suspect features in
assessment questionnaires, databases, or analysis datasets, especially outliers, inconsistencies and
strange patterns within the data while diagnosis involves identifying the nature of the missing data or
 31 STA1506/1

errors; treatment deals with correction of the data which can be either leaving it unchanged, deleting
or correcting it; and documentation involves reporting the actions taken as well as archiving the
original data and the changed data. Thus, data cleaning is the process of identifying and removing
the errors in data (Wangikar & Deshmukh, 2011).

Data storage
Once the data has been cleaned the next phase is data storage. Nowadays there are several ways
to store data. Data can be stored physically by filing data in cabinet files, or it can be stored on
a computer on a hard drive, a flash drive or cloud storage. Data storage is the method used to
record data using computers or other devices. Cloud Storage is when a user/customer/company
save their data within the cloud instead of on a local system (Obrutsky, 2016). Cloud computing is an
on-demand access to virtualised IT resources that are outside the organisation ( Marks & Lozano,
2010).

Definition 1.26
Data storage is the process of archiving or keeping information
electronically or by filing.

The main purpose of data storage is ensuring that it can be assessed in future. Multiple ways to store
data are advised. If one stores data by just filing in filing cabinets, then if fire breaks out everything
is lost. Storing data on computers on hard drive or flash drive might result in the device no longer
working resulting in loss of data. Nowadays cloud storage enables one to access data anywhere in
the world. Using cloud computing has its advantages and disadvantages. One of the major concern
is the security of information on cloud storage.

Data security
Once data has been stored the question that remains is how secure is the data. Securing data
is one of the ethical principals in research. You need to protect the identity of your participants
or respondents. Data security over the recent years has become one of the biggest concern from
researchers especially on cloud computing. How safe is your data?

Definition 1.27
Data security is the methods or measures used to prevent unauthorised,
access to computers, databases and websites.
 32

It is also involves protecting the data from being encrypted by viruses. Stories have come up in
history where patient information has been compromised and also state information. One of the
biggest breaches was reported in early October 2013 by security blogger Brian Krebs where Adobe
originally reported that hackers had stolen nearly 3 million encrypted customer credit cards records
plus login data for an undetermined number of users. This was estimated to have an impact on 153
million people in the world ([Link]
[Link].) Platforms like Yahoo and LinkedIn have also experienced in recent years
data security breaches.

Activity 1.5
1. Distinguish between the following terms:
(a) data coding and data editing
(b) data storage and data security.

2. Explain the major steps of the data management processes

after data has been captured.
 33 STA1506/1

Exercise 1.1

1. For each of the following random variables, indicate whether they are

(i) categorical, discrete or continuous, and

(ii) the measurement scale (i.e. nominal, ordinal, interval or ratio).

(a) The weight of a car.

(b) Ice cream flavours available.

(c) The number of people who recovered from the coronavirus.

(d) The names of teams in the ABSA premier league.

(e) The time at which the President is announcing procedures for the lockdown in South
Africa.

(f) The wood types that can be used to make a desk.

(g) The level of satisfaction of services by municipality as very dissatisfied, dissatisfied,

somewhat satisfied, satisfied and very satisfied.

(h) Models of cars in a Mercedes Benz range.

(i) The Scholastic Aptitude Test of students.

(j) Monthly revenue of a company.

(k) Names of people.

(l) Amount of time taken to off-load a lorry.

(m) Make of a sewing machine.

(n) Species of birds.

(o) Body temperature of a person suffering from coronavirus.

(p) Money in your mother’s handbag.

(q) The types of crops produced by a farmer.

(r) Levels of ranks in the South African police as (Student constable, Constables, Sergeant,
Warrant officer, Lieutenant, Captain, Lieutenant-Colonel, Colonel, Brigadier, Major-
General, Lieutenant-General and General).
 34

2. Suppose you are the Finance Director at UNISA and you request the following

statistics. Classify the following data sources as either:

(i) primary or secondary, and

(ii) internal or external.

(a) Amount of tuition fees paid by each student this semester.

(b) Past pass rates in the College of Science Engineering and Technology (CSET).

(c) Market research survey on prospective student preferences.

(d) Employment statistics published by StatsSA.

(e) The financial reports of other universities in the country.

(f) Financial data on the performance of companies on JSE.

(g) The academic performance of staff who have accessed support from UNISA in their
studies.

(h) The salary paid to contract workers this month..

3. Explain what you understand by the term data management and outline the data management
process.
 35 STA1506/1

1.9 Learning outcomes

Use the following learning outcomes as a checklist after you have completed this study unit to
evaluate the knowledge you have acquired.

After studying study unit 1, you should know (and understand!) the following definitions:

Statistics

descriptive statistics

inferential statistics

qualitative data

quantitative data

discrete data

continuous data

measurement

nominal scale

ordinal scale

interval scale

ratio scale

primary data

secondary data

population

population element

census
 36

After studying study unit 1, you should know (and understand!) the following definitions (cont’d) :

sample

representative sample

sampling frame

parameter

statistic

data

information

data management

data editing

data coding

data entry

data cleaning

data storage

data security
37 STA1506/1
 38

STUDY UNIT 2
Basics of Statistical Software
2.1 Introduction
Everywhere in the world people need facts to support their decisions. If you are a football fan and
your team is entering a crucial game, you would like to know the odds of your team winning against
the other team. The Soweto Derby is a soccer rivalry between Premier Soccer League’s Kaizer
Chiefs and Orlando Pirates and whenever the two teams play, statistics are brought in of their past
encounters and odds ratios are computed by those who bet to determine which team is most likely
to win or lose or whether it will be a draw. In the English Premier League the 2018 to 2019 season
had Manchester City and Liverpool fighting it all out for the championships with statisticians playing
around with past encounters of the two teams. The battle ended with Manchester City winning the
Championship with 98 points ahead of Liverpool with 97 points. What can we say about this 1 point
difference?

When ten students write a test or an examination its easy for a teacher or lecturer to find the summary
statistics like mean, median, mode, standard deviation etc. What happens when 2 000 students write
an examination? It will be difficult for one to compute summary statistics manually due to the large
number of observations. Nowadays data is being collected in volumes and the data will be big.
Such data needs a software package in order to analyse it. According to Hair, Black, Babin and
Anderson (2019: 49) Big Data is a term now being given to the explosion in secondary data typified
by increases in the volume, variety and velocity of the data being made available from a myriad set
of sources like social media, customer-level data and sensor data. The authors further indicated that
the age of Big Data is now impacting all aspects of academics and practitioner researcher efforts.

As a statistician one needs to know analysis of data using several softwares. There are lot of
softwares available for the analysis of data such as SAS, IBM SPSS (Statistical Packages for the
Social Sciences), Minitab, Stata, Statistica, R, Eviews, etc. Some of the softwares you will learn
to use if you pursue further studies in statistics. If you are given the 2 000 scores in a Statistics
examination and you are asked to comment on the performance of the students, it will be impossible
unless you analyse the data. This chapter is designed to assist you in making the first steps in the
analysis of data by analysing data using Microsoft Excel.

The purpose of this unit is to discuss what data analysis is, to familiarize you with Microsoft Excel
and to explain how to analyse data in Excel.
 39 STA1506/1

2.2 Objectives of the Unit

By the end of this unit, you should be able to:

Understand the meaning of data analysis.

Understand what Microsoft Excel is.

Determine how to add the Analysis ToolPak in Microsoft Excel.

Enter data in Microsoft Excel.

Do graphical representation in Microsoft Excel.

Do basic calculations in Microsoft Excel.

Do statistics analysis in Microsoft Excel using Analysis ToolPak.

Know the critical use of a software.

2.3 What is Data Analysis?

Data analysis involves methods of making sense of observations. Data analysis consists of methods
of examining data. This is the process where raw facts are turned into meaningful information.
Scientists spend their entire careers learning how to analyse data and developing new methods of
analysing data in an effort to disseminate their results to both their academic peers and the general
public (Wrench et al., 2016:16). The scientists in the last decades have been spending a lot of time
analysing data searching for an HIV and AIDs cure and in 2020, scientists started to analyse data in
an endeavour to find the cure for coronavirus (COVID-19).

Definition 2.1
Data analysis is the process of analysing data gathered for a research project and
the process involves describing and interpreting the data (Quinlan et al, 2019:399)

According to Quinlan et al., (2019), the process of data analysis also involves drawing conclusions
from the data, and in research undertaken in academic institutions (colleges or universities), it
involves theorising the data, that is connecting the data gathered with the theory laid out in the
literature review. Data analysis can be qualitative data analysis or quantitative data analysis.
According to Babbie (2017), qualitative data analysis are methods for examining data without
converting them into a numerical format while quantitative data analysis are the techniques of
converting data into a numerical form and subjecting it to statistical analysis. In this course we
 40

are concentrating on quantitative data analysis. According to Babbie (2017) one of the initial steps
in quantitative data analysis is quantification which involves the process of converting data into a
numerical format. In this course quantitative data analysis is done to:

1. Draw graphical presentations of data,

2. find the three common measures of central tendency or location (mean, median and mode),

3. find measures of position or relative standing (percentiles, quartiles, deciles etc.),

4. find measures of dispersion or variability (range, interquartile range, standard deviation and
variance),

5. find measures of relative dispersion (coefficient of variation),

6. determine the extent or degree of the relationship between variables (correlation analysis), and

7. determine the functional relationship between two variables (simple linear regression analysis).

The manual computation of these statistics have been done in STA1505 and in this course the
calculations will be done using Microsoft Excel.

2.4 What is Microsoft Excel?

Microsoft Excel is a spreadsheet application that displays data in rows and columns in the form of
a matrix or grid comprising individual cells (Duignan, 2014:7). In order for one to access Microsoft
Excel or to open a workbook one can either double click on the icon of Microsoft Excel or on

windows applications one goes to the start button and then selects Excel. Excel is one

of the packages where data can be easily exported into other statistical softwares. An Excel file is in
the form of a workbook which has individual pages called sheets as shown in Figure 2.1.
 41 STA1506/1

Figure 2.1: Microsoft Excel Workbook

 42

Individual workbooks are created for specific task with data being entered in the cells.

Essential Features of the Excel Workbook

Microsoft Excel window consists of the following features:

Title bar.

File.

Home.

Insert.

Page layout.

Formulas.

Data.

Review.

View.

Add-ins

Help.

Title Bar
The Title bar is the one that displays whether Autosave is off or on, the undo

and redo buttons, arrow to access the Customise Quick Access Bar , the

name of the worksheet opened and the last time it was modified ,

the name of the user in this case (Muchengetwa, Suwisa) , Ribbon

Display Options and the Minimise , Maximise and Close

buttons. The title bar is given in Figure 2.2.

Figure 2.2: Title Bar

However, this also depends on the version of Microsoft Excel you have got. These are the
characteristics of the Microsoft Excel 2016 title bar.
 43 STA1506/1

File Menu
The File Menu just like any other package like Microsoft Word is the one that allows one to create a
new file, to open an existing file, to save a file and other functions shown in Figure 2.3.

Figure 2.3: File Menu

The info option allows one to protect your workbook by putting a password on your workbook, restrict
access etc.; inspect workbook, version history and browser view options.

Home Menu
The Home Menu is the second menu and it consists of functions that enable one to copy, paste,
change font, edit a document, wrap text, merge and center, change type of columns like whether
its number, currency and etc., do conditional formatting, format tables, cell styles, insert and delete
rows and columns, editing, autosums, finding and selecting and ideas as shown in Figure 2.4.

Figure 2.4: Home Menu

This is the menu that allows one to format documents like centering headings and aligning text and
figures.
 44

Insert Menu
The Insert Menu is the third menu in Microsoft Excel 2016 and it allows one to insert tables, pictures,
draw graphs and is shown in Figure 2.5.

Figure 2.5: Insert Menu

This is one of the menus you should know very well since it allows you to do graphical presentation
of data.

Page Layout
The fourth menu in Microsoft Excel 2016 is Page Layout which is shown in Figure 2.6.

Figure 2.6 Page Layout Menu

If you are a colourful persons, this is the menu you can use to change theme colours, change page
orientation, size, inserting page breaks etc. If you go Page Layout then Colors; you select the
colours you need. In our case we are using Office 2007 - 2010 colors.

Formulas
The Formula bar is the one you should know by heart. It is the one that contains the formula
functions for computing summary statistics and it is shown in Figure 2.7.
 45 STA1506/1

Figure 2.7: Formulas Menu

It also consists of autosum, financial formulas, trigonometry & maths and also calculation options
among other things.

Data
The Data Menu is useful in importing data by connecting with the server and we can import data
automatically from web, MS Access etc. You can also sort data, filter it and it makes it easy to read
vast data and it is shown in Figure 2.8.

Figure 2.8: Data Menu

The data menu is one of the important menu option in that it allows one to do data analysis. It also
has data tools and forecast options among other things.

Review
The Review Menu is shown in Figure 2.9.

Figure 2.9: Review Menu

 46

It allows one to spell check worksheets, protect sheets and also put a password to your document.

View
The View Menu has functions like page break, page layout, zooming the document and freeze panes
and it is presented in Figure 2.10.

Figure 2.10: View Menu

Freeze panes allows one to freeze rows or columns. If you freeze a row it means as you go down
that row will be visible. It can be used in data entry where one freezes the first row which contains
label names. In view you can also divide your window into different panes that each scroll differently
and can also hide and unhide worksheets.

Adds -ins
The Add-ins Menu is shown in Figure 2.11.

Figure 2.11: Adds-ins

The add-ins allows one to add extra functions like Analysis ToolPak and solver (for linear
programming) in Microsoft Excel.

Help
The Help Menu it has the same function like in any other software where you can use the function
to get help on any information about the package. The Help Menu is shown in Figure 2.12.
 47 STA1506/1

Figure 2.12: Help Menu

On the Help Menu you can contact support, see what is new for Microsoft Excel. When you click the
Help Menu you get the functions get started, collaborate, formulas & functions, import & analyse,
format data and troubleshoot. One can get help 24/7.

2.5 Adding the Analysis ToolPak Function

The Analysis ToolPak is used for complex statistical analyses in Excel for Windows. The tool needs
to be added in Microsoft Excel using the following commands.

1. Click the File tab, click Options and then click Add-ins category.

2. In the Manage box select Excel Add-ins and then click Go.

If you are using Excel for Mac in the file menu go to Tools > Excel Add-ins.

3. In the Add-ins box, check the Analysis ToolPak check box, and then click OK.

If Analysis ToolPak is not listed in the Add-ins available box, click Browse to locate it.

If you are prompted that the Analysis ToolPak is not currently installed on your compute, click
Yes to install it.

2.6 Entering Data in Microsoft Excel

In the previous unit we learnt that before data analysis, there is data entry. In order to ensure
accuracy in data entry there should be error free data entry. One can enter data in Excel, import it
from the internet or Microsoft Access or other softwares or retrieve it from a text file like files with
extension csv. The aim of this section is to show you how to enter data in Excel.

Suppose a large manufacturing firm with 8 000 employees has designed a training programme that is
supposed to increase the productivity of the employees. The personal manager decides to examine
this claim by analysing the data results from the first group of 20 employees that attended the
 48

course. The data is given below with the percentage increase in productivity (Y) measured against
an evaluation score (X) to see how successful the training was.

Employee Evaluation % Increase in Employee Evaluation % Increase in

number score productivity (Y) number score productivity (Y)
1 47 4:2 11 67 5:7
2 71 8:1 12 57 5:4
3 64 6:8 13 69 7:5
4 35 4:3 14 38 3:8
5 43 5:0 15 54 5:9
6 60 7:5 16 76 6:3
7 38 4:7 17 53 5:7
8 59 5:9 18 40 4:0
9 67 6:9 19 47 5:2
10 56 5:7 20 23 2:2

We want to enter this data in Excel. First open Microsoft Excel and, if necessary, choose to create a
blank workbook. When you open it, there will be an open worksheet called "Book 1" in the title bar.
Now do the following commands.

> Click cell A1 to make it the active cell.

> Type Employee number and then press Tab or right arrow.

> Type Evaluation score in cell B1 and then press Tab or right arrow.

> Type % Increase in productivity in cell C1 and then press enter.

> Excel moves you to cell C2 and then use the left arrow to go to cell A2.

> Type 1 in cell A2 and then press Tab or right arrow.

> Type 47 in cell B1 and then press Tab or right arrow.

> Type 4.2 in cell C2 and then press enter.

> Excel moves you to cell C3 and then use the left arrow to go to cell A3.

> Using the same technique type the next 19 rows of the table so that the data for the 20 employees
is entered.

> Then go and highlight cells A1 to C1.

> Click Home > Format > AutoFit Column Width to ensure that the names of the columns are
clearly visible. You can also do that by dragging each of the line to the right of your column until
the name fits the column perfectly.
 49 STA1506/1

> The worksheet should look like the information in Table 2.1.
Table 2.1: Evaluation of Training Program

Now click on File and then Save and save the file as "Evaluation of training program".

2.7 Tables and Graphical Presentation in Excel

In STA1505 you learnt how to construct tables and draw graphs. In this section you are going to
learn how to use Microsoft Excel to create the following.

Tables.

Pie chart.

Bar Chart.

Multiple or compound bar chart.

Dotplot.

Histogram.

Frequency polygon.
 50

Cumulative relative frequency.

Box plot.

Scatter plot.

Time series plot.

We are going to demonstrate these graphical methods using the data in Table 2.1, a data set
called shoppers and an example of time series data.

Chocolate Perfection Website Transactions

Chocolate Perfection manufactures and sells quality chocolate products in Dubai. Two years ago the
company developed a Website and began selling its products over the internet. Website sales have
exceeded the company’s expectations, and management is now considering strategies to increase
sales even further. To learn more about the Website customers, a sample of 50 Chocolate Perfection
transactions was selected from the previous month’s sales. Data showing the day of the week each
transaction was made, the type of browser the customer used, the time spent on the Website, the
number of Website pages viewed and the amount spent by each of the 50 customers are contained
in the file named ‘[Link]’ under myunisa on additional resources under the folder study guide
data sets. Amount spent in the United Arab Emirates dirham (AED). (One euro is around 5 AED.) A
screenshot of a portion of the data is shown below.

We would like to use this data to demonstrate how to do tables and some of the graphical
presentations. First download the data set ‘[Link]’ from myUnisa and save it on your
computer.
 51 STA1506/1

Frequency Tables
Firstly we would like to make a frequency table of the Type of Browser for the data set in
‘[Link]’. First, open the data set in Excel Click on File > Open, browse to the file on your
computer, and click Open. We would like to find out how many people use Firefox, Internet explorer
or other. Our data is in column C labelled as Browser. Now we do the following commands.

> Click the Insert Menu and then Pivot Table and you obtain the following table.

Figure 2.13a: Creation of Pivot Table

> Check the box select a table or range and then click the upward arrow.

> Give the range by highlighting cell C1 to cell C51.

 52

> Press the downward arrow or enter and you obtain the following table.

Figure 2.13b: Creation of Pivot Table

> Click on Existing Worksheet.

> Click the upward arrow for Location and highlight the location as H1.

> Press the downward arrow or enter and you obtain the following table.
 53 STA1506/1

Figure 2.13c: Creation of Pivot Table

> Click OK and you obtain the following tables.

Figure 2.13d: Creation of Pivot

Table
 54

Figure 2.13e: Creation of Pivot Table

> From Figure 2.13e, we check browser under Pivot Table Fields. and we obtain the following
table.

Table 2.13f: Creation of Pivot Table

 55 STA1506/1

> On cell I1 type Frequency then press enter.

> On cell I2, we are going to use the command =COUNTIF(Range,Criteria), where Range is cells
C2 to C51 (with our data) and criteria is H2 (Firefox). We are telling the computer to count the
number of times Firefox appears and 16 appears. You can see the command by Formulas >
Insert Function > change Most Recently Used to ALL and then scrow down. The functions
are in alphabetical order.

> Apply the same command on cell I2 to Cell I3 (=COUNTIF(C2:C51,H3) and I4

(=COUNTIF(C2:C51,H4) and you obtain the frequencies 27 and 7 respectively.

> Then put your cursor on cell I5 and calculate the sum by using the command =SUM(I2:I4) and
you obtain 50.

> You can double click on Row Labels and replace with Browser.

> Then under cell J1 type Percent.

> On cell J2 type the command =(I2/50*100) then press enter.

> Copy the command on cell J2 and paste it on cell J3 to J5 and you obtain the following frequency
distribution.

Table 2.13g: Creation of Pivot Table

Now we would like to create a cross tabulation between Day and Type of Browser.

Cross Tabulation Tables

We want to make a crosstabulation of Day and Type of Browser so that we know how many people
used Firefox on a Sunday or Internet Explorer or other and etc.

> Select the range B1 to C51 and click Insert Menu and then Pivot Table and you obtain the
following table.
 56

Figure 2.14a: Crosstabulation

> Click on Existing Worksheet.

> Click the upward arrow for Location and highlight the location as L1.

> Press the downward arrow or enter and you obtain the following table.

Figure 2.14b: Crosstabulation

> Click OK and you obtain the following tables.

 57 STA1506/1

Figure 2.14c: Crosstabulation Figure 2.14d: Crosstabulation

> Then drag Day to columns, Browser to rows and you can drag either of the two in this case we
choose to drag Browser to Values. Then click enter and you will obtain the following table.

Figure 2.14e: Crosstabulation

 58

The table shows 3 people used Internet explorer on a Sunday, 6 people used Firefox on a Friday
etc. You can change this table to proportions by calculating the relative frequencies by dividing each
frequency by 50.

.Figure 2.14e shows the cross tabulations done. In this case we now want to generate the total
percentages, row percentages and column percentages. The following screen should be showing
after doing your cross tabulation.

Figure 2.14f: Pivot Table Fields

The above table will be used to make total percentages, row percentages and column percentages.

Making Total Percentages

To make total percentages, follow the following instructions:
 59 STA1506/1

> After making the cross tabulation in Figure 2.14e, under the PivotTable Fields, click on the
downward arrow on Count of Browser under Values.

> Choose the option Value Field Settings then select Show Values As.

> Click on the downward arrow on No Calculation.

> To make total percentages, choose the option % of Grand Total and you will obtain the following
output.

Figure 2.14g: Table of Total Percentages

> Similarly row percentages can be obtained by selecting the option % of Row Total and you will
obtain the following output.

Figure 2.14h: Table of Row Percentages

> Lastly, column percentages can be obtained by selecting the option % of Column Total and you
will obtain the following output.

Figure 2.14i: Table of Column Percentages

 60

You can always go back to the original frequencies by choosing the option No Calculation.

Pie Chart
We want to create a pie chart of Type of Browser. We need the type of browser and frequency. Take
the data to another sheet from cell H1 to I4 (leaving the total) and paste using the paste 123 icon;
then you use the following commands

> Highlight the range in our case cells B2 to C5 (containing the labels and frequencies).

> Then go to the Insert Menu and choose the icon for pie chart and choose 3D pie.

> Drag the pie chart to be below your data set. You can first choose the format you need. In our
case we need one which allows us to put the title and also the percentages inside so we select
chart style 1 and obtain the following output.

Figure 2.15a: Pie Chart

> Double click on Frequency and type Pie Chart of Type of Browser.

> You can click on change colours and indicate the colour you need. In this case we choose the
one with yellow at the start, then light blue then purple.

> Put the cursor outside the pie chart but within the box and right click. Select Format Plot Area
then click Fill. Under Fill you can choose Gradient Fill, or Picture or Texture Fill or Pattern Fill
and etc. and choose the background you want but as for us we choose Pattern Fill.

> On Pattern Fill we go to Foreground and select the colour we need and we choose purple on
the third row and then select the first pattern.
 61 STA1506/1

> With the cursor on the pie chart we choose Chart Tools > Format >Shape Outline > Weight >
4 21 pt: We have selected a border line which is 4 21 pt and we obtain this chart.

Figure 2.15b: Pie Chart

> You can also put the frequencies by highlight all the frequencies and you select Format Data
Labels and then check the box on Value and you obtain the following pie chart.

Figure 2.15c: Pie Chart

You can play around with the options available and see how your pie chart changes.
 62

Bar Chart
We are going to create a bar chart of Type of Browser. We need the type of browser and frequency.

> Highlight the range in our case cells B2 to C5 (containing the labels and frequencies).

> Then go to the Insert Menu and choose the icon for bar chart and choose a

vertical bar.

> Click on the + (Chart Elements) to the right of the graph and check Axis Title.

> Double click on Frequency and type A Simple Bar Chart of Type of Browser; change the vertical
axis title to Number of Shoppers and horizontal axis title to Type of Browser.

> You can click on change colours and indicate the colour you need. In this case we choose the
one with purple.

> Put the cursor outside the bar chart but within the box and right click. Select Format Plot Area
then click Fill. Under Fill you can choose Pattern Fill then select the first pattern.

> With the cursor on the bar chart we choose Chart Tools > Format >Shape Outline > Weight
> 4 21 pt and we obtain this chart.

Figure 2.16: Bar Chart

Play around with the options available and see how your bar chart changes.
 63 STA1506/1

Multiple or Compound Bar Chart

We are going to create a multiple or compound bar chart of Pages Viewed versus Type of Browser.
The pages viewed were grouped into 2 - 5 pages and 6 - 10 pages and the following cross tabulation
was obtained.

> Highlight the range in our case D2 to F5 (containing the labels and frequencies).

> Then go to the Insert Menu and choose the icon for compound bar chart and choose 3D
column.

> Click on the + (Chart Elements) to the right of graph and check Axis Title.

> Double click on Chart Title and type Compound Bar Chart of Pages Viewed and Type of Browser.
and change the vertical axis title to Number of Shoppers and the horizontal axis title to Type of
Browser.

> You can click on change colours and indicate the colour you need. In this case we choose the
one with brown and purple.

> Put the cursor outside the compound bar chart but within the box and right click. Select Format
Plot Area then click Fill. Under Fill you can choose Pattern Fill then select the first pattern.

> With the cursor on the bar chart we choose Chart Tools > Format >Shape Outline > Weight
> 4 21 pt:

> Highlight all the bars for 2 - 5 pages and then right click. Select Format Data Series then click
choose Column Shape Cylinder. Click on 6 - 10 pages and then do the same and we obtain this
chart.
 64

Figure 2.17a: Compound Bar Chart

> You can also put the frequencies on top of the bars by highlight the bars and then choosing Add
Data Labels and we obtain the following chart.

Figure 2.17b: Compound Bar Chart

Play around with the options available and see how your chart changes.
 65 STA1506/1

Dot Plot
We are going to create a dot plot of Pages Viewed. First we do the commands of creating a bar chart
Firstly, we obtain the following frequency distribution.

> Then we do as if we are creating a bar chart. Highlight the range in our case cells K1 to L10
(containing the labels and frequencies).

> Then go to the Insert Menu and choose the icon for bar chart and choose a

vertical bar.

> Drag the bar chart to below the data . Choose Quick Layout number 5.

> Highlight all the bars and right click, then choose Format Data Series then Fill then Picture or
texture fill.

> On Picture Source click Insertthen choose Online pictures and type Dot and press enter.

> Choose the dot you want in our case the purple one and then click insert The output in Figure
2.18a is obtained. .
 66

Figure 2.18a: Dot Plot

> Click on Stack and Scale Width and then on Units/Pictures put 1.

> Click on the + (Chart Elements) to the right of graph and check the box on Axis Title.

> Double click on Chart Title and type Dotplot of Pages Viewed and change the vertical Axis title
to Number of Shoppers and the horizontal Axis Title to Type of Browser.

> Put the cursor out side the dot plot but within the box and right click. Select Format Plot Area
then click Fill. Under Fill you can choose Pattern Fill then select the first pattern.

> With the cursor on the dot plot we choose Chart Tools > Format >Shape Outline > Weight >
4 12 pt and we obtain the chart in Figure 2.18b.

Figure 2.18b: Dot Plot

 67 STA1506/1

Play around with the options available and see how your bar chart changes.

Histogram
We would like to construct a histogram of Amount Spend (AED).

> First we create a new worksheet by using the commands Home > Insert > New Sheet and copy
the data on Amount Spend onto column A of the new worksheet.

> Secondly, we use the MIN and MAX command to check the smallest and highest value by typing
on cell A54 and Cell A55 the commands; =MIN(C2:C51) and =MAX(C2:C51) respectively and
we obtain the values 65:47 and 581:73 respectively.

> Now we would like to create a histogram with classes 0 to less than 100, 100 to less than 200; up
to 500 to less than 600. On .Column D on cell D1 we type Bin. Then we enter the figure 0, 100,
200, 300, 400, 500, 600 in cells D2 to D8 to obtain the following table.

> Then go to the Data Menu and choose Data Analysis and then Histogram.

> Click the upward arrow on Input Range and highlight the cells A1 to A51 and click downward
arrow or enter.

> Click the upward arrow on Bin Range and highlight the cells D1 to D8 and click downward arrow
or enter.

> Click Labels since we have included cells A1 and D1.

> Click the upward arrow on Output Range and highlight the cell G1 and click downward arrow or
enter.

> Click on Chart Output and then select OK and you obtain the following output.
 68

Figure 2.19a: Histogram

> Delete the row under Bin with More and frequency 0.

> It can be observed that the histogram has gaps instead of being continuous. Highlight all the bars
and right click, then choose Format Data Series and reduce Gap Width to 0%.

> Change the horizontal label Bin to Amount Spend (AED), the Frequency to Number of Shoppers
and the title to Histogram of Amount Spend.

> Put the cursor inside the histogram and right click. Select Format Plot Area then click Fill. Under
Fill you can choose Pattern Fill and enter. Then do the same for the part inside the box but
outside the histogram

> Highlight all the bars and right click, then choose Outline and select first column third row and
obtain the following output.

Figure 2.19b: Histogram

 69 STA1506/1

> Highlight all the bars and then choose Add Data Labels and we obtain the following chart.

Figure 2.19c: Histogram

Practice by changing the bins and number of intervals and also changing the histogram bins to 0
- 99.99 then 100 to 199.99 and etc. after you plot the histogram and see the difference. Normally
a histogram has the upper limits as boundaries; we want you to experiment with Excel to do that.
Find out how to do it using help on the Internet through youtube. However we are warning you
that its a very long and tedious process but its worth trying.

Percentage Histogram
We would like to construct a percentage histogram of Amount Spend (AED).

> After creating the histogram, besides the column Frequency, create a column called Percent

> Then take each frequency and divide by the total no of observations by doing the command
=Cell/n(total number of observations). In this case it will be =Cell Number/50.

> Highlight the frequency, then click on the home page and change General to Percent

> Insert between Bin and Frequency a column called Midpoint and the following output is obtained
 70

> Highlight the percent column, then right click, format cells and change the number of decimals
to 0.

> Right click on the histogram and select data. Click on Edit the Horizontal (Category) Axis
Labels and highlight the midpoint series as your Series values and click OK.

> Right click on the histogram and select data. Click on Frequency and then Edit and highlight the
percent series as your series value and click OK.

> Change the vertical label from Frequency to Percentage.

> Highlight all the bars and then choose Add Data Labels and we obtain the following chart.

> Click on the histogram and then change the vertical label from Number of Shoppers to
Percentage of Shoppers and you obtain the following output.

Figure 2.19c: Histogram

You can practice by changing the bins and following the same instructions.
 71 STA1506/1

Frequency Polygon
We want to create a frequency polygon of Amount Spend. When plotting a frequency polygon we
use midpoints. We enter our data into columns J and K to obtain the following table.

> Then go to the Insert Menu and choose the icon for line graph and select first option of
basic simple line graph and we obtain the following chart.

Figure 2.20a: Frequency Polygon

> The horizontal axis is 1 to 7 and we need to change them to the midpoints values. Highlight all
the horizontal axis labels and right click and choose Select Data.

> Go to Horizontal (Category Axis Labels) and click Edit.

> Select Axis Label Range as cells J2 to J8 and click OK twice.

 72

> Click on the + (Chart Elements) to the right of the graph and check the box on Axis Title.

> Double click on Chart Title and type Frequency Polygon of Amount Spend and change the vertical
Axis title to Number of Shoppers and the horizontal Axis Title to Amount Spend (AED).

> Put the cursor out side the frequency polygon but within the box and right click. Select Format
Plot Area then click Fill. Under Fill choose Pattern Fill and go to Foreground and select the
colour you need and in our case we choose purple on the third row and then select the first
pattern.

> With the cursor on the frequency polygon we choose Chart Tools > Format >Shape Outline >
Weight > 4 21 pt and we obtain this chart.

Figure 2.20a: Frequency Polygon

> Highlight any point on the line and then choose Add Data Labels and we obtain the following
chart.
 73 STA1506/1

Figure 2.20b: Frequency Polygon

You can play around with the options available and see how your frequency polygon changes.

Cumulative Relative Frequency

We would like to construct a cumulative frequency curve (OGIVE) of Amount Spend (AED).

> Since we created a histogram already we use the upper limits of the frequency histogram and
create cumulative percentages in cell S1 to obtain the following table.

> Then go to the Data Menu and choose Data Analysis and then Histogram.

> Click the upward arrow on Input Range and highlight the cells A1 to A51 and click downward
arrow or enter.

> Click the upward arrow on Bin Range and highlight the cells S1 to S8 and click downward arrow
or enter.
 74

> Click labels since we have included cell A1 and S1.

> Click the upward arrow on Output Range and highlight the cell Y12 and click downward arrow or
enter.

> Check box on Cumulative Percentage and Chart Output and the select OK. The following output
is obtained.

Figure 2.21a: Ogive Curve

> Delete the newly created row under Upper limits with More, frequency 0 and Cumulative %
100%.

> Then highlight the series labels and then delete. Also Highlight all the histograms bars and delete
 75 STA1506/1

to obtain the following output in Figure 2.21b.

Figure 2.21b: Ogive Curve

> Then highlight the vertical labels and right click, then select Format Axis and on Axis Options
change maximum to 1 and press enter.

> Click on the + (Chart Elements) to the right of the graph and check box on Axis Title.

> Double click on Chart Title and type Ogive Curve of Amount Spend and change the vertical Axis
title to Cumulative % and the horizontal Axis Title to Amount Spend (AED).

> Put the cursor inside the ogive curve and right click. Select Format Plot Area then click Fill.
Under Fill you can choose Pattern Fill and click. Then do the same for the part inside the box
but outside the ogive curve

> With the cursor on the ogive curve we choose Chart Tools > Format >Shape Outline > Weight
> 4 12 pt and obtain the following output in Figure 2.21c.
 76

Figure 2.21c: Ogive Curve

> Click any point on the line and then choose Add Data Labels and we obtain the following chart.

Figure 2.21d: Ogive Curve

Practice by changing the format, pattern fill and etc.

 77 STA1506/1

Boxplot
Constructing a boxplot in Microsoft Excel 2016 is very easy compared to the previous versions.
Those with Microsoft 2019 are going to find some of the things very easy to do. In order to create a
boxplot we use the following commands.

> Highlight the range in our case cells C2 to C52 (containing the data for Amount Spend.

> Then go to the Insert Menu and choose the icon for histogram and choose Box and
Whisker and we obtain the following output in Figure 2.22a.

Figure 2.22a: Boxplot

> Click on the + (Chart Elements) to the right of the graph and check the box on Axis Title.

> Double click on Chart Title and type Boxplot of Amount Spend and change the vertical Axis title
to Amount Spend and delete the horizontal Axis Title and also the horizontal label 1.

> Put the cursor outside the boxplot and right click. Select Format Plot Area then click Fill. Under
Fill you can choose Pattern Fill and click.

> With the cursor outside the boxplot but within the boxplot we choose Chart Tools > Format
>Shape Outline > Click on any colour to obtain the borderline in this case blue and obtain
the following output in Figure 2.22b.
 78

Figure 2.22b: Boxplot

> Click any line on the box plot and then choose Add Data Labels and we obtain the following
chart.

Figure 2.22c: Box Plot

> The box plot shows that the maximum value 569:95 is an outlier; 493:25 is now the maximum value
which is not an outlier; upper quartile, Q3 is 308:77; median the second quartile, Q2 is 228:09;
the lower quartile, Q1 is 160:92; the minimum value is 65:47 and the mean, that is, the average is
250:03.
 79 STA1506/1

Scatterplot
We want to create a scatterplot of the data in Table 2.1 for the data set Evaluation of the Training
Program. A scatter plot assists in determining whether a linear relationship between two variables
exists. The commands for creating a scatterplot in Excel are:

> Highlight the range in our case cells B1 to C21 (containing the data for X and Y).

> Then go to the Insert Menu and choose the icon for scatterplots and select the first option
and we obtain the following output.

Figure 2.23a: Scatterplot

> Click on the + (Chart Elements) to the right of the graph and check on the box on Axis Title.

> Double click on % Increase in Productivity (Y) and type Scatterplot of % Increase in Productivity
(Y) versus Evaluation Score (X) and change the vertical Axis title to % Increase in Productivity
(Y) and the horizontal Axis Title to Evaluation Score (X).

> Put the cursor outside the scatter plot but within the box and right click. Select Format Plot Area
then click Fill. Under Fill choose Pattern Fill and go to Foreground and select the colour we
need and we choose purple on the third row and then select the first pattern.

> With the cursor on the scatterplot we choose Chart Tools > Format >Shape Outline > Weight
> 4 12 pt and we obtain this chart in Figure 2.23b.
 80

Figure 2.23b: Scatterplot

You can play around with the options available in formatting the graph and see how your scatter
plot changes.

Time Series Plots

A time series consists of data that has been taken at regular successive intervals. It is a set of
measurements, ordered over time on a particular quantity of interest. Suppose we are given the
following scenario.

The following set of data that shows the number of passengers for African Outward Passenger
Movements by Sea over the last three years.

2017 2018 2019

Quarter 1 2 3 4 1 2 3 4 1 2 3 4
Number of
2:2 5:0 7:9 3:2 2:9 5:2 8:2 3:8 3:2 5:8 9:1 4:1
passengers (millions)

Plot a time series plot of the data.

 81 STA1506/1

We need to put the data in a format we can draw a time series plot. We enter the data in Excel to
obtain this worksheet.

> Highlight the Y range in our case cells B1 to B12 (containing the data for Y).

> Then go to the Insert Menu and choose the icon for line graphs and select the fourth
option and we obtain the following output.

> Click on the + (Chart Elements) to the right of the graph and check on the box on Axis Title.

> Highlight the horizontal axis label and right click, Select Data then on Horizontal (Category)
Axis Labels click on Edit and choose the Axis Label Range as cells B2 to B13 and then click
the downward arrow or enter and then OK.
 82

> Double click on the title Number of Passengers (Y) and change to Time Series Plot of
African Outward Passenger Movements by Sea and change the vertical Axis title to Number
of Passengers and the horizontal Axis Title to Quarters.

> Put the cursor inside the box and right click. Select Format Plot Area then click Fill. Under Fill
choose Pattern Fill and go to Foreground and select the colour we need and we choose purple
on the third row and then select the first pattern.

> With the cursor on the time series plot we choose Chart Tools > Format >Shape Outline >
Weight > 4 21 pt: We have selected a border line which is 4 12 pt and we obtain this chart.

Figure 2.24b: Time Series Plot

You can play around with the options available in formatting the graph and see how your time
series plot changes.
 83 STA1506/1

Activity 2.1
On myunisa website under additional resources on the folder on study guide datasets
you will find the [Link] data set that was customised from the SPSS IBM
version 26 sample data sets. The data set contains nine variables which are gender,
age of employee, educational level (years), employment category, current salary,
beginning salary, months since hire, and previous experience (months).
Use this data to plot a
(a) Frequency distribution of educational levels.
(b) pie chart of employee category
(c) bar chart of employee category.
(d) multiple bar chart of gender distribution of employment category.
(e) Cross tabulation of minority and educational levels.
(f) Histogram of the ages of employee.
(g) Frequency polygon of the ages of employees.
(h) Ogive curve of the ages of employees
(i) Scatter plot of current salary and salary at the beginning.

2. The age of 28 applicants for a graduate program management trainee post are:
21 23 21 21 23 21 24 22 21 24 21 26 23 22
21 22 23 21 22 21 22 25 21 22 21 22 21 24
Produce a dot plot of the ages of applicant.

3. Sales of cold and flu treatment (in millions of Rands) in a Clicks Pharmacy during
the last three years are:
Quarter
Year 1 2 3 4
1 11:3 5:1 3:9 9:5
2 12:6 7:9 3:7 8:8
3 10:9 6:2 4:7 9:3
Plot a time series plot of the data.

2.8 Basic Calculations in Excel

In STA1505 you learnt about computing descriptive measures like measures of central tendency
or location (mean, median and mode), measures of relative standing or position (percentiles and
quartiles), measures of dispersion or variability (range, interquartile range, standard deviation and
variance), measures of relative dispersion (coefficient of variation), the correlation coefficient and
fitting a simple linear regression to a data set. In this section, you are going to learn how to
 84

calculate these summary measures using the Excel Formula function. If you go to Formulas >
Insert Function (fx) > change Most Recently Used to ALL and then you scrow down you will see
the following commands in alphabetical order.

AVERAGE
CORREL
INTERCEPT
MAX
MEDIAN
MIN
MODE
PERCENTILE
QUARTILE
[Link]
SLOPE
STDEV
VAR

These are the commands we can use to obtain the summary measures above. For the correlation
analysis and regression we are going to use the data given in Table 2.1 on Evaluation of Training
Program. Suppose we are given the following scenarios

Scenario A

1. A sample of 30 engineering graduates had the following starting salaries in thousands of Rands.

36.8 34.9 35.2 37.2 36.2 35.8 36.8 36.1 36.7 36.6
37.3 38.2 36.3 36.4 39.0 38.3 36.0 35.0 36.7 37.9
38.3 36.4 36.5 38.4 39.4 38.8 35.4 36.4 37.0 36.4

(a) What is the

(i) mean starting salary?

(ii) median starting salary?

(iii) mode?

(b) What is the

(i) minimum value?

(ii) maximum value?

(iii) first quartile?

(iv) second quartile?

(v) third quartile?

 85 STA1506/1

(vi) 20th percentile?

(vii) 57th percentile?

(viii) 70th percentile?

(c) Calculate the following summary statistics;

(i) range,

(ii) interquartile range,

(iii) variance, and

(iv) standard deviation.

(d) Calculate the coefficient of variation of the starting salaries.

Scenario B

2. Using the data in Table 2.1,

(a) calculate correlation coefficient between % Increase in Productivity (Y) and Evaluation Score
(X).

(b) estimate the simple linear regression equation of predicting % Increase in Productivity given
Evaluation Score.

We are going to use the two scenarios to learn how to use Excel to calculate our summary measures
using the Formula function.
 86

For Scenario A, we need to fill this template in Figure 2.25a.

Figure 2.25a: Scenario A Template

We are going to answer the questions on Scenario A using the template in Figure 2.25a.

Measures of central tendency or location

The measures of central tendency is the value that is at the centre or middle of a data set. As the
name implies it is the measure that locates the centre of a set of observations. The three most
common measures of central tendency are the mean, median and mode.
 87 STA1506/1

A. Mean
The mean is the average that is obtained by summing all the observations and then dividing by the
number of observations. In this case the command is obtained by Formulas > Insert Function (fx)
> change Most Recently Used to ALL and then AVERAGE which is

Figure 2.25b: Mean

You can calculate the average of two, three, four and so on numbers and just put the numbers or you
can give an array, that is, the cell range with the values. In our case our data has been entered in
Column A from cells A1 to A31 where Cell A1 contains the label name in this case Salary and cells
A2 to A31 contains the observations. The average will be obtained using the formula

=AVERAGE(A2:A31)

Whenever one does a formula in Excel you start with an equal sign (=) and then the command. In
this case we are telling Excel to calculate the average for the observations in cells A2 to A31. The
command will give the mean equal to 36.88. It is affected by extreme values or outliers.

B. Median
The median is the number which is at the middle of the distribution when data has been ordered in
descending or ascending order. It is the value that has 50% of the observations less or equal to it
and 50% of the observations greater or equal to it. It is the second quartile, Q2 or 50th percentile.
The command of the median is obtained by Formulas > Insert Function (fx) > MEDIAN and is

Figure 2.25c: Median

Thus the median will be

=MEDIAN(A2:A31)

The command will give a median of 36.65. It is not affected by extreme values or outliers.
 88

C. Mode
The mode is the number that occurs the most, that is, the number that occur with the greatest
frequency. Sometimes a data set does not have a mode, that is, if no value occurs more than once
and sometimes a data set has more than one mode. If two values occurs with the same greatest
frequency, then each one is a mode and the data set is bimodal. If three values occurs with the same
greatest frequency then each value is a mode and the data set is said to be trimodal and if there are
many values that occur with the same greatest frequency then all of them are modes and the data
set is multimodal. In Excel, the mode is obtained by Formulas > Insert Function (fx) > MODE
which is

Figure 2.25d: Mode

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25e: Mode

Thus the mode will be obtained using the formulas

=MODE(A2:A31) or =[Link](A2:A31)

The command will give a mode equal to 36.4.

It can be observed that the mean, median and modal starting salaries are R36 880, R36 650 and
R36 400: Experiment with the following functions AVERAGEA, AVERAGEIF, AVERAGEIFS and

[Link] and see the difference values the commands give you if any

Measures of relative standing or position

The measures of relative standing or position describe the position or place occupied by a data value
as compared to the rest of the data. The median is one such example that shows that 50% of the
scores are situated below it. Measures of relative standing are of practical value only when data sets
are large. When finding the value, the data is ordered either in descending or ascending order just
 89 STA1506/1

as we do when calculating the median. Besides the quartiles, other measures of relative standing
are the deciles and percentiles.

A. Minimum
The minimum value is the lowest value in a data set. The command is obtained by Formulas >
Insert Function (fx) > MIN which is

Figure 2.25f: Minimum

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25g: Minimum

The [Link] is inclusive it calculates the quartiles on the percentile range 0 to 1 inclusive.
The quart can take the values 0 (Minimum), 1 (Q1 ), 2 (Q2 ), 3 (Q3 ) and 4 (Maximum). Thus the
minimum will be obtained using the formulas.

=MIN(A2:A31) or =[Link](A2:A31, 0)

The command will give a minimum equal to 34.9.

B. Maximum
The maximum value is the largest value in a data set. The command is obtained by Formulas >
Insert Function (fx) > MAX which is

Figure 2.25h: Maximum

 90

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25i: Maximum

Thus the maximum will be obtained using the formulas.

=MAX(A2:A31) or =[Link](A2:A31, 4)

The command will give a minimum equal to 39.4.

C. Quartiles
There are three quartiles denoted by Q1 , Q2 and Q3 that divide the ordered data set into four equal
parts.
The first quartile Q1 , which is the 25th percentile divides the data set into 25% below and 75% above,
the second quartile, Q2 , is the median, which is the 50th percentile that is, the number that divides
the bottom 50% of the data from the top 50% and finally the third quartile,Q3 is the 75th percentile
which divides the data set into 75% below and 25% above.

Lower quartile, Q1
The lower quartile, Q1 is the value that has 25% of the observations less or equal to it and 75% of the
observations above or equal to it. The command is obtained by Formulas > Insert Function (fx) >
QUARTILE which is

Figure 2.25j: Lower Quartile

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25k: Lower Quartile

 91 STA1506/1

Thus the lower quartile, Q1 will be obtained using the formula.

=QUARTILE(A2:A31,1) or =[Link](A2:A31,1)

The command will give a lower quartile, Q1 equal to 36.225.

Second quartile, Q2
The second quartile, Q2 is the value that has 50% of the observations less or equal to it and 50% of
the observations above or equal to it. The command is obtained by Formulas > Insert Function
(fx) > QUARTILE which is

Figure 2.25l: Second Quartile

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25m: Second Quartile

Thus the second quartile, Q2 will be obtained using the formula.

=QUARTILE(A2:A31,2) or =[Link](A2:A31,2)

The command will give a lower quartile, Q2 equal to 36.65.

Upper quartile, Q3
The lower quartile, Q3 is the value that has 75% of the observations less or equal to it and 25% of the
observations above or equal to it. The command is obtained by Formulas > Insert Function (fx) >
QUARTILE which is

Figure 2.25n: Upper Quartile

 92

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25o: Upper Quartile

Thus the upper quartile, Q3 will be obtained using the formula.

=QUARTILE(A2:A31,3) or =[Link](A2:A31,3)

The command will give a lower quartile, Q3 equal to 37.75.

D. Percentiles
The 100k th percentile, Pk , is the value such that at least 100k% of the observations are at or below
this, value and at least 100(1 k)% of the observations are at or above this value. The command of
the percentile is obtained by Formulas > Insert Function (fx) > PERCENTILE and is

Figure 2.25p: Percentile

OR

Formulas > Insert Function (fx) > [Link] which is

Figure 2.25q: Percentile

The [Link] is inclusive it calculates the percentiles on the percentile range 0 to 1

inclusive. The value k can take the values 0:01 (1 percentile), to 1 (100th ). Thus the percentiles
will be obtained using the formulas.

Percentile k PERCENTILE [Link] Value

20th 0:2 PERCENTILE(A2:A31, 0.2) [Link](A2:A31, 0.2) 36:08

57th 0:57 PERCENTILE(A2:A31, 0.57) [Link](A2:A31, 0.57) 36:753

70th 0:7 PERCENTILE(A2:A31, 0.57) [Link](A2:A31, 0.7) 37:23

 93 STA1506/1

If you are in Excel and you would like to copy the formula and change only the second part you need
to make A2:A31 fixed, that is, to fix the cell reference so that there are absolute cell references rather
than relative cell references. You highlight A2:A31 and press fn on your keyboard and then press F4
and the range will be written as $A$2:$A$31. If not, the next cell will change A2:A31 to A3:A32.

Thus the 20th , 57th and 70th percentile are 36:08, 36:753 and 37:23 respectively.

Experiment with the following functions MAXA, MAXIFS, MINA, MINIFS, [Link], and
[Link] and see the difference values the commands give you if any.

Measures of Dispersion or Variability

Spread or dispersion is the extent by which the observations of a random variable are scattered
about the central value. It expresses how scores are far away from the central value. When values in
a data set are close to the mean of the data set, they exhibit less dispersion than when some of the
values are much larger and/or smaller than the mean. Two distribution might have the same mean
but different variability. For example the values 149, 150 and 151 have a mean of 150 while the values
50 150 and 250 have the same mean but far apart. The measures of dispersion that will be discussed
are the range, interquartile range, standard deviation and variance.

A Range
The range is the difference between the highest and lowest values in a data set. There is no direct
in build formula to calculate the range in Excel. Rather locate the cells with the highest and lowest
values. In our case its E10 and E14. Then the formula for calculating the range in this case is

Range = E14 E10

This gives a range of 4:5, that is, 39:4 34:9 = 4:5.

B. Interquartile Range
The interquartile range (IQR) represents the length of the interval covered by the centre half of the
observations, that is, it is the range of the middle 50% of the observations. It is the difference between
the lower and upper quartiles, that is Q3 Q1 . Similarly like the range there is no direct in build formula
to calculate the IQR in Microsoft Excel 2016. Locate the cells with the lower quartile and the upper
quartile in this case E11 or F11 and E13 or F13 respectively. Then the formula for calculating the
interquartile range in this case is

IQR = E13 E11 or IQR = F13 F11

 94

This gives an IQR = 1:525, that is, 37:75 36:225 = 1:525. One of the advantages of the interquartile
range is that it is not affected by extreme values or outliers.

C. Variance
The variance is a measure of the variation around the mean and it is one of the most used measure
of variation. In Excel, the variance is obtained by Formulas > Insert Function (fx) > VAR which is

Figure 2.25r: Variance

OR

Formulas > Insert Function (fx) > VAR.S which is

Figure 2.25s: Variance

Thus the variance will be obtained using the formula.

=VAR(A2:A31) or =VAR.S(A2:A31)

The command will give a variance equal to 1.413379. It is affected by extreme values or outliers.

D. Standard Deviation
The standard deviation is the square root of the variance. The standard deviation has the desirable
property of measuring variability in the original units of the variable of interest. In Excel, the standard
deviation is obtained by Formulas > Insert Function (fx) > STDEV which is

Figure 2.25t: Standard Deviation

 95 STA1506/1

OR

Formulas > Insert Function (fx) > STDEV.S which is

Figure 2.25u: Standard Deviation

Thus the standard deviation will be obtained using the formula.

=STDEV(A2:A31) or =STDEV.S(A2:A31)

The command will give a standard deviation equal to 1.188856. It is also affected by extreme values
or outliers.

C. Coefficient of Variation
The coefficient of variation (CV) is one that is used to compare the variability of different sets of data.
Coefficient of variation is unitless. The coefficient of variation is the ratio of the standard deviation
to the mean. We locate the cells with the mean and standard deviation which are cells E3 and E28
respectively. Then the formula for calculating the coefficient of variation is
standard deviation
CV =
mean
E28
=
E3

This gives a coefficient of variation of 0:0322:

Experiment with the following functions VAR.P, VARA, VARP, VARPA, STDEV.P, STDEVA, and
STDEVP and see the difference values the commands give you if any.
 96

In summary the worksheet with the solutions is

These are formulas that take long to do. The Analysis ToolPak gives the summary statistics much
faster and will be dealt with in Section 2.10.

Correlation Analysis

The Pearson correlation coefficient measures the extent or degree to which two variables are related.
It is suitable for assessing the strength of the association between quantitative variables, whose
values are interval or ratio data. The correlation coefficient ranges from 1 to +1 with 1 being a
perfect negative relationship, 0 no relationship at all and +1 being a perfect positive relationship. The
command is obtained by Formulas > Insert Function (fx) > CORREL which is

Figure 2.26: Correlation

 97 STA1506/1

Using the data in Table 2.1 given below

Thus the correlation will be obtained using the formula.

CORREL(B2:B21,C2:C21)

The command will give a correlation coefficient of 0:8902. which is a strong positive correlation
between % Increase in Productivity (Y) and Training Score (X).

Regression Analysis

Regression analysis is a technique used to estimate the functional relationship between two sets
of paired data. The purpose of simple linear regression analysis involve estimation of the equation
representing the relationship between the two variables, that is, in terms of a mathematical equation.
The estimated equation is

yb = a + bx

where a is the intercept and b is the slope. The command of the intercept is obtained by Formulas
> Insert Function (fx) > INTERCEPT which is
 98

Figure 2.27a: Regression

Using the data in Table 2.1, the command will be

INTERCEPT(C2:C21,B2:B21)

The intercept will be equal to 0.6712.

The command for the slope is obtained by Formulas > Insert Function (fx) > SLOPE which is

Figure 2.27b: Regression

Using the data in Table 2.1, the command will be

SLOPE(C2:C21,B2:B21)

The slope will be equal to 0.0915. This means the estimated equation is

yb = 0:6712 + 0:0915x

The following statistics will be done again using the Analysis ToolPak as you make your journey as
statisticians.
 99 STA1506/1

Activity 2.2
1. Using the employee data set in Activity 2.1 and the variable current salary,
calculate the following summary measures using the formula function.

(a) mean current salary.

(b) median current salary.
(c) modal current salary.
(d) minimum current salary.
(e) maximum current salary.
(f) first quartile.
(g) second quartile.
(h) third quartile.
(i) 30th percentile.
(j) 69th percentile.
(k) 99th percentile.
(l) range.

(m) interquartile range.

(n) variance, and.

(o) standard deviation.

(p) coefficient of variation.

2. Using the current salary and salary at the beginning,

(a) estimate the correlation coefficient.
(b) determine the dependent and independent variable.
(c) calculate the intercept.
(d) calculate the slope.
(e) give the estimated regression equation.
 100

2.9 Statistics analysis in Microsoft Excel Using Analysis

ToolPak
The purpose of this section is show you how to obtain the solutions to the questions in Scenario A
and Scenario B using the Analysis ToolPak.

Descriptive Statistics

In order to get the statistics for Scenario A, the descriptive statistics we go Data > Data Analysis
> Descriptive Statistics and you obtain the following box.

Figure 2.28a. Descriptive Statistics

Then click OK.

> Click the upward arrow on input range and highlight cells A1 to A31.

> Press the downward arrow or enter.

> Check the boxes Labels in first row and Summary statistics and you obtain the following table.
 101 STA1506/1

Figure 2.28b. Descriptive Statistics

> Click OK to obtain the following output.

Figure 2.28c. Descriptive

Statistics
 102

The following output does not give you the coefficient of variation and the quartiles and the
s
percentiles. It gives you more information like Standard Error of the Mean p , Kurtosis,
n
X
Skewness, Sum x and Count (n). There is the option of Rank and Percentile which will
give you the percentile for each of the values. You can do it as an exercise.

Correlation Analysis

In order to get the statistics for Scenario B, the correlation coefficient we go Data > Data Analysis
> Correlation and you obtain the following box.

Figure 2.29a. Correlation

Then click OK.

> Click the upward arrow on Input Range and highlight cells B1 to C21.

> Press the downward arrow or enter.

> Check the box Labels in First Row and you obtain the following table.

Figure 2.29b. Correlation

 103 STA1506/1

> Click OK to obtain the following output.

Figure 2.29c. Correlation

The following output gives you the correlation matrix indicating that the correlation between
Evaluation Score (X) and % Increase in Productivity (Y) is 0:8902 which is similar to the one obtained
using the CORREL function. You can find the correlation matrix of more than two variables at once.

Regression Analysis

In order to get the statistics for Scenario B, the regression equation we go Data > Data Analysis >
Regression and you obtain the following box.

Figure 2.30a: Regression

Then click OK.

> Click the upward arrow on Input Y Range and highlight cells C1 to C21.

> Press the downward arrow or enter.

> Click the upward arrow on Input X Range and highlight cells B1 to B21.

> Press the downward arrow or enter.

> Check the box Labels and you obtain the following table.
 104

Figure 2.30b: Regression

> Click OK to obtain the following output.

Figure 2.30c: Regression

 105 STA1506/1

Looking at the output it can be noted that the regression equation is yb = 0:6712 + 0:0915X , that is,

% Increase\
in Productivity = 0:6712 + 0:0915Evaluation Score
A lot of output is obtained. The coefficient of determination, R2 is equal to 0:7924.

The next chapter will discuss how these statistics are interpreted.

Activity 2.3
1. Using the employee data set in Activity 2.1 and calculate,
summary statistics of the continuous variables.

2. Find the correlation between

(a) current salary and salary at the beginning.
(b) salary at the beginning and previous experience .
(c) current salary and previous experience.

3. Estimate the simple linear regression model between

(a) current salary and salary at the beginning.
(b) salary at the beginning and previous experience .
(c) current salary and previous experience.

2.10 Use of a Software

Software applications are an integral part of any type of research (Comerford, 2012:1).
Using a software has many advantages. Some of the advantages of using a software are:.

Useful in the management, exploration, and analysis of datasets that are related to each other
(Annechino, Antin & Lee, 2010).

Improves accuracy.

Less time consuming.

Can use large data sets.

One can easily merge files.

Primary data can be easily converted to secondary data.

 106

Can assess the reliability and validity of data given.

The next chapter present the interpretation of data and what you should consider when writing a
report.

Activity 2.4
1. Explain what you understand by the term Big Data.

2. What are the advantages of using a software in data analysis?

 107 STA1506/1

Exercise 2.1

1. The Student Survey data found on the R MASS package is a data frame that contains the
responses of 237 students. The components of the data frame are:

Sex The sex of the student. (Factor with levels "Male" and "Female".)
span (distance from tip of thumb to tip of little finger
[Link]
of spread hand) of writing hand, in centimetres.
[Link] span of non-writing hand.
[Link] writing hand of student. (Factor, with levels "Left" and "Right".)
“Fold your arms! Which is on top” (Factor, with levels
Fold
"R on L", "L on R", "Neither".)
Pulse pulse rate of student (beats per minute).
‘Clap your hands! Which hand is on top?’ (Factor,
Clap
with levels "Right", "Left", "Neither".)
how often the student exercises. (Factor, with
Exer
levels "Freq" (frequently), "Some", "None".)
how much the student smokes. (Factor, levels "Heavy", "
Smoke
"Regul" (regularly), "Occas" (occasionally), Never".)
Height height of the student in centimetres.
whether the student expressed height in imperial (feet/inches) or metric
M.I
(centimetres/metres) units. (Factor, levels "Metric", "Imperial".)
Age age of the student in years.
The data was cleaned by removing cases (students) with missing information on some variables
and leaving the total number of observations of 168. The data is under myUnisa Additional
Resources on a folder called Study guide data sets with the name "R Survey (Mass) Dataset
No Missing Data". Use the file to do the following exercises in Microsoft Excel.

(a) Construct frequency distributions of all the categorical variables (sex, gender, fold, clap, Exer,
Smoke and M.I).

(b) Do a crosstabulation of sex and clap.

(c) Construct the following plots.

(i) pie chart of fold.

(ii) bar chart of exer.

(iii) multiple or compound bar chart of gender distribution across clap.

(iv) Histogram of pulse.

(v) Frequency polygon of the ages of students.

(vi) Ogive curve of heights of students.

(vii) Scatter plot of [Link] and heights of students.

 108

(viii) Scatter plot of [Link] and heights of students.

(d) Do a correlation matrix of [Link], [Link], Pulse, Height and Age using the correlation
command on the Analysis ToolPak.

(e) Estimate simple linear regression model of

(i) Predicting the span of writing head ([Link]) using the height of a student (Height).

(ii) Predicting the span of non-writing head ([Link]) using the height of a student (Height).

2. The ‘To Let’ column in the accommodation pages of a local newspaper contains details of 20
houses available to rent. The numbers of bedrooms in these properties are:
2 3 5 2 4 2 4 4 4 3
2 5 3 2 3 4 4 3 2 4

Using Microsoft Excel

(a) arrange this data into a frequency distribution.

(b) construct a dotplot of the data.

3. A machine vending company operates three hot drinks machines, one at a bus station, a second
at a train station and a third at a leisure center. The numbers of coffee, tea and chocolate drinks
dispensed from each in a single day are given in the following table.

Number of drinks
Location of machine Coffee Tea Chocolate
Bus station 68 91 23
Train station 105 74 47
Leisure center 49 67 89

Using Microsoft Excel,

(a) plot a simple bar chart to portray the total number of drinks dispensed at each location.

(b) plot a cluster bar chart (multiple or compound) to portray the number of drinks dispensed at
each location by the type of drink.
 109 STA1506/1

4. The management of an office complex in central Johannesburg wants to understand the pattern
of energy consumption (i.e., energy costs related to heating and air conditioning) in the complex.
They have assembled quarterly data on energy costs for the past three years (In R100 000s).

Year Quarter
Summer Autumn Winter Spring
2009 2:4 3:8 4:0 3:1
2010 2:6 4:1 4:1 3:2
2011 2:6 4:5 4:3 3:3

Using Microsoft Excel,

(i) plot the pattern of energy costs graphically.

(ii) Find the least squares trendline for energy usage in this office complex (Hint: Take the X-
values as 1, 2, ....12; where 1 is summer 2009, 2 is autumn 2009 and etc.).
 110

2.11 Learning outcomes

Use the following learning outcomes as a checklist after you have completed this study unit to
evaluate the knowledge you have acquired.

After studying study unit 2, you should know (and understand!) the following issues:

data analysis

qualitative data analysis

quantitative data analysis

what is Microsoft Excel

discuss the essential features of the Excel Workbook.

adding the Analysis ToolPak function in Microsoft Excel

entering data in Microsoft Excel

construction of tables, charts and plots in Microsoft Excel

Doing descriptive statistics, correlation and regression analysis in Excel.
using the formula function and the Analysis ToolPak.
 111 STA1506/1

STUDY UNIT 3
Report Writing
3.1 Introduction
Every human being on earth has two sides the good side and the evil side. If you are a good person
it means the good side rules and if you are a bad person it means the bad side rules. When a Pastor
or Priest is in front of a congregation, his or her aim is to ensure that people understand the works
of God and be able to use them in their society, that is, to be good citizens. Pastors all over the world
they can read the same script in the bible but they interpret it differently according to how each one
views the gospel. This means a text can be interpreted differently but saying the same message.
This shows that we might be given the same information but tend to interpret differently or use it
differently. In order for us to be able to use it similarly, our objectives have to be the same. Thus
objectives are very important when it comes to report writing. We need to have a purpose in mind of
doing something.

In order to be able to do a good research you should have an aim or objective of that research.
Report writing is a technique which one should learn. When doing a report you need to know who
are your recipients. Suppose you are given data on the COVID-19 and you want to tell the community
how deadly the disease is, then what would you do? You do not just write the report but rather you
need to now who this report is targeting. If you are targeting the poor community with a lot of people
not educated then the way you write the report should be made in such a way that they understand.
This is where you see people doing the reports tend to use a lot of diagrams and less technical
language. If you are writing the report to fellow statisticians then the style of writing changes. In all
this you can not relay information from data if you do not know how to interpret that information.

The purpose of this unit is to able to guide you on how you write your report and be able to interpret
your statistical findings in a way that your audience will understand.
 112

3.2 Objectives of the Unit

By the end of this unit, you should be able to:

understand what report writing is .

know the criteria for producing professional reports.

interpret the results.

present a managerial report.

3.3 Report Writing

If you witness a political event, then the next day when you read the local papers you will see the
different style of writing of the reporters regarding that event. Newspapers attract customers in the
way they write their stories. People want writers that are very good in writing and they can capture
the readers’ attention. When you are in the academic field, you can not have people writing stories
which are not supported by facts when it comes to statistics. You need to relay correct information
from your data and if you mislead people in your report writing then that’s unethical. This means
when you write your report you should be able to interpret your findings correctly. Then, what is
report writing?

Definition 3.1

Report writing is the process of creating an account or statement describing

events in detail, situations or occurrence usually as a result of observation
or inquiry ([Link])

Report writing is the culmination of the assessment process (Wiener & Costaris, 2012). There
are two most common forms of report writing which are news report writing and academic report
writing. Report writing is different from other forms of writing because it only includes facts, not the
opinion or judgement of the writer ([Link]). In addition the website said that report
writing is the creation of a structured document that precisely describes, and examines an event
or occurrence. According to Kalucy and Mcintyre (2004) when one is writing a report one should
consider the following global reporting initiative reporting principles.
 113 STA1506/1

Source: Kalucy and Mcintyre (2004:5)

Thus when writing a report according to Kalucy and Mcintyre (2004:4), the principles of transparency
and inclusiveness are the beginning point for reporting process and are entwined into the fabric of all
the other principles and all the decisions about reporting (e.g., how, when, what) should take these
two principles and associated practices into consideration.

3.4 Criteria of Producing a Professional Report

In order for one to be able to produce a professional report there are some guidelines you can follow.
A professional report is a report written in a professional manner following ethical considerations and
addressing a particular audience. Examples are reports done by academics.

Definition 3.2 =

Reports are factual documents which give accounts of events, processes,

methods or systems (Anigbogu & Ahumareze, 2017: 3)
 114

A professional report is a report written in a systematic way that delivers specific information about a
topic to a particular audience. According to Leedy and Ormrod (2015:349) the report should achieve
six main objectives:

~ It should give readers a clear understanding of the problem and why it merited an in-depth
investigation.

~ Methods used in an attempt to resolve the problem should be clearly described.

~ Data should be presented precisely and completely and the data presented in the report should
substantiate all the interpretations and conclusions that will follow.

~ The data should be interpreted in a way that readers will understand and one should demonstrate
either how the data resolve the problem or why they do not completely resolve the problem. A
report that merely presents raw data and uninterpreted facts (in the form of tables, graphs, and
other data-summary devices) is of little help to readers in deriving meaning from those data.

~ The report should alert readers to possible weaknesses of the study (e.g., what its delimitations
and limitations may have been, what assumptions and biases might have affected results and
interpretations).

~ The report should conclude by summarising the findings and connecting them to contexts beyond
the study itself like relating them to current theories about the topic or drawing implications for
future policies or practices.

The next section is going to show you how to interpret your findings and the last section will deal with
examples of a managerial report.

3.5 Interpretation of Results

One of the goals of a Statistician is to be able to interpret the information given after you have
analysed your data. Statistics are used everywhere like in sports, churches, financial institutions,
government and etc. The purpose of this section is to be able to teach you how to interpret your
statistics. From chapter two you now know how to use Microsoft Excel to come up with the statistics.
In our last chapter we are going to present outputs where we interpret the findings. Lets begin the
journey of interpreting the information in front of us. Lets the game begin towards the journey of
being Statisticians.
 115 STA1506/1

Problem A

A survey of 1 085 adults asked “Do you enjoy shopping for clothing for yourself?” The results indicated
that 51% of the females enjoyed shopping for themselves as compared to 44% of the males. Suppose
that the results were summarised in the following table:
ENJOY SHOPPING
FOR CLOTHING FOR GENDER
YOURSELF Male Female Total
Yes 238 276 514
No 304 267 571
Total 542 543 1085

(a) Construct contingency tables on total percentages, row percentages, and column percentages.

(b) What conclusions do you reach from these analyses?

(c) Construct a multiple bar chart of enjoying shopping and gender.

(d) What conclusions do you reach from this chart?

Solution A

(a) The contingency tables with total percentages, row percentages, and column percentages are
shown below.
Table of total percentages
ENJOY SHOPPING
FOR CLOTHING FOR GENDER
YOURSELF Male Female Total
Yes 21:94% 25:43% 47:37%
No 28:02% 24:61% 52:63%
Total 49:95% 50:05% 100:00%

Table of row percentages

ENJOY SHOPPING
FOR CLOTHING FOR GENDER
YOURSELF Male Female Total
Yes 46:30% 53:70% 100:00%
No 53:24% 46:76% 100:00%
Total 49:95% 50:05% 100:00%

Table of column percentages

ENJOY SHOPPING
FOR CLOTHING FOR GENDER
YOURSELF Male Female Total
Yes 43:91% 50:83% 47:37%
No 56:09% 49:17% 52:63%
Total 100:00% 100:00% 100:00%
 116

(b) A higher proportion of females tend to enjoy shopping for clothing for themselves as compared to
males.

(c) The multiple bar chart of enjoying shopping and gender is

Multiple Bar Chart of Enjoying Shopping and

Gender

Female

Male

0 50 100 150 200 250 300 350

No Yes

Figure 3.1: Multiple Bar Chart

(d) Females tend to enjoy shopping for clothing more than males.

Problem B

A supermarket sells kilogram bags of pears. The numbers of pears in 21 bags were:
7 9 8 8 10 9 8 10 10 8 9
10 7 9 9 9 7 8 7 8 9

(a) Construct a dot plot of the data.

(b) What can you say about the shape of the distribution?

(c) Find the mode, median and mean for these data.

(d) Interpret the three measures of central tendency?

(e) Compare your results and comment on the likely shape of the distribution.
 117 STA1506/1

(f) Plot a simple bar chart to portray the data and comment.

Solution B

(a) The dotplot of the data is shown below

Figure 3.2: Dotplot of Number of Pears

(b) The data is almost symmetrical.

(c) The measures of central tendency are found in Figure 3.3

Figure 3.3: Summary Statistics

 118

Thus the values of the mean, median and mode are 8:52, 9 and 9 respectively.

(d) Mean: On average the number of pears in a bag is approximately 9.

Median: About 50% of the bags have 9 pears or less or half of the bags contain up to 9 pears.

Mode: The most common number of pears in a bag is 9 pears.

(e) The data is almost symmetrical since the mean median = mode.

(f) The simple bar chart is

Figure 3.4: Bar Chart of Number of Pears

The dotplot and the bar chart are almost similar and in this case the data is almost symmetrical.
 119 STA1506/1

Problem C

(a) The owner of a restaurant that serves Continental-style entrees has the business objective of
learning more about the patterns of patron demand during Friday to Sunday weekend time period.
Data were collected from 630 customers on the type of entree ordered and are shown in the
following table.
Type of Entree Number Served
Beef 187
Chicken 103
Mixed 30
Duck 25
Fish 122
Pasta 63
Shellfish 74
Veal 26
Total 630

(i) Construct a percentage summary table for the types of entrees ordered.

(ii) Construct a bar chart and pie chart for the types of entrees ordered.

(iii) Do you prefer using a bar chart or pie chart of the data and why?

(iv) What conclusions can the restaurant owner reach concerning demand for the different types
of entree.

(b) Suppose that the owner of the restaurant wanted to study the demand for desert during the
same time period. She decided that in addition to studying whether a dessert was ordered, she
would also study the gender of the individual and whether a beef entree was ordered. Data were
collected from 630 customers and organised in the following contingency tables:

GENDER
DESERT ORDERED Male Female Total
Yes 50 96 146
No 250 234 484
Total 300 330 630

BEEF ENTREE
DESERT ORDERED Male Female Total
Yes 74 68 142
No 123 365 488
Total 197 433 630

(i) For each of the two contingency tables, construct contingency tables of row percentages,
column percentages, and total percentages.
 120

(ii) Which type of percentage (row, column, or total) do you think is most informative for each
gender? For beef entrée? Explain

(iii) What conclusions concerning the pattern of desert ordering can the restaurant owner reach?

Solution C

(a) (i) The percentage summary table is.

Type of Entree Number Served

Beef 29:68%
Chicken 16:35%
Mixed 4:76%
Duck 3:97%
Fish 19:37%
Pasta 10:00%
Shellfish 11:75%
Veal 4:13%
Total 100:00%

(ii) The bar chart for the types of entrees ordered is

Figure 3.5: Bar Chart of Type of Entree

 121 STA1506/1

The pie chart for the type of entree is

Figure 3.6: Pie Chart of Type of Entree

(iii) The pie chart since we are concerned about relative frequencies (proportions).

Note: The bar chart will be appropriate if we are concerned with frequencies.

(iv) Almost half of the weekend patrons of a continental restaurant prefer beef and fish which
account for nearly 50% of all entrees ordered. If chicken is included, then nearly two-thirds of
the entrees are accounted for. Veal and duck are the least ordered.

(b) (i) The contingency tables with total percentages, row percentages, and column percentages are
shown below.

For Gender:

Table of total percentages

GENDER
DESERT ORDERED Male Female Total
Yes 7:94% 15:24% 23:17%
No 39:68% 37:14% 76:83%
Total 47:62% 52:38% 100:00%
 122

Table of row percentages

GENDER
DESERT ORDERED Male Female Total
Yes 34:25% 65:75% 100:00%
No 51:65% 48:35% 100:00%
Total 47:62% 52:38% 100:00%
Table of column percentages
GENDER
DESERT ORDERED Male Female Total
Yes 16:67% 29:09% 23:17%
No 83:33% 70:91% 76:83%
Total 100:00% 100:00% 100:00%

For Beef Entree:

Table of total percentages
GENDER
DESERT ORDERED Male Female Total
Yes 11:75% 10:79% 22:54%
No 19:52% 57:94% 77:46%
Total 31:27% 68:73% 100:00%

Table of row percentages

GENDER
DESERT ORDERED Male Female Total
Yes 52:11% 47:89% 100:00%
No 25:20% 74:80% 100:00%
Total 31:27% 68:73% 100:00%
Table of column percentages
GENDER
DESERT ORDERED Male Female Total
Yes 37:56% 15:70% 22:54%
No 62:44% 84:30% 77:46%
Total 100:00% 100:00% 100:00%

(ii) The table of total percentages will be most informative if the owner is interested in finding out
the percentage of joint proportions of gender and ordering of dessert or the percentage of
joint occurrence of ordering a beef entrée and a dessert among all patrons. If the owner is
interested in the effect of gender on ordering of dessert or the effect of ordering a beef entrée
on the ordering of dessert, the table of column percentages will be most informative. Since
dessert will usually be ordered after the main entree and the owner has no direct control over
the gender of patrons, the table of row percentages is not very useful here.

(iii) Looking at the column percentages, about 16:67% of the men sampled ordered desserts
compared to 29:09% of the women. Women are almost twice as likely to order desserts as
men. About 37:56% of the patrons ordering a beef entree ordered dessert compared to less
than 15:70% of patrons ordering all other entrees. Patrons ordering beef are better than 2:3
 123 STA1506/1

times as likely to order dessert as patrons ordering any other entree.

Problem D

Bakers Ltd runs a chain of bakery shops and is famous for the quality of its pies. The management
of the company are concerned about the number of complaints from a particular branch about the
amount of time that it takes to serve customers. The motto of the company is "Have your pies in 2
minutes." The manager of the branch concerned has been told to provide data on the time it takes
for customers to enter the shop and be served by the staff. The following data was obtained.

0:37 1:18 0:51 1:93 1:83 1:33 0:22 0:97 0:91 1:08
0:19 1:83 0:66 1:14 0:78 1:38 0:89 1:57 1:56 0:37
1:21 0:71 0:81 1:22 0:65 0:86 1:08 1:53 1:14 0:44
0:59 0:89 1:08 1:05 2:33 0:88 1:64 0:85 0:83 1:20
1:35 0:52 1:10 0:28 1:00 1:93 0:96 0:63 0:39 2:07
1:08 0:77 0:55 1:14 1:90 0:63 1:16 1:52 0:63 1:33
0:70 1:39 1:31 0:62 0:41 0:48 1:11 0:93 1:42 0:80
0:31 1:28 0:86 0:06 0:54 0:85 0:09 1:07 1:54 1:49
1:12 0:48 0:67 0:97 1:34 1:15 1:64 0:36 0:72 1:38
1:24 1:23 1:70 0:56 1:47 0:37 1:09 0:43 0:33 1:78

(a) Construct a frequency table from these data.

(b) Use the frequency table from part (a) to construct the histogram.

(c) Describe the shape of the distribution?

(d) From the data set, calculate the mean, median and mode.

(e) Interpret the mean, median and mode.

(f) Calculate the standard deviation and interpret it.

(g) Calculate the quartiles (Q1 , Q2 and Q3 ) and the interquartile range and interpret each statistics.

(h) Estimate the 13th and 79th percentiles and interpret them.

(i) Do the results suggest that there was a large variation in the amount of time it took to serve
customers?

(j) Which measures would you recommend the shop manager uses to describe the variation in the
time taken to serve customers?
 124

(k) What conclusions can you draw from these results?

Solution D

(a) The frequency table is

Class Frequency Class Frequency

0:0 < x 0:5 17 (0:0; 0:5] 17
0:5 < x 1:0 34 (0:5; 1:0] 34
OR
1:0 < x 1:5 33 (1:0; 1:5] 33
1:5 < x 2:0 14 (1:5; 2:0] 14
2:0 < x 2:5 2 (2:0; 2:5] 2

(b) The histogram is shown below.

Figure 3.7: Histogram of Time Taken to Serve Customers

(c) Data is almost symmetrical; although slightly positively skewed.

 125 STA1506/1

(d) The descriptive statistics are

Figure 3.8: Summary Statistics

From the data set, the mean, median and mode are 0:9992, 0:985 and 1:08 respectively.

(e) Mean: On average the time taken to serve customers is approximately 0:9992 minutes.

Median: About 50% of the customers take up to 0:985 minutes to serve.

Mode: The most common time taken to serve customers is 1:08 minutes.

(f) Looking at Figure 3.8, the standard deviation is 0:4827. On average most of the observations
deviate 0:4827 from the mean. In this case 68:26% of the orders take between 0:5165 to 1:4819
minutes (x s) while 95:5% of the orders take between 0:0338 to 1:9646 minutes (x 2s)

(g) Using the Excel’s function key QUARTILE:

Lower Quartile, Q1 = 0:63 "=QUARTILE(data range,1)"

Second Quartile, Q2 = 0:985 "=QUARTILE(data range,2)"

Upper Quartile, Q3 = 1:33 "=QUARTILE(data range,3)"

The interquartile range (IQR) = 1:33 0:63 = 0:70.

Lower quartile: About 25% of the customers take up to 0:63 minutes to serve.

Second quartile, median: About 50% of the customers take up to 0:985 minutes to serve.
 126

Upper quartile: About 25% of the customers take at least 1:33 minutes to serve..

Interquartile range: The middle 50% of the customers take between 0:63 up to 1:33 minutes to
serve.

(h) Using the Excel’s function key - [Link]:

P13 = 0:4274 "=PERCENTILE(B2:B101,0.13)"

P79 = 1:38 "=PERCENTILE(B2:B101,0.79)"

P13 : About 13% of the customers take up to 0:4274 minutes to be served.

P79 : About 21% of the customers take at least 1:38 minutes to be served.

(i) No, data is almost symmetrical.

(j) I would recommend the standard deviation since data is almost symmetrical.

(k) From the frequency distribution only 2 orders out of the 100 are above 2 minutes. Most of the
statistics shows that the majority of the people are served within 2 minutes. The motto of the
company that "Have your pies in 2 minutes" is being fulfilled most of the time. The company is
performing within its motto.
 127 STA1506/1

Activity 3.1

1. The numbers of credit cards carried by 25 shoppers are

2 5 2 0 4 3 0 1 1 7 1 4 1
3 9 4 1 4 1 5 5 2 3 1 1
(a) Construct a dot plot of the data.

(b) Determine the mode and median of this distribution .

(c) Calculate the mean of the distribution and compare it to the mode and
median. What can you conclude about the shape of distribution?

(d) Draw a bar chart to represent the distribution and confirm your conclusions in (c):

2. Each day at a large hospital, several hundred laboratory tests are performed. The rate
at which these tests are done improperly (and therefore need to be redone) seems steady,
at about 4%. In an effort to get to the root cause of these nonconformance, tests that
need to be redone, the director of the lab decided to keep records over a period of one
week. The laboratory tests were subdivided by the shift of workers who performed the

lab tests, The results are as follows:

LAB TESTS SHIFT
PERFORMED Day Evening Total
Nonconforming 14 26 40
Conforming 656 304 960
Total 670 330 1000
(a) Construct contingency tables on total percentages, row percentages, and

column percentages.

(b) Which type of percentage; row, column or total do you think is most informative

from these data. Explain.

(c) Construct a compound bar chart of nonconformance and shift.

(d) What conclusions concerning the pattern of nonconforming laboratory tests can the

laboratory director reach?

.
 128

Activity 3.1 (cont’d)

3. Afrisoft supplies a range of computer software to 50 schools within Limpopo province.
When Afrisoft won the contract, the issue of customer service was considered to
be central to the company being successful at the final bidding stage. The
company has now requested that its customer service director creates a
series of graphical representations of the data to illustrate customer satisfaction
with the service. The data in the following table has been collected during the
past six months and measures the time to respond to complaints (days).
5 24 34 6 61 56 38 32
87 78 34 9 67 4 54 23
56 32 86 12 81 32 52 53
34 45 21 31 42 12 53 21
43 76 62 12 73 3 67 12
78 89 26 10 74 78 23 32
26 21 56 78 91 85 15 12
15 56 45 21 45 26 21 34
28 12 67 23 24 43 25 65
23 8 87 21 78 54 76 79
(a) Construct a frequency table from these data.

(b) Use the frequency table in part (a) to construct a histogram.

(c) Describe the shape of the distribution?

(d) From the data set, calculate the mean, median and mode.

(e) Interpret the mean, median and mode.

(f) Calculate the standard deviation and interpret it.

(g) Calculate the quartiles ( Q1 , Q2 and Q3 ), the interquartile range and interpret

each statistics.

(h) Estimate the 13th and 79th percentiles and interpret them.

(i) Do the results suggest that there is a great deal of variation in the time taken to

respond to customer complaints?

(j) Which measures would you recommend the customer service manager uses to

describe the variation in the time taken to respond to customer complaints?

(k) What conclusions can you draw from these results?

 129 STA1506/1

Problem E

The manager at BIG JIMS restaurant is concerned by the time it takes to process credit card
payments at the counter by staff. The manager has collected the processing time data (time in
minutes) in the table below and requested that summary statistics are calculated.
1:57 1:38 1:97 1:52 1:39
1:09 1:29 1:26 1:07 1:76
1:13 1:59 0:27 0:92 0:71
1:49 1:73 0:79 1:38 2:46
0:98 2:31 1:23 1:56 0:89
0:76 1:23 1:56 1:98 2:01
1:40 1:89 0:89 1:34 3:21
0:76 1:54 1:78 4:89 1:98

(a) Calculate a five-number summary for this data set.

(b) Is there any evidence for a symmetric distribution?.

(c) Construct a boxplot of the data.

(d) Use the Excel Analysis ToolPak to calculate descriptive statistics?

(e) Which measures would you use to provide a measure of average and spread?

Solution E

(a) Using the Excel’s function key QUARTILE, the five-number summary is:
Minimum = 0:27 "=QUARTILE(A2:A41,0)"

Lower Quartile, Q1 = 1:085 "=QUARTILE(A2:A41,1)"

Second Quartile, Q2 = 1:395 "=QUARTILE(A2:A41,2)"

Upper Quartile, Q3 = 1:765 "=QUARTILE(A2:A41,3)"

Maximum = 4:89 "=QUARTILE(A2:A41,4)"

(b) The distances of the four quarters are:

Minimum to Q1 = 1:085 0:27 = 0:815

Q1 to Q2 = 1:395 1:085 = 0:31

Q2 to Q3 = 1:765 1:395 = 0:37

Q3 to Maximum = 4:89 1:765 = 3:125

 130

The distance from median to Q3 is approximately equal to distance from Q1 to median. Distance
from median to largest value is greater than distance from minimum to Q1 indicating that the data
is positively skewed.

(c) The box plot is

Figure 3.9: Box Plot of Time Taken to Process Credit Card Payments

(d) The descriptive statistics are:

Figure 3.10: Summary Statistics

 131 STA1506/1

(e) The data is positively skewed due to the two outliers on the right. When data is skewed the median
is the best measure of central tendency since its not affected by outliers. When median is used as
a measure of central tendency, then for dispersion or spread we use the semi-interquartile range
IQR
:
2

Problem F

The number of bicycles sold monthly by a bicycle dealer was:

25 18 30 18 20 19 30 16 36 24

(a) Find the mean and median number of bicycles sold monthly. Interpret each descriptive statistics
measure.

(b) Find the range, variance and the standard deviation of the number of bicycles sold monthly.
Interpret the range and standard deviation measures.

(c) Calculate the lower and upper quartiles of monthly bicycle sales. Interpret.

(d) Construct a box plot of monthly bicycle sales. Interpret the plot.

(e) If the dealer uses the formula ‘mean plus one standard deviation’ to decide on the opening stock
level of bicycles at the beginning of next month, will he run out of stock during the month if he
receives orders for 30 bicycles next month? Assume no extra bicycles can be ordered.

Solution F

(a) The mean and median are

Mean = 23:6 "=AVERAGE(A2:A11)"

Median, Q2 = 22 "=QUARTILE(A2:A11,2)"

Mean: On average, approximately 24 (23:6) bicycles are sold each month.

Median: For half of the months (i.e., 5 months), bicycle sales were not more than 22
(20 + 0:5 (24 20)) bicycles per month.
 132

(b) The range, variance and standard deviations are

Minimum = 16 "=QUARTILE(A2:A11,0)"

Maximum = 36 "=QUARTILE(A2:A11,4)"

Range = 20 36 16 = 20

Variance = 43:6 "=VAR(A2:A11)"

Standard deviation = 6:6030 "=STDEV(A2:A11)"

Range: The range of sales between the worst and best months was 20 bicycles.

Standard deviation: About 68:26% of all monthly bicycles are likely to lie between 17 and 30:2:
(x s = 23:6 6:6).

(c) The lower and upper quartiles are

Lower Quartile, Q1 = 18:25 "=QUARTILE(A2:A11,1)"

Upper Quartile, Q3 = 28:75 "=QUARTILE(A2:A11,3)"

Lower quartile: About 25% of monthly bicycles sales were less than or equal to 18:25 bicycles or
no more than 18 bicycles per month were sold in 25% of the months.

Upper quartile: About 25% of monthly bicycles sales were above 28:75 or more than 29 (28:75)
bicycles per month were sold in 25% of the months.

(d) The box plot is

Figure 3.11: Box Plot of Bicycle Sales

 133 STA1506/1

Monthly bicycle sales range between 16 and 36. The median monthly sales was 22. There is a
longer tail to the right suggesting that data is positively skewed. The positive skewness shows a
wider spread of monthly sales toward the months of high sales.

(e) Opening monthly stock level = 23:6 6:6 = 30:2 bicycles in stock. If orders = 30, then the dealer
will have sufficient bicycle stock to meet demand.

Problem G

A restaurant owner randomly selected and recorded the value of meals enjoyed by 20 diners on a
given day. The values of meals (in rand) were:
144 165 180 172 190 158 144 147 148 135
165 156 136 169 148 162 151 155 150 144

(a) Define the random variable and its data type.

(b) Compute the mean and standard deviation of the value of meals at the restaurant and interpret
them.

(c) What is the median value of a meal at the restaurant? Interpret its meaning?

(d) What meal value occurs most frequently and interpret it?

(e) Which central location measure would you choose? Why?

Solution G

(a) The random variable is value of a restaurant meal (in Rands) and the data type is quantitative,
continuous and ratio-scaled.
 134

(b) The summary statistics are

Figure 3.12: Summary Statistics

The mean and standard deviation are R155:95 and R14:33 respectively.

Mean: On average, the value of a restaurant meal is R155:95.

Standard deviation: About 68:26% of the meal values are likely to lie between R141:62 and
R170:28: (x s = 155:95 14:33).

(c) The median value is R153.

Median: Half of the meals were valued at R153 or less.

(d) The value that occurs most frequently, that is the modal value is R144.

Mode: The most common value of a restaurant meal is R144.

(e) There is moderate skewness caused by two high meal values (i.e., R180 and R190). We would
recommend selecting the median as the most representative central location measure since it is
not affected by outliers.
 135 STA1506/1

Problem H

A company that manufacturers wooden products (e.g. garden furniture, ladders, benches) regularly
maintains its lathe machines, which are used for cutting and shaping components. The manager
would like to know whether the cost of machine maintenance is related to the age of the machines.
For a random sample of 12 lathe machines in the company’s factory, the annual maintenance cost
(in R100s) and age of each machine was recorded.

Maintenance costs analysis

Machine Age (yrs) Annual cost (R)
1 4 45
2 3 20
3 3 38
4 8 65
5 6 58
6 7 50
7 1 16
8 1 22
9 5 38
10 2 26
11 4 30
12 6 35

(a) Identify the independent variable and the dependent variable.

(b) Show the data graphically in a scatter plot. What relationship is observed?.

(c) Calculate the correlation coefficient between the age of lathe machines and their annual
maintenance costs. Comment on the strength of the association.

(d) Use the method of least squares to find the best fitting line between the age of lathe machines
and their annual maintenance costs.

(e) Interpret the meaning of the regression coefficient (b1 ) of the independent variable.

(f) What is the expected average maintenance cost of a lathe machine that is five years old.

Solution H

(a) The independent variable is age (in years) and the dependent variable is annual cost (in
Rands).
 136

(b) The scatter plot is shown below.

Figure 3.13: Scatter Plot of Annual Maintenance Costs Vs Age of Machines

There is a strong positive linear relationship between the age of machines and annual
maintenance cost. Aged machines tend to have high maintenance cost.

(c) The correlation coefficient is

Figure 3.14: Correlation Matrix

There is a very strong positive correlation or association between the age of machines and their
annual maintenance cost (in Rands).
 137 STA1506/1

(d) The regression output is

Figure 3.15: Simple Linear Regression Model of Annual Cost vs Age of Machines

[ = 12:6272 + 5:8295Age.
The least squares regression line is Cost

(e) b1 = 5:8295: For every increase in age of 1 year, the annual maintenance cost increase by
R582:95 or each year, the annual maintenance cost of machines increase by R582:95

Note: Costs is in R100s.

[ = 12:6272 + 5:8295 (5)

(f) Age = 5 years. Cost 41:7747: For a five year old machine, the annual
maintenance cost are expected to be R4 177:47:

Problem I

The price of a veal cordon bleu meal (in rand) was taken from the menus of 28 Durban restaurants
in a survey conducted by Lifestyle magazine into the cost of ‘dining out’. The prices are:
148 166 160 190 158 168 153 163 164 155 164 158 154 172
156 180 155 162 175 148 155 145 148 172 152 168 156 170

Use Excel’s Data > Data Analysis > Descriptive Statistics option and, where necessary, the
function keys QUARTILE and PERCENTILE, to answer the following questions:
 138

(a) Define the random variable and its data type.

(b) Find the mean and median price of a veal cordon bleu meal. Interpret each measure.

(c) Can you identify a modal price? Give its value and discuss its usefulness.

(d) Identify the value of the standard deviation of the price of veal cordon bleu and interpret it.

(e) Does the data appear to be skewed? If so, why?

(f) Which central location measure would you choose to report in the article on ’dining out’? Why?

(g) What is the least price that a patron to one of these restaurants would pay if they dined out at any
one of the most expensive 25% of restaurants?

(h) The least expensive 25% of restaurants do not charge above what price for the veal cordon bleu
meal?

(i) What is the least price to be paid for the most expensive 10% of veal cordon bleu meals?

Solution I

(a) The random variable is cost of a veal cordon bleu meal at a Durban restaurant and the data type
is quantitative, continuous and ratio-scaled.

(b) Using the Excel’s Descriptive Statistics option in Data Analysis the summary statistics are

Figure 3.16: Summary Statistics

 139 STA1506/1

The mean and median are R161:25 and R159 respectively.

Mean: On average, the value of a veal cordon blue meal at a Durban restaurant is R161:25 or on
average, a patron expects to pay R161:25 of a veal cordon blue meal at a Durban restaurant.

Median: For half of the Durban restaurants, a veal cordon bleu meal costs not more than R159 or
50% of Durban restaurants charge no more than R159 for a veal cordon bleu meal.

(c) Yes, however they are two. When there are more than one modal values, Excel tends to take
the smallest one, that is, the first occurrence. In this case there are two modal values (R148 and
R155), since both occur with a frequency of 3. It is a misleading value because of its low frequency
of occurrence.

Note: When interpreting its better to chose the one closer to the other measures of central
tendency, in this case R155.

(d) Standard deviation = R10:54. About 68:26% of the Durban restaurants are likely to charge
between R150:71 and R171:79: (x s = 161:25 10:54) for a veal cordon bleu meal.

(e) Mean > Median > Mode. The data is positively skewed. Two of the restaurants tend to be
charging high prices (i.e., R180 and R190) for a veal cordon bleu meal.

(f) The data is positively skewed due to the two restaurants charging high prices. When data is
skewed the median cost of R159 is the best measure of central tendency or location since its not
affected by outliers.

(g) The most expensive 25% of the restaurants is the upper quartile. Using the Excel’s function key
QUARTILE, then the upper quartile is:

Upper Quartile, Q3 = 168 "=QUARTILE(A2:A29,3)"

About 25% of the Durban restaurants charge at least R168 for a veal cordon bleu meal.

(h) The least expensive 25% of the restaurants is the lower quartile. Using the Excel’s function key
QUARTILE, then the lower quartile is:

Lower Quartile, Q1 = 154:75 "=QUARTILE(A2:A29,3)"

About 25% of the Durban restaurants charge not more than R154:75 for a veal cordon bleu meal.
 140

(i) The most expensive 10% of the restaurants is the 90th percentile. Using the Excel’s function key
PERCENTILE, then the 90th percentile is:

90th percentile = 172:9 "=PERCENTILE(A2:A29,0.9)"

About 10% of the Durban restaurants charge at least R172:9 for a veal cordon bleu meal.

Problem J

Bakers Ltd is concerned about the possible relationship between the amount of fat (grams) and the
number of calories in a popular pie:

Amount of Amount of
Pie ID Calories Pie ID Calories
Fat(g) Fat(g)
1 19 410 16 33 597
2 31 580 17 31 583
3 34 590 18 37 589
4 35 570 19 39 640
5 39 640 20 23 456
6 39 680 21 43 660
7 43 660 22 22 448
8 22 465 23 30 577
9 28 567 24 34 594
10 38 610 25 35 590
11 35 576 26 41 638
12 22 434 27 34 560
13 40 690 28 43 660
14 43 660 29 45 680
15 21 435 30 29 587

(a) Plot a scatter plot and comment on a possible relationship between calories and the amount of fat
in the pies.

(b) Use Excel Analysis ToolPak to undertake the following tasks.

(i) State the least squares regression model equation.

(ii) Interpret the meaning of the regression coefficient (b1 ).

(iii) Comment on model reliability (r and R2 ).

 141 STA1506/1

Solution J

(a) The scatter plot is shown below.

Figure 3.17: Scatter Plot of Calories vs Amount of Fat

There is a strong positive linear relationship between the amount of fat and calories. Pies with high
fat content tend to be associated with more calories. Thus a linear regression model is possible
between the two variables.

(b) The regression output is

Figure 3.18: Simple Linear Regression of Calories vs Amount of Fat

 142

\ = 242:2248 + 10:0786Amount of fat.

(i) The least squares regression line is Calories

(ii) b1 = 10:0786: For every increase in amount of fat of one gram, the calories in a pie increases
by 10:08.

(iii) The correlation matrix is

Figure 3.19: Correlation Matrix

r = 0:9504 and R2 = 0:9033 (from regression output).

There is a very strong positive (direct) correlation or association of 0:9504 between the amount
of fat and calories.

About 90:33% of the variability in calories is being accounted for by the amount of fat or by the
model. This indicates a very good fit.

Problem K

The number of new franchise dealers recorded over 10 periods by the Franchise Association of South
Africa is shown below. .
Period 1 2 3 4 5 6 7 8 9 10
New dealers 28 32 43 31 38 47 40 45 55 42

(a) Draw a time series graph to represent this data and comment.

(b) Calculate the trend line yb = b0 + b1 x and interpret the slope.

(c) Calculate the trend estimates of the number of new franchise dealers in periods 11, 12 and 13.
 143 STA1506/1

Solution K

(a) The time series plot is

Figure 3.20: Time Series Plot of Number of New Franchise

Dealers

There is an increasing trend (secular trend).

(b) The trend output is.

Figure 3.21: The Trendline of New Franchise Dealers

 144

The trend line yb = 29 + 2:0182x; x = 1; 2; 3; : : : ; 10.

b1 = 2:0182: Every year the number of new franchise dealers increases by 2.

(c) The trend line estimates for periods 11, 12 and 13 are:

Period 11 yb = 29 + 2:0182 (11) = 51:2002 51

Period 12 yb = 29 + 2:0182 (12) = 53:2184 53

Period 13 yb = 29 + 2:0182 (13) = 55:2366 55

Thus the estimates are 51, 53 and 55 respectively.

Problem L

Consider the following quarterly demand levels of electricity (in 1000 megawatts) in Cape Town from
2016 to 2019.
Months 2016 2017 2018 2019
Jan-Mar 21 35 39 78
Apr-Jun 42 54 82 114
Jul-Sep 60 91 136 160
Oct-Dec 12 14 28 40

(a) Plot the time series of quarterly electricity demand graphically.

(b) Find the least squares trend line for quarterly electricity demand in Cape Town and interpret the
slope.

(c) Estimate demand for 2020 all quarters.

 145 STA1506/1

Solution L

(a) The time series plot is

Time Series Plot of Electricity Demand for Cape

Town
180
160
140
120
100
80
60
40
20
0
Electricity Demand (1000 MW)
0 2 4 6 8 10 12 14 16 18
Quarters (2016 Q1 - 2019 Quarter 4)

Figure 3.22: Time Series Plot of Electricity Demand for Cape Town

Electricity demand in Cape Town depicts a seasonal variation. Demand peaks in quarter 3 and
lowest in quarter 4.

(b) The trend output is.

Figure 3.23: The Trendline of Electricity Demand

 146

The trend line yb = 20:8 + 4:95x; x = 1; 2; 3; : : : ; 16.

b1 = 4:95: Every quarter, electricity demand in Cape Town increases by 4 950 Megawatts:

(c) The trend line estimates for 2020 all quarters are 17, 18; 19 and 20 are:

Period 17 yb = 20:8 + 4:95 (17) = 104:95

Period 18 yb = 20:8 + 4:95 (18) = 109:9

Period 19 yb = 20:8 + 4:95 (19) = 114:85

Period 20 yb = 20:8 + 4:95 (20) = 119:8

Thus the electricity demand estimates for 2020 quarter 1, 2, 3 and 4 are 104 950M W , 109 900M W ,
114 850M W and 119 800M W respectively.
 147 STA1506/1

Activity 3.2
1. A supermarket has one checkout for customers who wish to purchase 10 items or less.

The numbers of items presented at this checkout by 19 customers were:

10 8 7 7 6 11 10 8 9 9
9 6 10 9 8 9 10 10 10
(a) Find the mode, median and mean for these data..

(b) Interpret the three measures of central tendency computed in (a).

(c) What do your results for (a) tell you about the shape of the distribution?
(d) Calculate the range and standard deviation of the data and interpret them.

(e) Draw a box plot of the data and confirm your conclusions in (c):

2. The human resources department of a company recorded the number of days

absent of 23 employees in the technical department over the past nine months:
5 4 8 17 10 9 30 5 6 15 10 9
2 16 15 18 4 12 6 6 15 10 5
(a) Find the mean, median and modal number of days absent over this nine-month

period. Interpret each central location measure.

(b) Compute the first quartile and the third quartile of the number of days absent.

Interpret these quartile values for the human resources manager.

(c) The company’s policy is to keep its absenteeism level to within an average of
one day per employee per month. Based on the findings in (a), is the company
successful in managing its absenteeism level? Explain.

3. A manufacturing company produces electric insulators. If the insulators break

when in use, a short circuit is likely to occur. To test the strength of the
insulators, destructive testing is carried out to determine how much force
is required to break the insulators. Force is measured by observing how
many pounds must be applied to an insulator before it breaks. Data are
collected from a sample of 30 insulators. The strength are as follows:
1870 1728 1656 1610 1634 1784 1522 1696 1592 1662
1866 1764 1734 1662 1734 1774 1550 1756 1762 1866
1820 1744 1788 1688 1810 1752 1680 1810 1652 1736
.
 148

Activity 3.2 (cont’d)

(a) Compute the mean, median, mode, range and standard deviation for the
force needed to break the insulator.

(b) Interpret the measures of central tendency and variability computed in (a).

(c) Comment on the usefulness of the mode.

(d) Construct a box plot and describe its shape.

(e) What can you conclude about the strength of the insulator if the company

requires a force of at least 1500 pounds before breakage?

4. The Gauteng chamber of business conducted a survey amongst 17 furniture

retailers to identify the percentage of bad debts in each company’s debtors’

book. The bad debts percentage are as follows:

2:2 4:7 6:3 5:8 5:7 7:2 2:6 2:4 6:1
6:8 2:2 5:7 3:4 6:6 1:8 4:4 5:4
(a) Find the average and standard deviation of the percentage of bad debts
amongst the 17 furniture retailers surveyed.

(b) Find the median percentage of bad debts amongst the 17 furniture

retailers surveyed.

(c) Interpret the findings from (a) and (b).

(d) Is there a modal percentage of bad debts? If so, identify it and comment

on its usefulness.
(e) Compute the first quartile and the third quartile of the percentage of bad

debts amongst the furniture retailers surveyed. Interpret these

quartile values.

(f) Draw a box plot of the data and comment on the shape of the distribution.
(g) The chamber of business monitors bad debt levels and will advise an
industry to take corrective action if the percentage of bad debts, on
average, exceeds 5%. Should the chamber of business send out an
advisory note to all furniture retailers based on these sample findings?

Justify your answer.

.
 149 STA1506/1

Activity 3.2 (cont’d)

5. The following data gives the assignment mark and examination mark of 14
undergraduate statistics students.
Examination 77 66 65 65 80 71 78 75 70 60 67 61 59 58
Assignment 69 42 43 40 100 80 100 90 77 47 68 50 45 41
(a) Determine the independent variable and dependent variable.

(b) Plot a scatter plot and comment on a possible relationship .

between the examination and assignment marks.

(c) Calculate the correlation coefficient.

(d) Use the Excel regression function to do the following tasks.
(i) Fit a simple linear regression model.
(ii) Comment on model reliability (r and R2 ).
(iii) Interpret the slope of the regression line (b1 ).
(e) Predict the examination mark for a student with an assignment

mark in statistics of 75.

(f) As a fellow student what advice can you give to the students

regarding assignments.

6. Opinion polls are often criticised for their lack of predictive validity (i.e. the ability
to reliably estimate the actual election result). In a recent election in each of 11
regions, the percentage of votes predicted by opinion polls for the winning
political party are recorded together with the actual percentage of votes received.
Region 1 2 3 4 5 6 7 8 9 10 11
Opinion poll (%) 42 34 59 41 53 40 65 48 59 38 62
Actual election (%) 51 31 56 49 68 35 54 52 54 43 60
(a) Determine the degree of association between the opinion poll results and the

results of the actual election by calculating Pearson’s correlation coefficient.

(b) Set up a least squares regression equation to estimate actual election

results based on opinion poll results.

(c) Calculate the coefficient of determination for this regression equation.

Interpret its value.

(d) If an opinion poll showed 58% support for the winning party, what is

the actual election result likely to be?

(e) If an opinion poll showed 82% support for the winning party, what is the actual
election result likely to be? Is this a valid and reliable result? Comment.
 150

Activity 3.2 (cont’d)

7. The following data represent the yearly movie attendance (in billions) from
2001 to 2013.
Year 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Attendance 1:44 1:60 1:52 1:48 1:38 1:40 1:40 1:36 1:42 1:35
Year 2011 2012 2013
Attendance 1:28 1:36 1:15
(a) Construct a time series plot for the movie attendance (in billions)

(b) What patterns, if any, is present in the data.

(c) Calculate the trend line yb = b0 + b1 x and interpret the slope.

(d) Calculate the trend estimates of the attendance for the years 2014
and 2015:

8. A hotel’s monthly occupancy rate (measured as a percentage of rooms

available) is reported as follows for a 10-month period.
Month Sep Oct Nov Dec Jan Feb Mar Apr May Jun
Occupancy (%) 74 82 70 90 88 74 64 69 58 65
(a) Produce a line graph of the hotel’s occupancy rate per month.

(b) Fit a least squares trend line to the hotel occupancy rate data.

(c) What is the trend estimate of the hotel’s occupancy rate for July
and August. Comment on your findings.
.
 151 STA1506/1

3.6 Managerial Report

You have been analysing and interpreting data in bits and pieces. Now you need to know how you
report information in a coherent manner. The purpose of this section is to enable you to write a
managerial report. Three scenarios will be given and you will be shown how a managerial report can
be presented. In this case the statistical interpretation, conclusions and recommendations are the
ones that will be presented.

Scenario A:

CardioGood Fitness:

The market research team at AdRight is assigned the task to identify the profile of the typical
customer for each treadmill product offered by CardioGood Fitness. The market research team
decides to investigate whether there are differences across the product lines with respect to
customer characteristics. The team decides to collect data on individuals who purchased a treadmill
at a CardioGoodFitness retail store during the prior three months. The data are stored in the
CardioGoodFitness file on myunisa under the folder "Study Guide Data Sets". The team identifies
the following customer variables to study:

- Product purchased (TM195, TM498, TM798)

- Gender (Male, Female)

- Education (in years)

- Marital status (Single, Partnered)

- Usage (average number of times the customer plans to use the treadmill each week)

- Fitness (self-rated fitness on an Likert scale ranging from 1 (poor shape) to 5 (excellent shape))

- Income (Monthly household net income (R10s)

- Miles (Average number of miles the customer expects to walk/run each week)
 152

The first fifteen cases are shown below.

You need to present a report to the management of CardioGood Fitness which includes:

(a) Creating a customer profile for each CardioGood Fitness treadmill product line by developing
appropriate tables and charts.

(b) Determine whether product purchased has a relationship with gender by looking at the total
percentages, row percentages and column percentages.

(c) Compute descriptive statistics to create a customer profile for each CardioGood Fitness treadmill
product line.

(d) Determine the relationship between age, education, usage, fitness, income and miles and
comment on the relationships.

(e) Use the method of least squares to determine whether fitness can be predicted by:

(i) usage and

(ii) miles the customer expects to walk/run in a weak.

In each case also present the scatter plots, the reliability of the model (r and R2 ) and where
reliable interpret the slope of the regression model.

Ensure that the report is given a title and all charts and tables are labelled and detailed findings
are presented and recommendations are made to the management of CardioGood Fitness.
 153 STA1506/1

Solution to Scenario A

Managerial Report CardioGood Fitness Customer Profile:

A total of 180 customers managed to indicate the treadmill product purchased, age, gender,
education level, marital status, usage (average number of times the customer plans to use the
treadmill each week), fitness level (self-rated), annual household income and miles done on the
treadmill. The frequency distributions of the categorical variables are shown in Figure 3.24.

Figure 3.24: Descriptive Statistics

About 44:44% (n = 80) of the customers purchased the TM195 treadmill, 33:33% (n = 60) purchased
the TM498 treadmill and 22:22% (n = 40) purchased the TM798 treadmill. In terms of gender, about
42:22% (n = 76) were females and 57:78% (n = 104) were males. This means that the majority of
the customers were males. Close to 60%, that is, 59:44% (n = 107) were partnered while 40:56%
(n = 73) were [Link] customer profile by gender is shown in Figure 3.25.

Table of Frequency Distribution of Product by Gender

Figure 3.25a: Crosstabulation of Product by Gender

 154

Table of Total Percentages Product by Gender

Table of Row Percentages of Product by Gender

Table of Column Percentages of Product by Gender

Figure 3.25b: Crosstabulation of Product by Gender

Looking at the row percentages, for treadmill TM195, equal proportion of males and females, that
is, 50% prefer it. Almost the same proportion prefer TM498, that is, about 48:33% of the customers
who purchased the TM498 treadmill are females as compared to 51:67% of males. This means that
males and females prefer the treadmills TM195 or TM498 almost equally. For the treadmill TM798,
about 17:50% of the females prefer it as compared to 82:50% males. Men are almost five times as
likely to purchase the TM798 treadmill than women.

Looking at the column percentages,about 52:63% of the females purchased the TM195, 38:16%
purchased the TM498 and 9:21% purchased the TM798. Thus for females the majority prefers the
TM195. For males about 38:46% purchased the TM195, 29:41% purchased the TM498 and 31:73%
purchased the TM798. The males seems to prefer the TM195 more and the TM498 and TM798
seem to be preferred equally.
 155 STA1506/1

The multiple bar chart is shown in Figure 3.26.

Figure 3.26: Multiple Bar Chart of Product by Gender

Thus, the TM195 is preferred by majority of the females and TM798 is not a popular treadmill for
females. For males generally the treadmills are most preferred evenly although TM198 is preferred
the most.

The customer profile by marital status is shown in Figure 3.27.

Table of Frequency Distribution of Product by Marital Status

Table of Total Percentages Product by Marital Status

Figure 3.27a: Crosstabulation of Product by Marital Status

 156

Table of Row Percentages of Product by Marital Status

Table of Column Percentages of Product by Marital Status

Figure 3.27b: Crosstabulation of Product by Marital Status

Looking at the row percentages, 60% of the customers who purchased the treadmill TM195 are
partnered while 40% are single. The same pattern is observed for TM498. For TM798, 57:50% of the
customers who purchased it are partnered as compared to 42:50% single. The ratio of the preference
of the treadmills by marital status for partnered and single are in the ratio 3 : 2.

Looking at the column percentages, for those with partners, about, 44:86% purchased TM195, 33:64%
purchased the TM498 and 21:50% purchased the TM798. Thus for partnered customer, the largest
proportion prefers the TM195. For those single 43:84% purchased the TM195, 32:84% purchased
the TM498 and 23:29% purchased the TM798. For those single, TM195 is preferred more. The
preference of treadmills by marital status seem to be the same across. The ratio of preferring the
treadmills are the same for the partnered and the single with both groups preferring the TM195 more.
 157 STA1506/1

The multiple bar chart is shown in Figure 3.28.

Figure 3.28: Multiple Bar Chart of Product by Marital Status

Looking at Figure 3.28, it can be noted that those partnered and those single seem to prefer the
treadmill the same as evidenced by the bars that are almost equal. Both groups prefer TM195 more.
However the ratio of proportion of preference across treadmills are almost the same for partnered
and single.

Box plot were done to determine variability of treadmill preference by age, education, usage, fitness,
income and miles. The boxplot of product by age is shown below.

Figure 3.29: Boxplot of Age by Product

The medians for all the treadmills seem to be the same. It can be observed that at least 50% of the
customers are more than 28 years of age. TM798 tend to be preferred mostly by people below 30
 158

years of age. All boxplots depict that the distribution of product by age depict distributions that are
positively skewed, that is few older people are utilising these products.

The boxplot of product by education is shown below.

Figure 3.30: Boxplot of Education by Product

In terms of education level, preference of TM195 and TM498 are the same. The boxplot are almost
similar. The majority of the people preferring them has an education level between 14 years and 16
years. The majority of the customers who prefer TM798 are those with education level between 16
and 19 years. Thus TM798 seem to be preferred mostly by people more educated. The education
distributions for TM195 and TM498 are almost symmetrical while that for TM798 has some positive
skewness.

The boxplot of product by usage is shown below.

Figure 3.31: Boxplot of Usage by Product

 159 STA1506/1

The majority of the customers indicated that the average number of times the customers plan to use
the TM195 treadmill and the TM498 treadmill each week are concentrated between 3 and 4 times.
For the TM798 treadmill, the majority of the people indicated that they can use it between 4 and
5 times a week. Those with TM798 prefer to use it more per week than the other treadmill. All
treadmills have distributions that are positively skewed with few customers indicating that they would
like to use them at least 5 times a week for TM195 and TM498 and at least six times a week for
TM798.

The boxplot of product by fitness is shown below.

Figure 3.32: Boxplot of Fitness by Product

The self-rated fitness of customers who purchased the treadmills TM195 or TM498 is concentrated
around 3 while for the Treadmill TM798 is concentrated around 5. Thus those who use TM798 seem
to indicate high levels of fitness.

The boxplot of product by income is shown below.

Figure 3.33: Boxplot of Income by Product

 160

Most of the customers who purchased TM195 or TM195 seem to have an annual income not more
than R50 000 while those who purchased the TM798 seem to be having an income of at least R75 000.
Those with high income are the ones who purchased the TM798. The distribution of income for
TM195 has positive skewness. TM498 has almost a symmetrical distribution and TM798 has a
distribution that is negatively skewed.

The boxplot of product by miles is shown below.

Figure 3.34: Boxplot of Miles by Product

All box plots depicts positive skewness with outliers to the right. The majority of the customers who
use the treadmills TM195 or TM498 expects to run not more than 85 miles while those who use the
TM798 expects to run at least 160 miles Those who use the TM798 expects to run more lines as
compared to the others.
 161 STA1506/1

A detailed customer profile by treadmill is as follows:

TM195:

The TM195 customer profile of age is shown below:

Figure 3.35: Stem and Leaf Plot of Age for TM195

Figure 3.36: Descriptive Statistics of Age for TM195

The ages of the purchasers of TM195 is concentrated around 21 29 years. The youngest is 18
years old while the oldest is 50 years old. The middle 50% of the customers fall between 23 (lower
quartile) and 33 years (upper quartile). The mean and standard deviation are 28:55 and 7:22 years
 162

respectively. On average, the age of those who purchases TM195 is 28:55 years. The median and
the modal values are 26 and 23 years. Thus, half of the customers who purchased TM195 are not
more than 26 years old and the most common age for those purchasing a TM195 is 23 years. About
68:26% of the customers who purchases the TM195 have ages that range from 21:33 and 35:77 years
(x s = 28:55 7:22). The coefficient of variation was 25:29% which is not that far apart from 0% (no
variability). Thus, the ratio of standard deviation to mean is almost 1 : 4. Age is right-skewed, and
slightly more peaked and thicker tailed than a normal distribution.

The TM195 customer profile of education is shown below:

Figure 3.37: Stem and Leaf Plot of Education for TM195

Figure 3.38: Descriptive Statistics of Education for TM195

The education in years of TM195 customers is concentrated around 15 years. The number of years
of education of TM195 customers has a mean of 15:04 years. Thus, on average, the education of
 163 STA1506/1

TM195 customers is 15:04 years. The median and the modal values are both 16 years. Thus, half
of the TM195 customers having not more than 16 years of education and the most common time
taken by customers to do their education is 16 years. The mean spread around the mean (standard
deviation) is 1:22 with the lowest value of 12 years and the highest value of 18 years. About 68:26%
of the customers who purchases the TM195 have undertaken between 14:82 and 16:26 years in
education (x s = 15:04 1:22). The coefficient of variation was 8:09% indicating that there was not
much variability in educational levels and the ratio of the standard deviation to the mean is 1 : 12.
The years in education is almost symmetrical, and only slightly less peaked and thinner tailed than a
normal distribution.

The TM195 customer profile of usage is shown below:

Figure 3.39: Stem and Leaf Plot of Usage for TM195

Figure 3.40: Descriptive Statistics of Usage for TM195

The average number of times the TM195 customers plans to use the treadmill each week is
concentrated around 3. The mean, median and mode are 3:09, 3 and 3 respectively. On average the
 164

TM195 customers expect to use the treadmill 3 times a week, half of the customers have a usage
at most 3 and the most common time customers expect to use the treadmill is 3 times a week. The
lowest value was 2 and the highest value was 5. The middle 50% of the usage falls between 3
and 4. The standard deviation, that is, the mean spread around the mean fitness was 0:78 giving
a coefficient of variation of 25:35% indicating that there was not much variability in usage and the
ratio of the standard deviation to the mean is 1 : 4:. About 68:26% of the customers who purchases
the TM195 expect to use the treadmill from 2:22 to 3:87 a week, that is expect to use the treadmill
TM195 2 to 4 times a week. The usage value is almost symmetrical, and only slightly less peaked
and thinner tailed than a normal distribution.

The TM195 customer profile of fitness is shown below:

Figure 3.41: Stem and Leaf Plot of Fitness for TM195

Figure 3.42: Descriptive Statistics of Fitness for TM195

The self-rated fitness of customers who purchased the TM195 is concentrated around 3. The mean,
median and mode are 2:96, 3 and 3 respectively. On average the TM195 customers gave a self-rating
 165 STA1506/1

fitness of almost 3, half of the customers gave a self rating of at most 3 and 3 was the most common
self-rated fitness. The mean spread around the mean was 0:66 with the lowest value of 1 and the
highest value of 5. The middle 50% of the fitness value equals to 3. The coefficient of variation was
22:43% indicating that there was not much variability in self-rated fitness and the ratio of the standard
deviation to the mean is 2 : 9:.About 68:26% of the customers who purchases the TM195 have self-
rated fitness that range from 2:30 to 3:62, that is, will give a self rating fitness that range from 2 to 4.
The fitness value is almost symmetrical, and more peaked and thicker than a normal distribution.

The TM195 customer profile of income is shown below:

Figure 3.43: Histogram of Income form TM195

Figure 3.44: Descriptive Statistics of Income for TM195

 166

The monthly household income of TM195 customers is concentrated around R47 500. The mean
and standard deviation are R46 418:03 and R9 075:78 respectively. On average, the monthly income
of TM195 customers is R46 418:03: The median and the modal values were each R46 617. Thus, half
of the customers who purchased TM195 have a monthly household of R46 617 or less and the most
common income earned by those purchasing a TM195 is R46 617. The minimum score is R29 562
and the maximum income is R68 220. The middle 50% of the customers have a monthly income that
falls between R38 658 and R53 439. About 68:26% of the customers who purchases the TM195 have
monthly household income that range from R37 342:25 to R55 493:81. The coefficient of variation was
19:55% indicating that there was not much variability in monthly income and the ratio of the standard
deviation to the mean is 1 : 5. The monthly household income is almost symmetrical, and only slightly
less peaked and thinner tailed than a normal distribution.

The TM195 customer profile of miles is shown below:

Figure 3.45: Histogram of Miles form TM195

 167 STA1506/1

01/

Figure 3.46: Descriptive Statistics of Miles for TM195

The number of miles a TM195 customers expects to walk/run each week is concentrated around 80
miles. The mean and standard deviation are 82:71 and 28:85 miles respectively. On average, the
miles a TM195 customer expects to walk/run per week on the treadmill is 82:71 miles. The median
and the modal values were both 84:6 miles respectively. Thus, half of the customers who purchased
TM195 expect to walk/run per week for not more than 84:6 miles and the most common walk/runs per
week a TM195 customer expect to do is 84:6 miles. The lowest value is 37:6 miles and the highest
value is 188 miles. The middle 50% of the customers expect to walk/run an average between 65:8
and 94 miles per week. About 68:26% of the customers who purchases the TM195 expect to walk/run
from 53:86 to 111:56 miles. The coefficient of variation was 34:88% indicating that there was some
variability in miles expected to be run or walked and the ratio of the standard deviation to the mean
is almost 1 : 3. The average number of miles the customer expects to walk/run each week is right
skewed, and more peaked and thicker tailed than a normal distribution.
 168

The TM195 correlation matrix is shown in Figure 3.47.

Figure 3.47: Correlation Coefficient for TM195

There is a strong positive correlation between miles and fitness (r = 0:8273) and age and income
(r = 0:7518) : There is a moderate positive correlation between usage and miles (r = 0:6450) and a
weak positive correlation between fitness and usage (r = 0:4688), between education and income
(r = 0:4466) and between education and age (r = 0:3363) : All other correlations are very small.

The TM195 scatter plot of fitness against usage are shown in Figure 3.48.

Figure 3.48: Scatterplot of Fitness vs Usage for TM195

There is a weak positive relationship between usage and fitness. High levels of fitness are associated
with high levels of usage, however the relationship is weak. The simple linear regression for TM195
customers output is shown in Figure 3.49.
 169 STA1506/1

Figure 3.49: Simple Linear Regression of Usage on Fitness for TM195

\ = 1:73 + 0:40Usage.
The least squares regression line is Fitness

b1 = 0:4: For every increase of one unit in usage, fitness increase by 0:4. R2 = 0:2198: About
21:98% of the variability in fitness is being accounted for by usage or by the model. This indicates a
very poor fit.

The TM195 scatterplot of fitness against miles is shown in Figure 3.50

Figure 3.50: Scatterplot of Fitness vs Miles for TM195

There is a strong positive linear relationship between miles and fitness. High levels of fitness are
associated with high levels in miles.
 170

The simple linear regression TM195 customers output is shown in Figure 3.51.

Figure 3.51: Simple Linear Regression of Fitness vs Miles for TM195

\ = 1:39 + 0:02Miles.
The least squares regression line is Fitness

b1 = 0:02: For every increase of 1 mile, fitness increases by 0:02. R2 = 0:6844: About 68:44% of
the variability in fitness is being explained by miles or by the model. This indicates a good fit. Only
31:56% remain unexplained.

TM498:

The TM498 customer profile of age is shown below:

Figure 3.52: Stem and Leaf Plot of Age for TM498

 171 STA1506/1

Figure 3.53: Descriptive Statistics of Age for TM498

The ages of the of TM498 customers is concentrated around 23 25 years. The mean and standard

deviation are 28:9 and 6:65 years respectively. On average, the age of those who purchases TM195
is 28:9 years. The median and the modal values are 26 and 25 years. Thus, half of the customers who
purchased TM498 are not more than 26 years old and the most common age for those purchasing a
TM498 is 25 years. The youngest is aged at 19 years and the oldest is aged 48 years. The middle
50% of the customers fall between 24 and 34 years. About 68:26% of the customers who purchases
the TM498 have ages that range from 22:25 years and 35:55 years. The coefficient of variation was
22:99% which indicates not much variability. Thus, the ratio of standard deviation to mean is almost
1 : 4. Age is right-skewed, and has a peak and tails almost identical to be a normal distribution.

The TM498 customer profile of education is shown below:

Figure 3.54: Stem and Leaf Plot of Education for TM498

 172

Figure 3.55: Descriptive Statistics of Education for TM498

The education in years of TM498 customers is concentrated around 15 years. The mean and
standard deviation are 15:12 years and 1:22 years respectively. On average, the education of TM498
customers is 15:12 years. The median and the modal values are both 16 years. Thus, half of the
TM498 spend not more than 16 years of education and the most common time taken by customers
to do their education is 16 years. The lowest value is 12 years and the highest value is 18 years.
About 68:26% of the customers who purchases the TM498 have undertaken between 13:9 years
and 16:34 years in education. The coefficient of variation was 8:09% indicating that there was not
much variability in educational levels and the ratio of the standard deviation to the mean is 1 : 12.
The years in education is almost symmetrical, and only slight less peaked and thinner tailed than a
normal distribution.

The TM498 customer profile of usage is shown below:

Figure 3.56: Stem and Leaf Plot of Usage for TM498

 173 STA1506/1

Figure 3.57: Descriptive Statistics of Usage for TM498

The average number of times the TM498 customers plans to use the treadmill each week is
concentrated around 3. The mean, median and mode are 3:07, 3 and 3 respectively. On average
the TM498 customers expect to use the treadmill 3 times a week, half of the customers expect to
use the treadmill not more than 3 times a week and the most common time customers expect to use
the treadmill is 3 times a week. The lowest value is 2 and the highest value is 5. The middle 50% of
the usage value falls between 3 and 4. The mean spread around the mean usage is 0:80 giving a
coefficient of variation of 26:08% indicating that there was not much variability in usage and the ratio
of the standard deviation to the mean is 1 : 4:.About 68:26% of the customers who purchases the
TM498 expect to use the treadmill from 2:27 to 3:87 a week, that is expect to use the treadmill TM498
2 to 4 times a week. The usage value is only slightly right-skewed, and has a peak and tails almost
identical to a normal distribution.

The TM498 customer profile of fitness is shown below:

Figure 3.58: Stem and Leaf Plot of Fitness for TM498

 174

Figure 3.59: Descriptive Statistics of Fitness for TM498

The self-rated fitness of customers who purchased the TM498 is concentrated around 3. The mean,
median and mode are 2:9, 3 and 3 respectively. On average the TM498 customers gave a self-rating
fitness of 3, half of the customers gave a self rating of not more than 3 and 3 was the most common
self-rated fitness.. The lowest value was 1 and the highest value was 4. The middle 50% of the fitness
value equals to 3. The mean spread around the mean fitness value was 0:63 giving a coefficient of
variation of 21:72% indicating that there was not much variability in self-rated fitness and the ratio of
the standard deviation to the mean is almost 1 : 5:.About 68:26% of the customers who purchases
the TM498 have self-rated fitness that range from 2:27 to 3:53, that is will give a self rating fitness
that range from 2 to 4. The fitness value is slightly left-skewed, and more peaked and thicker tailed
than a normal distribution.
 175 STA1506/1

The TM498 customer profile of income is shown below:

Figure 3.60: Histogram of Income form TM498

Figure 3.61: Descriptive Statistics of Income for TM498

The monthly household income of TM498 customers is concentrated around R47 500. The mean
and standard deviation are R48 973:65 and R8 653:99 respectively. On average, the monthly income
of TM498 customers is R48 973:65: The median and the modal values were R49 459:5 and R45 480
respectively. Thus, half of the customers who purchased TM498 have a monthly household income
of not more than R49 459:5 and the most common income earned by those purchasing a TM498 is
R45 480: The minimum income is R31 836and the maximum income is R67 083. The middle 50% of
 176

the customers have a monthly income that falls between R43 206 and R53 439. About 68:26% of the
customers who purchases the TM498 have a monthly household income that range from R40 319:66
to R57 627:64. The coefficient of variation was 17:67% indicating that there was not much variability
in monthly income and the ratio of the standard deviation to the mean is almost 1 : 6. The monthly
household income is almost symmetrical and only slightly less peaked and thinner tailed than a
normal distribution.

The TM498 customer profile of miles is shown below:

Figure 3.62: Histogram of Miles form TM498

Figure 3.63: Descriptive Statistics of Miles for TM498

 177 STA1506/1

The average number of miles a TM498 customers expects to walk/run each week is concentrated
around 90 miles. The mean and standard deviation are 87:98 and 33:25 miles respectively. On
average, the miles a TM498 customer expects to walk/run per week on the treadmill is 87:98 miles.
The median and the modal values were 84:8 and 95:4 miles respectively. Thus, half of the customers
who purchased TM498 expect to walk/run per week for not more than 84:8 miles and the most
common walk/runs per week a TM498 customer expect to do is 95:4 miles. The lowest value was 21:2
miles and the highest value was 212 miles. The middle 50% of the customers expect to walk/run an
average between 63:6 miles and 106 miles per week. About 68:26% of the customers who purchases
the TM498 expect to walk/run from 54:73 miles to 121:23 miles. The coefficient of variation was 37:8%
indicating that there was some variability in miles expected to be run or walked and the ratio of the
standard deviation to the mean is almost 1 : 3. The average number of miles the customer expects to
walk/run each week is right skewed, and more peaked and thicker tailed than a normal distribution.

The TM498 customer correlation matrix is shown in Figure 3.64.

Figure 3.64: Correlation Coefficient for TM498

There is a strong positive correlation between age and income (r = 0:8523) and between usage
and miles (r = 0:7432) : There is a moderate positive correlation between education and income
(r = 0:5640) and between fitness and miles (r = 0:6928) : A weak positive correlation was found
between fitness and usage (r = 0:4000) and between education age and education (r = 0:3562). All
other correlations are small.
 178

The scatter plot for TM498 customers for fitness against usage are shown in Figure 3.65.

Figure 3.65: Scatterplot of Fitness vs Usage for TM498

There is a weak positive relationship between usage and fitness. High levels of fitness are associated
with high levels of usage but the relationship is weak. The simple linear regression output is shown
in Figure 3.66.

Figure 3.66: Simple Linear Regression of Fitness vs Usage for TM498

\ = 2:05 + 0:28Usage.
The least squares regression line is Fitness
 179 STA1506/1

b1 = 0:28: For every increase of one unit in usage, fitness increase by 0:28. R2 = 0:1225: About
12:25% of the variability in fitness is being accounted for by usage or by the model. This indicates a
very poor fit since it leaves 87:75% unexplained.

The scatterplot for TM498 customers of fitness against miles is shown in Figure 3.67

Figure 3.67: Scatterplot of Fitness vs Miles for TM498

There is a strong positive linear relationship between miles and fitness. High levels of fitness are
associated with high levels in miles. The simple linear regression output is shown in Figure 3.68.

Figure 3.68: Simple Linear Regression of Fitness vs Miles for TM498

 180

\ = 1:77 + 0:01Miles.
The least squares regression line is Fitness

b1 = 0:01: For every increase of 1 mile, fitness increases by 0:01. R2 = 0:4574: About 45:74% of the
variability in fitness is being explained by miles or by the model. This indicates a poor fit since more
than 50% in this case 54:26% remain unexplained.

TM798:

The TM798 customer profile of age is shown below:

Figure 3.69: Stem and Leaf Plot of

Age for TM798

Figure 3.70: Descriptive Statistics of Age for TM798

The ages of the TM798 customers is concentrated around 22 30 years. The mean and standard
deviation are 29:1 years and 6:97 years respectively. On average, the age of TM798 customers is
 181 STA1506/1

29:1 years. The median and the modal values are 27 and 25 years. Thus, half of the customers who
purchased TM798 are not more than 27 years old and the most common age for those purchasing a
TM798 is 25 years. The youngest is aged at 22 years and the oldest at 48 years. The middle 50% of
the customers fall between 24 and 31 years of age. About 68:26% of the customers who purchases
the TM798 have ages that range from 22:13 years and 36:07 years. The coefficient of variation was
23:96% which is not that far apart from 0% (no variability). Thus, the ratio of standard deviation to
mean is almost 1 : 4. Age is right-skewed, and more peaked and thicker than a normal distribution.

The TM798 customer profile of education is shown below:

Figure 3.71: Stem and Leaf Plot of Education for TM798

Figure 3.72: Descriptive Statistics of Education for TM798

The education in years of TM798 customers is concentrated around 17 years. The mean and
standard deviation are 17:33 years and 1:64 years respectively. On average, the education of TM798
 182

customers is 17:33 years. The median and the modal values are both 18 years respectively. Thus,
half of the TM798 spend not more than 18 years of education and the most common time taken by
customers to do their education is 18 years. The lowest value is 14 years and the highest value of 21
years. About 68:26% of the customers who purchases the TM798 have undertaken between 15:69
and 18:97 years in education. The coefficient of variation was 9:46% indicating that there was not
much variability in educational levels and the ratio of the standard deviation to the mean is almost
1 : 11. The years in education is only slightly right-skewed, and only slight more peaked and thicker
tailed than a normal distribution.

The TM798 customer profile of usage is shown below:

Figure 3.73: Stem and Leaf Plot of

Usage for TM798

Figure 3.74: Descriptive Statistics of Usage for TM798

The average number of times the TM798 customers plans to use the treadmill each week is
concentrated around 5. The mean, median and mode are 4:78, 5 and 4 respectively. On average the
 183 STA1506/1

TM798 customers expect to use the treadmill almost 5 times a week, half of the customers expect
to use the treadmill not more than 5 times a week and the most common time customers expect to
use the treadmill is 4 times a week. The mean spread around the mean usage was 0:95 with the
lowest value of 3 and the highest value of 7. The middle 50% of the usage value falls between 4 and
5. The coefficient of variation was 19:83% indicating that there was not much variability in usage and
the ratio of the standard deviation to the mean is almost 1 : 5:.About 68:26% of the customers who
purchases the TM798 expect to use the treadmill from 3:83 to 5:73 a week, that is expect to use the
treadmill TM798 4 to 6 times a week. The usage value is only slightly right-skewed, and slightly less
peaked and thinner tailed than a normal distribution.

The TM798 customer profile of fitness is shown below:

Figure 3.75: Stem and Leaf Plot of Fitness for TM798

Figure 3.76: Descriptive Statistics of Fitness for TM798

The self-rated fitness of customers who purchased the TM798 is concentrated around 5. The mean,
median and mode are 4:63, 5 and 5 respectively. On average the TM798 customers gave a self-rating
fitness of 5, half of the customers gave a self rating of not more than 5 and 5 was the most common
 184

self-rated fitness. The lowest value was 3 and the highest value was 5. The middle 50% of the fitness
value falls between 4 and 5. The mean spread around the mean fitness was 0:67 giving a coefficient
of variation of 22:43% indicating that there was not much variability in self-rating fitness and the ratio
of the standard deviation to the mean is 1 : 7:.About 68:26% of the customers who purchases the
TM798 have self-rated fitness that range from 3:96 to 4:30, that is will give a self rating fitness that
range from 4 to 5: The fitness value is left-skewed, and more peaked and thicker tailed than a normal
distribution.

The customer profile of income is shown below:

Figure 3.77: Histogram of Income form TM498

 185 STA1506/1

Figure 3.78: Descriptive Statistics of Income for TM798

The monthly household income of TM798 customers is concentrated around R76 000. The mean
and standard deviation are R75 441:58 and R18 505:84 respectively. On average, the monthly income
of TM798 customers is R75 441:58: The median and the modal values were R76 568:5 and R90 886
respectively. Thus, half of the customers who purchased TM798 have a monthly household income
of not more than R76 568:5 and the most common income earned by those purchasing a TM798
is R90 886. The minimum income was R48 556 and the maximum income was R104 581. The
middle 50% of the customers have a monthly income that falls between R57 271 and R90 886. About
68:26% of the customers who purchases the TM798 have monthly household income that range from
R56 935:74 to R93 947:42. The coefficient of variation was 24:53% indicating that there was not much
variability in monthly income and the ratio of the standard deviation to the mean is 1 : 4. The monthly
household income is almost symmetrical, and only slightly less peaked and thinner tailed than a
normal distribution.
 186

The TM798 customer profile of miles is shown below:

Figure 3.79: Histogram of Miles form TM498

Figure 3.80: Descriptive Statistics of Miles for TM798

The average number of miles a TM798 customers expects to walk/run each week is 170 miles. The
mean and standard deviation are 166:9 miles and 60:07 miles respectively. On average, the miles
a TM798 customers expects to walk/run per week on the treadmill is 166:9 miles. The median and
the modal values were 160 and 100 miles respectively. Thus, half of the customers who purchased
TM798 expect to walk/run per week for not more than 160 miles and the most common walk/runs per
week a TM798 customer expect to do is 100 miles. The lowest value was 80 miles and the highest
 187 STA1506/1

value was 360 miles. The middle 50% of the customers expect to walk/run an average between 120
miles and 200 miles per week. About 68:26% of the customers who purchases the TM798 expect
to walk/run from 106:83 miles to 226:97 miles. The coefficient of variation was 34:88% indicating
that there was some variability in miles expected to be run or walked and the ratio of the standard
deviation to the mean is almost 1 : 3. The average number of miles the customer expects to walk/run
each week is right-skewed, and more peaked and thicker tailed than a normal distribution.

The TM798 customer correlation matrix is shown in Figure 3.81.

Figure 3.81: Correlation Coefficient for TM498

There is a strong positive correlation between income and age (r = 0:7135) : There is a moderate
positive correlation between usage and miles (r = 0:5031) and a weak positive correlation between
fitness and miles (r = 0:3911) and miles and income (r = 0:3093). All other correlations are small.

The scatter plot for TM798 customers for fitness against usage are shown in Figure 3.82.

Figure 3.82: Scatterplot of Fitness vs Usage for TM798

There is no linear relationship between usage and fitness. The simple linear regression output is
shown in Figure 3.83.
 188

Figure 3.83: Simple Linear Regression of Fitness vs Usage for TM798

\ = 3:88 + 0:16Usage.
The least squares regression line is Fitness

b1 = 0:16: For every increase of one unit in usage, fitness increase by 0:16. R2 = 0:0521: About
5:21% of the variability in fitness is being accounted for by usage or by the model. This indicates a
very poor fit. About 94:79% of variation is unexplained.

The scatterplot for TM498 customers of fitness against miles is shown in Figure 3.84.

Figure 3.84: Scatterplot of Fitness vs Miles for TM798

There is a no linear relationship between miles and fitness. The simple linear regression output is
shown in Figure 3.85.
 189 STA1506/1

Figure 3.85: Simple Linear Regression of Fitness vs Miles for TM798

\ = 3:9 + 0:004Miles.
The least squares regression line is Fitness

b1 = 0:004: For every increase of 1 mile, fitness increases by 0:004. R2 = 0:1529: About 15:29% of
the variability in fitness is being explained by miles or by the model. This indicates a poor fit. About
84:81% remain unexplained.

CONCLUSIONS:

TM195: TM195 is likely an entry-level, mainstream product catering to the masses. Utilisation
of this item is the same for females and males thus, its customer base represents an even split
between males and females, and the greater proportion of customers who use this are married.
However, looking at females, the majority of the females utilises this treadmill compared to the other
treadmills. In terms of age, their ages range widely from 18 years old to 50 years old.. Also, the
results demonstrated that the average ages of the customers who are utilising this item are around
28:6 years old. Their median income is the lowest across the three treadmill products at R46 617, and
years of education ranges between 12 and 18. They have a median fitness of 3, and there is a wide
variation in the number of times they plan to use the treadmill per week, ranging from 2 times to 5
times a week. On average, they plan to run the least each time they use the treadmill.

The usage and education of the individuals who utilising TM195, are less than the utilisation of
TM798. Those with lower income levels are utilising TM195 as compared to the TM798. The total
crossed miles in this item is 6 616:8 miles. Fitness was seen highly correlated to miles, with this
treadmill, more miles done are likely to lead to high level of fitness. About 68:44% of the variability
 190

in fitness is being explained by miles or by the model which indicates a good fit. Usage is weakly
related to fitness as it only explain 21:98% of the variation in fitness.

TM498: TM498 is likely a small upgrade from TM195 in terms of functionality and price. Almost the
same number of males and females are using this product, that is, its customer base represents
a relatively even split between males and females. The results indicates that partnered are most
utilising this result more than singles. Their age ranges between 19 and 48 years of age, and their
median income is slightly higher at R48 973:65. Years of education range similarly between 12 and
18 and the median fitness is also 3. There is a lower variation in the number of times they intend to
use the treadmill per week, ranging only between 3 to 4 times a week. On average,they plan to run
slightly more each time they use the treadmill.

Age, fitness and the education of people who are utilising this product are almost the same as TM195.
Thus, buyers of TM798 rate their fitness levels more highly than buyers of TM195 andTM498. Those
with lower income levels are utilising TM498 as compared to TM798. The total crossed miles in this
product are 5 278:8 miles. Fitness was seen to be moderately correlated to miles, with this treadmill,
more miles done are likely to lead to high level of fitness. About 45:74% of the variability in fitness
is being explained by miles or by the model. This indicates a poor fit since more than 50% in this
case 54:26% remain unexplained. Usage is weakly related to fitness as it only explain 12:25% of the
variation in fitness.

TM798: TM798 is likely a top-end treadmill with full, advanced functionality catered to the Fitness
and running enthusiasts. TM798 has a distinct customer profile that separates it clearly from that
of TM195 and TM498. The 82:5% of the people who are utilising this product are males, thus, its
customer base is dominated by males, with a female: male ratio of about 1 : 5. The results indicates
that partnered are more utilising this result more than singles. The median fitness is 5 (the highest
possible rating), the median number of times they plan to use the treadmill per week is 5, and the
median number of miles they intend to log is 160 miles, which is almost twice the median of TM195
and TM498 customers. These enthusiasts also earn more than TM195 and TM498 buyers. They
have a median income of R76 568:50;almost 1:5 times that of the buyers of TM195 andTM49, and
they receive more years of education, ranging between 14 to 21 years.

Additionally, the levels of usage, fitness, education, and income who are utilising this product are
higher as compared to the other treadmills. It is the treadmill that is being utilised by the high income
earning group. The total miles of this product are 6 676 miles which would seem likewise higher over
those two previous results. Fitness was weakly related to usage and miles. Thus these two are not
good predictors for fitness. About 5:21% of the variability in fitness is being accounted for by usage
or by the model. This indicates a very poor fit as 94:79% of variation is unexplained. About 15:29%
of the variability in fitness is being explained by miles or by the model. This indicates a poor fit since
84:81% remain unexplained.
 191 STA1506/1

RECOMMENDATIONS TO THE MANAGEMENT OF CARDIOFOOD FITNESS:

=) Treadmill TM195 is sought the most and used the least, and Treadmill TM798, is sought the least
but used the most.

=) TM195 can be useful for acquiring capital and TM798 useful for contributing to a fitness mission
statement. It would be beneficial in both aspects of capital and fitness promotion could either sell
more of TM 798 or get TM195 to be used more.

=) TM798 has a distinct customer profile that separates it clearly from that of TM195 [Link]
customer base is dominated by males, with a female: male ratio of about 1 : 5.

=) As a company, if management of CardioGood Fitness are to change what products they sell they
must answer themselves if they want to sell low priced quantity of goods which may be the more
profitable route, or focus their attention on their higher end products who the consumers use more.

=) With the latter decision, Cardio Fitness may have the option in the future to promote its goods as
showing proven results to entice more buy.
 192

Scenario B :

Foreign Beers:

Suppose that the National Consumer Commission of South Africa is worried that the way people
are drinking beer is causing most people to increase weight and then have heart problems. The
commission feels that beers produced in South Africa do not present problems but rather beers
imported from outside. The National Consumer Commission would like to look at the best selling
156 foreign beers in terms of percentage of alcohol, number of calories per 12 ounces and number
of carbohydrates (in grams) per 12 ounces. The aim is to be able to make recommendations on the
characteristics of descriptive statistics of these variables and the relationship between the variables.
The first fifteen brands are shown below.

You need to present a report to the management of the National Consumer Commission of South
Africa taking into consideration these issues.

(a) Compute a percentage histogram for percentage alcohol, number of calories per 12 ounces, and
number of carbohydrates (in grams) per 12 ounces.

(b) Construct three scatter plots: percentage alcohol versus calories, percentage alcohol versus
carbohydrates and calories versus carbohydrates.

(c) Discuss what you have learnt from the graphs you have plotted.

(d) Compile a descriptive evaluation of each of the three numerical variables and also include boxplots
of the data.
 193 STA1506/1

(e) Determine the correlation matrix between the variables and comment on the relationships.

Ensure that the report is given a title and all charts and tables are labelled and detailed findings
are presented and recommendations are made to the management of the National Consumer
Commission of South Africa.

Solution to Scenario B

Managerial Report of Foreign Beers:

A total of 180 foreign beers were evaluated. The percentage histogram for percentage alcohol is
shown in Figure 3.86.

Figure 3.86: The Percentage Histogram on Alcohol Content

The alcohol percentage is concentrated between 4% and 6%, with more between 4% and
5%. There are outliers in the percentage of alcohol in both tails of the distribution. However, the
alcohol content distribution is right-skewed, that is, positively skewed indicating that there are few
beers with alcohol content as high as around 11:5%.
 194

The percentage histogram for carbohydrates is shown in Figure 3.87.

Figure 3.87: The Percentage Histogram on Carbonhydrates

The carbohydrates are concentrated between 12 and 15. There are a few
beers with carbohydrates as high as 32 making the distribution slight positively skewed. .

The percentage histogram for calories is shown in Figure 3.88.

Figure 3.88: The Percentage Histogram on Carolies

The calories are concentrated between 140 and 160. There are a few beers with calorie content as
high as around 320 making the distribution a right-skewed, that is, positively skewed
 195 STA1506/1

The three histograms shows that they foreign beers with alcohol content, high carbohydrates and
high calories. The question is are these the same beers or not. The scatterplot of percentage
alcohol versus calories is shown in Figure 3.89.

Figure 3.89: The Scatterplot of Alcohol vs Calories

There is a positive relationship between percentage of alcohol and calories, High values in alcohol

content are associated with high values in calories.

The scatterplot of percentage alcohol versus carbonhydrates is shown in Figure 3.90.

Figure 3.90: The Scatterplot of Alcohol vs Cabonhydrates

There is a positive relationship between percentage of alcohol and carbohydrates. High values in

alcohol content are associated with high values in carbohydrates.

 196

The scatterplot of calories and carbohydrates is shown in Figure 3.91.

Figure 3.91: The Scatterplot of Alcohol vs Calories

There is a positive relationship between percentage carbonhydrates and calories, High values in

carbohydrates are associated with high values in calories.

It can be concluded that there is a positive relationship between percentage of alcohol and calories
and between calories and carbohydrates, and positive relationship between percentage alcohol and
carbohydrates.

The alcohol content descriptive statistics are shown below:

Figure 3.92: The Boxplot of Alcohol Content

 197 STA1506/1

Figure 3.93: Descriptive Statistics for Alcohol Content

The amount of % alcohol is right skewed with an average at 5:24%. Half of the beers have %
alcohol of not more than 4:9%. The middle 50% of the beers have alcohol content spread over a
range of 1:20%. The highest alcohol content is at 11:5% while the lowest is at 0:1% giving a range of
11:4%: The typical spread of alcohol content around the mean is 1:41% and a coefficient of variation
of 26:91% indicating that there was not much variability in alcohol content and the ratio of the standard
deviation to the mean is 1 : 4:.About 68:26% of the alcohol content ranges from 3:83 to 6:65.

The calories content descriptive statistics are shown below:

Figure 3.94: The Boxplot for Calories

 198

Figure 3.95: Descriptive Statistics for Calories

The number of calories is symmetric with an average at 151:40. Half of the beers have calories
of not more than 150:00: The middle 50% of the beers have calories spread over a range of 35:25. The
highest number of calories is 330 while the lowest is 55. The typical spread of calories around the
mean is 44:07 and a coefficient of variation of 29:11% indicating that there was not much variability
in calories content and the ratio of the standard deviation to the mean is 2 : 7:.About 68:26% of the
calories content ranges from 107:33 to 195:47.

The carbohydrates content descriptive statistics are shown below:

Figure 3.96: The Boxplot of Carbonhydrates

 199 STA1506/1

Figure 3.97: Descriptive Statistics for Carbohydrates

The number of carbohydrates is almost symmetric however it had three high outliers from the boxplot
with an average at 11:93, which is almost identical to the median at 12:01 and a mode of 12. Half of
the beers have carbohydrates not more than 12:01. The middle 50% of the beers have carbohydrates
spread over a range of 5:55. The highest number of carbohydrates is 32:10 while the lowest is 1:9.
The typical
spread of carbohydrates around the mean is 4:87 and a coefficient of variation of 40:82% indicating
that there was some variability in carbohydrates and the ratio of the standard deviation to the mean
is almost 2 : 5:.About 68:26% of the alcohol content ranges from 7:06 to 16:8:

The correlation matrix is shown in Figure 3.98.

Figure 3.98: The Correlation Matrix

There is a very strong positive linear correlation between alcohol content and calories (r = 0:9094) ;
a strong positive correlation between carbohydrates and calories (r = 0:7952) : There is a moderate
positive correlation between alcohol and carbohydrates (r = 0:5203:)
 200

The correlations have shown that there is positive relationship between variables. Alcohol content
and calories are highly correlated.

CONCLUSIONS:

=) The statistics reveal that there are some beers with high alcohol content, high carolies and high
carbohydrates.

=) All the distributions are positively skewed.

=) It can be concluded that there is a positive relationship between percentage of alcohol and calories
and between calories and carbohydrates, and positive relationship between percentage alcohol
and

carbohydrates

=) Due to the positive relationship between the variables, this might mean that some of the beers
with high alcohol content, might be the ones with high carbohydrate content and high calories.

RECOMMENDATIONS:
We recommend further investigation into the beers with high levels of alcohol, high calories and high
carbohydrates so that some screening can be done and these beers can not be imported into South
Africa.
 201 STA1506/1

Scenario C:

AfriSoft Supply for Houses for Sale:

AfriSoft is considering diversifying and entering the South African housing market. It is only interested
in a short-term investment. To help it with the decision and to assess the market, its analysts have
extracted a time series that covers South African supply of houses for sale at current sales rate. The
time series is not adjusted for seasonality and the data set reflects true market movements. The
table below covers data from January 2015 to June 2019 (Year Y , Month M , Value V ):

Write a report to the management of AfriSoft by analysing the data and producing forecasts. In your
report pay specific attention to the following.

(a) Graph the time series.

(b) Decide on the type of time series.

(c) Use the best suited method and produce forecasts until the end of 2019 (six time periods in the
future).

(d) Decide what would be your recommendations, from the data analysis point of view, to AfriSoft.

Solution to Scenario C

Managerial Report of AfriSoft Supply for Houses for Sale:

The data depicts South African supply of houses for sale at current sales rate from January 2015
to June 2019. The time series is not adjusted for seasonality and the data set reflects true market
 202

movements. The time series plot is shown in Figure 3.99.

Figure 3.99: Time Series Plot of House Values for Sale

The graph shows that over the years there is an increasing trend (secular) in sales of houses although
there seem to be a decline in 2019 and the last three months seem to show constant value of houses.
Looking at the graph it can be noted that the time series has a basic tendency to grow over a period
of time. Thus, the sale value show a general increase in the value of the sales over time and these
are steady movements over a long time. The increase in this case is assumed to form a line since a
line can be fitted. Thus it can be concluded that the sales value of houses follow a secular trend or
long term trend.

A trend line using the least squares method was fitted to the data and the output in Figure 3.100 was
obtained.
 203 STA1506/1

Figure 3.101: The Trendline of House Values for Sale

1. The estimated trend line is yb = 2:5618 + 0:1349x; x = 1; 2; 3; : : : ; 54. The slope

b1 = 0:1349: Every month, the house sale value increases by 0:1349: The trend line estimates for
2019 the remaining six months are shown in Figure 3.101.

Figure 3.101: Forecast Values July 2019 to December

2019

The forecast values show that the house sales value will be from 9.98 to 10.66 showing a steady
increase.

CONCLUSIONS:

=) The house sale values seem to be increasing steadily as we move into the future.
 204

=) However, the last six months of 2019 seem to show a decline in sales which is a cause of concern.

RECOMMENDATIONS:
Generally the trend has been showing an increase over the years however the last six months are
a cause of concern. Since the times series trend line shows an overall increase from 2015 to 2018
but a decline in the first six months of 2019, Afrisoft should wait and see whether the house sales
values will continue to decrease in the last six months of 2019 since there are looking for a short
term investment. They will then invest if the sales show an upward trend.

3.7 Conclusion
We hope this has given you a starting point in the field of data analysis. We also challenge you to go
and look for free packages like R on the Internet which are used for data analysis and start practising
some of the statistics we have done. As you go through your courses you do not cover all the things
and all the softwares. In order for you to be a good Statistician you need to go out there and learn as
many data analysis softwares as possible. Nowadays they are many courses online that can assist
you. WISH YOU WELL IN YOUR JOURNEY AS STATISTICIANS WHO WHISPER TO DATA AND
IT DOES WONDERS.
 205 STA1506/1

Activity 3.3
1. The student news service at Clear Mountain State University (CMSU) has
decided to gather data about the undergraduate students that attend CMSU
They create and distribute a survey of 14 questions and receive responses

from 62 undergraduates. The questions were

Q1 What is your gender? (Female, Male)

Q2 What is your age (as of last birthday)?
Q3 What is your class? (Freshman, Sophomore, Junior, Senior)
Q4 What is your major area of study? (Management, Economics/Finance and etc.)
Q5 At the present time, do you plan to attend graduate school? (Yes, No, Undecided)
Q6 What is your current cumulative grade point average (GPA)?
Q7 What is your current employment status? (Unemployed, Part-time, Full-time)
Q8 What would your expect your starting annual salary (in US$1000) to be if you.
were to seek full-time employment immediately after obtaining your
bachelor’s degree?
Q9 For how many social networking sites are you registered?
Q10 How satisfied are you with the food and dining services on campus?
(1 - extremely unsatisfied to 7 - extremely satisfied)
Q11 About how much money did you spend this semester for textbooks and supplies?
Q12 What type of computing device do you prefer to use for your studies?
(Laptop, Tablet, Desktop)
Q13 How many text messages do you send in a typical week?
Q14 How much wealth (income, savings, investment, real estate and other assets)
would you have to accumulate (in millions of dollars)
before you would you are rich?
The file is stored under myunisa, additional resources in the folder Study
Guide Data Sets as UndergradSurvey
You need to write a report to the management of CMSU on the profile of the
undergraduate student and their ambitions. In your report ensure that you:

(a) Determine whether the variable is categorical or numerical. If you determine

that the variable is numerical, identify whether it is discrete or continuous
(b) Construct all the appropriate tables and charts.

(c) Compute all the appropriate descriptive statistics.

(d) Construct a correlation matrix of all the continuous variables.

.
 206

Activity 3.3 (cont’d)

(e) Use the method of least squares to estimate the relationship on how
(i) Salary expected can predict wealth ambitions.
(ii) Age can predict the text messages send.
(iii) Age can predict the satisfaction level with food and dining services.

(iv) Money spend on textbooks can predict accumulated GPA.

(f) Write a report summarising your conclusions and ensure that you interpret.

all your findings.

2. The file CEO-Compensation2013 (Study Guide Data Sets) includes the total
compensation (in millions of US$) for CEOs of 200 public companies and
the investment return in 2013. There are only two numeric variables:
compensation and return. The first ten cases are shown below

You would like to write a report to management and you main objective is to find out
whether there is a relationship between compensation and return. Write a report
and ensure that you:
(a) Construct frequency distributions and percentage distributions. .

(b) Construct histograms and percentage polygons.

(c) Construct cumulative percentage distributions and plot cumulative.

percentage polygons (ogives).

(d) What conclusions can you reach concerning CEO compensation and return in 2013.

(e) Construct scatter plot of total compensation and investment return in 2013.

(f) What is the relationship between the total compensation and investment return in 2013?.
.
 207 STA1506/1

Activity 3.3 (cont’d)

(g) Compute the mean, median, first quartile, and third quartile. Comment on the results.

(h) Compute the range, interquartile range, variance, standard deviation, and
coefficient of variation. Comment on the results.
(i) Compute a boxplot. Are the data skewed? If so, how?.

(j) Compute the correlation coefficient between compensation and the investment

return. Comment on the finding.

(k) Write a short summary of your findings on the relationship between the two.
Do the results surprise you? Ensure that in your recommendations you advice the

management on the relationship between these two variables.

3. Statistics South Africa (Stats SA) is concerned about the regulations that have
been put by Home Affairs regarding overseas tourists requirements on entry into
South Africa. Stats SA has been tasked by the Tourism Minister to use data
from 2013 to 2019 to come up with predictions for 2020 and 2021. The
Minister is concerned on determining whether the predictions for 2020 and
2021 will show decline in overseas tourists. The number of overseas visitors

to South Africa (’000s) estimated by Stats SA for the years 2012 to 2019 are:
2012 2013 2014 2015 2016 2017 2018 2019
5990 6065 6369 6574 6977 7709 8012 7500
Write a report ensuring that you:

(a) Graph the data and assess its suitability for linear trend projection.

(b) Estimate the trend line equation.

(c) Interpret the slope of the regression equation and what it means.

for future overseas tourists visits to South Africa.

(d) Use the trend line equation to forecast time series for 2020 - 2021.

(e) Ensure that you interpret your findings thoroughly and make
recommendations
 208

Exercise 3.1

1. A gender activists investigated a study on issues on whether the boy child or girl child should be
taught the same household chores as they grow up. The gender activists did a survey where she
asked university students across the universities in South Africa for their opinion whether there
are in support or do not support. The following data was obtained:

GENDER
Opinion Male Female Total
Support 14678 28202 42880
Do not support 13921 11264 25185
Total 28599 39466 68065

(a) Construct contingency tables on total percentages, row percentages, and column
percentages.

(b) Which type of percentage, row, column, or total; do you think is the most informative for theses
data?. Explain.

(c) What conclusions do you reach from these analyses?

(d) Construct a multiple bar chart of opinion and gender.

(e) What conclusions do you reach from this chart?

(f) In your opinion is the unit of analysis of a university student, appropriate for this study. Explain
your response.

2. Twenty-six dental patients require the following numbers of fillings during their current course of
treatment.:
2 3 2 2 3 1 2 2 1 3 2 2 2
2 4 3 2 2 2 2 2 1 1 0 1 1

(a) Construct a dot plot of the data.

(b) What can you say about the shape of the distribution?

(c) Find the mode, median and mean for these data.

(d) Interpret the three measures of central tendency?

 209 STA1506/1

(e) Compare your results and comment on the likely shape of the distribution.

(f) Plot a simple bar chart to portray the data and comment..

3. Thirty households in a Paarl suburb were surveyed to identify their average water usage per month
(in kilolitres, kl). The usage per household was:

10 18 30 13 42 14 9 15 19 20
25 15 24 12 15 16 22 22 8 33
50 26 16 32 25 26 16 26 25 12

Use Excel’s Data > Data Analysis > Descriptive Statistics option and, where necessary, the
function key QUARTILE to answer the following questions:

(a) Find the mean, median and modal water usage across the 30 households.

(b) Find the variance and standard deviation of water usage per household.

(c) Find the first and the third quartile of water usage amongst the 30 households.

(d) Interpret the findings from (a) to (c) for the municipal officer who conducted this survey.

(e) If there are 750 households in the Paarl suburb, what is the most likely total water usage (in kl)
amongst all these households:

(i) in a month?

(ii) in a year.

3. Home-Made Beers Ltd is a brewery that is undergoing a major expansion after being bought out
by a large national brewery. Home-Made Beers Ltd produces a small range of beers and stouts,
but it is reowned for the quality of its beer - having won a number of prizes at trade fairs. The
new parent company is reviewing the quality-control mechanisms of Home-Made Beers Ltd and
is concerned by the quantity of the lager contained in the bottles produced at the brewery. These
bottles should contain a mean of 330 ml with a standard deviation of 15 ml. The bottling plant
manager provided the parent company with quantity measurements from 100 bottles for analysis.
 210

348 343 337 333 324 349 348 352 335 374
337 346 347 333 315 346 356 362 340 348
354 356 343 337 356 322 361 352 360 354
344 341 324 339 339 353 342 336 331 355
355 345 345 334 349 331 351 343 346 339
339 351 318 363 334 348 336 332 375 365
328 350 330 367 316 350 345 336 322 329
351 328 343 349 361 337 325 328 360 347
360 358 365 354 326 341 360 342: 346 351
360 350 338 324 342 333 338 340 350 335

(a) Construct a frequency table from these data.

(b) Use the frequency table in part (a) to construct a histogram.

(c) From the data set, calculate the mean, median and mode.

(d) Interpret the three measures of central tendency?

(e) Compare your results and comment on the likely shape of the distribution.

(f) Calculate the standard deviation and interpret it.

(g) Calculate the quartiles (Q1 , Q2 and Q3 ), the interquartile range, semi-interquartile range and
interpret each statistics.

(h) Do the results suggest that there is a great deal of variation in quantity in the bottle
measurements. Compare the assumed average bottle average and spread with the measured
average and spread.

(i) What conclusions can you draw from these results?

5. What was the average price of a room at a two-star, tree-star, and four-star hotels in cities around
the world in summer of 2010? The file Hotelprice (Study Guide Data Sets) contains the prices in
English pounds (about US$1.56 as of January 2011).

(a) Construct a frequency distribution and a percentage distribution of each variable.

(b) Construct a histogram and a percentage polygon of each variable.

(c) Construct a cumulative percentage distribution and plot a cumulative percentage polygon
(ogives) for each variable.
 211 STA1506/1

(d) What conclusions can you reach about the cost of two-star, three-star, and four-star hotels.?

(e) Construct scatter plots of the cost of two-star, three-stars, and four-stars hotels?

(f) What conclusions can you reach about the relationship of the price of two-star hotels versus
three-stars hotel versus four-stars hotels.

(g) Compute the mean, median, first quartile, and third quartile of each variable.

(h) Compute the range, interquartile range, variance, standard deviation, and coefficient of
variation of each variable.

(i) Interpret the measures of central tendency and variation within the context of this problem.

(j) Construct boxplots. Are the data skewed? If so, how?.

(k) Compute the coefficient of correlation between the average price a two-star nand three-stars
hotels, between two-star and four-star hotels, and between three-star and four-stars hotels.

(l) Based on (k), what conclusions can you reach about the relationship between the average
price of a room at two-star, three-star, and four star hotel .

(m) Consider the price difference between two-star and three-star hotels and compute its mean
and median. What can you conclude from this.

(n) Consider the price difference between three-star and four-star hotels and compute its mean
and median. What can you conclude from this.

6. The monthly fuel bills of a random sample of 75 Paarl motorists who commute to work daily by
car were recorded in a 2012 survey recorded by Wegner (2012). The data is on
myunisa under additional resources in the directory "Study Guide Data Sets"

(a) Use Data > Data Analysis > Descriptive Statistics option in Excel to find the mean, median,
variance and standard deviation of the monthly fuel bill of the sample of car commuters.

(b) Interpret the meaning of each descriptive statistics in (a).

(c) Is the data skewed and if so, in what direction? What could be the cause of the skewness?

(d) Find the coefficient of variation for monthly fuel bills. Is the relative variability between the
sampled motorists’ monthly fuel bills low?
 212

(e) Use the Excel function key QUARTILE to find the lower and upper quartiles of monthly fuel
bills of motorists and interpret each quartile.

(f) Compile the five-number summary table for the monthly fuel bills and construct a box plot of
the monthly fuel bills.

(g) Describe the profile of the monthly fuel bills of Paarl motorists who use their cars to commute
daily to work and back.

(h) Assume that the cost of fuel is R13 per litre and that there are 25 000 motorists who commute
to work daily in Paarl by car. Estimate the most likely total amount of fuel used (in litres) by all
car commuters in Paarl in a month.

7. The file Thickness (Study Guide Data Sets) contains data on thickness of 50 metal plates and
corresponding values of four predicting variables, pH, Pressure, Temperature and Voltage, all of
which are thought to contribute towards thickness of the plates

(a) Construct a scatter plot using Thickness as the dependent variable and pH as the independent
variable. Comment on the scatter plot.

(b) Assuming a linear relationship, use the least squares method to compute the regression
coefficients b0 and b1 .

(c) Interpret the regression coefficients computed in (b) .

(d) Compute the coefficient of determination, R2 and interpret its meaning.

(e) Repeats parts (a) to (d) with Pressure as the independent variable.

(f) Repeats parts (a) to (d) with Temperature as the independent variable.

(g) Repeats parts (a) to (d) with Voltage as the independent variable.

(h) Compare the results obtained from part (a) to (g) If you have to choose the ’best’ predictor of
Thickness among pH, Pressure, Temperature and Voltage which one would you choose?

8. GDP (Singapore $) for 1990 - 2007 are tabulated below (Statistics Singapore, 2009).
 213 STA1506/1

Year S$ Year S$
1990 66778 1999 140022
1991 74570 2000 159840
1992 80984 2001 153398
1993 93971 2002 158047
1994 107957 2003 162288
1995 119470 2004 184508
1996 130502 2005 199375
1997 142341 2006 216995
1998 137902 2007 243169

(a) Plot the time series.

(b) Does a linear trend appear to be present?

(c) Develop a linear trend equation for this time series?

(d) Plot the estimated line on the same plot with the time series plot.

(e) Use the trend equation to estimate the GDP for the years 2008 - 2010.

9. Home-Made Beers Ltd employs a local transport company to deliver beers to local supermarkets.
To develop better work schedules, the managers want to estimate the total daily travel time for
their drivers’ journeys. Initially, the managers believed that the total daily travel time would be
closely related to the number of kilometers travelled in making the daily deliveries.

Distance Travel Distance Travel

Journey Travelled Time Journey Travelled Time
(km), x (hours), y (km), x (hours), y
1 100 9:3 11 85 7:4
2 50 4:8 12 62 6:4
3 100 8:9 13 98 8:4
4 100 6:5 14 58 4:9
5 50 4:2 15 73 6:8
6 80 6:2 16 81 7:8
7 75 7:4 17 66 6:2
8 65 6 18 72 7:3
9 90 7:6 19 53 4:4
10 90 6:1 20 56 4:6

(a) Plot a scatter plot and comment on a possible relationship between travel time and distance
travelled.

(b) Use Excel Analysis ToolPak to undertake the following tasks.

(i) State the least squares regression model equation.

 214

(ii) Interpret the meaning of the regression coefficient (b1 ).

(iii) Comment on model reliability (r and R2 ).

10. You want to develop a model to predict the selling price of homes based on assessed value. A
sample of 30 recently sold single-family houses in a large city is selected to study the relationship
between selling prices (in ten thousands of dollars) and assessed value (in ten thousands of
dollars). The houses in the city were reassessed at full value one year prior to the study. The
results are in the file called House1 (Study Guide Data Sets).

(a) Determine the independent and dependent variables.

(b) Construct a scatter plot and assuming linear relationship, use the least squares method to
compute the regression coefficients b0 and b1 .

(c) Interpret the regression coefficients computed in (b).

(d) Use the prediction line developed in (a) to predict the selling price for a house whose assessed
value is R1700000:

(e) Compute the coefficient of determination, R2 and interpret its meaning.

11. The data below shows the calories and total fat (in grams) for a sample of 12 veggie burgers made
by a restaurant.
Calories Fat
110 3:5
110 4:5
90 3:0
90 2:5
120 6:0
130 6:0
120 3:0
100 3:5
140 5:0
70 0:5
100 1:5
120 1:5

(a) Construct a scatter plot with calories on the X -axis and total fat on the Y axis.

(b) For each variable, compute the mean, median, first quartile and third quartile.

(c) For each variable, compute the range, interquartile range, variance, standard deviation and
coefficient of variation.
 215 STA1506/1

(d) For each variable, construct a boxplot. Are the data skewed? if so how?

(e) Compute the coefficient of determination between calories and total fat.

(f) What conclusions can you reach about the relationship between the calories and total fat
in veggie burgers? To answer this question, write a short report of your findings. In your
report ensure that you interpret each statistics and indicate what conclusions can you reach
concerning calories and total fat?

12. The dean of students at Clear Mountain State University (CMSU) (Problem 1, activity 3.3)
has learned about the undergraduate survey and has decided to undertake a similar survey for
graduate students at CMSU. She creates and distributes a survey of 14 questions and receives
responses from 44 graduate students (The file is stored under myunisa, additional resources in
the folder Study Guide data Sets as GradSurvey ). The following questions were asked.

Q1 What is your gender? (Female, Male)

Q2 What is your age (as of last birthday)?
Q3 What is your current major area of study? (Accounting, Economics/Finance, and etc.)
Q4 What is your current graduate cumulative grade point average?
Q5 What was your undergraduate major? (Biological sciences, Business, Computers, and etc.)
Q6 What was your undergraduate cumulative grade point average (GPA)?
Q7 What is your current employment status? (Unemployed, Part-time, Full-time)
Q8 How many different full-time jobs have you held in the past 10 years?
Q9. What do you expect your annual salary (in US$000) to be immediately after completion
of your graduate studies if you are employed full time?
Q10 About how much money did you spend this semester for textbooks and supplies?
Q11 How satisfied are you with the MBA program advisory services on campus?
(1 - extremely unsatisfied to 7 - extremely satisfied)
Q12 What type of computer do you prefer to use for your studies?(Laptop, Tablet, Desktop)
Q13 How many text messages do you send in a typical week?
Q14 How much wealth (income, savings, investment, real estate and other assets)
would you have to accumulate (in millions of dollars) before you would you are rich?

You need to write a report to the management of CMSU on the profile of the graduate student and
their ambitions. In your report ensure that you:.

(a) Determine for each variable whether it is categorical or numerical. If you determine that the
variable is numerical, identify whether it is discrete or continuous.

(b) Construct all the appropriate tables and charts.

 216

(c) Compute all the appropriate descriptive statistics.

(d) Construct a correlation matrix of all the continuous variables.

(e) Construct a scatter plot of Graduate GPA and Undergraduate GPA and comment on the
relationship.

(f) Use the method of least squares to estimate the relationship on how

(i) Salary expected can predict wealth ambitions.

(ii) Undergraduate GPA can predict Graduate GPA

(iii) Number of full-time jobs can predict wealth ambition.

(iv) Advisory rating can predict Graduate GPA.

(g) Write a report summarising your conclusions and ensure that you interpret all your findings.

13. The Royal Automobile Club (RAC) is one of the major motoring organisations that offer emergency
breakdown cover in UK. The RAC aims in allocating patrols to meet future demand for vehicle
rescue. The data below summarise actual monthly demands for RAC rescue services over a
five year time period. To meet the national demand for its services in the coming year, the
RAC’s human resources planning department forecasts the number of members expected, using
historical data and market forecasts. It then predicts the average number of breakdowns and
number of rescue calls expected, by referring to the probability of a member’s vehicle breaking
down each year. In year 1, an establishment of approximately 1400 patrols was available to deal
with the expected workload. This figure tend to be reviewed monthly since it was an average
for the year and did not take into account fluctuations in demand in different seasons. Monthly
demand for RAC rescue services years 1 - 5 are given below.
 217 STA1506/1

In your course you did basic time series and you did not calculate seasonal indices or
deaseasonalise a series. Using your basic time series, you need to write a managerial report
advising the RAC on the patrol allocation of year 6. In your managerial report:

(a) Do a profile analysis for each year (yearly demands) by doing descriptive statistics for each
year and constructing box plots.

(b) What can you say about the shape of the distribution of yearly demands?

(c) Plot the time series plot over the five years and comment on the pattern.

(d) Fit a trend line to the data and interpret your regression coefficients.

(e) Use the trend line to estimate the monthly demand for year 6.

(f) State an assumptions you make.

(g) Comment on the validity of your results or otherwise.

14. The file Restaurants (Study Guide Data Sets) contains the cost per meal and the ratings of 50 city
and 50 surburban restaurants on their food, decor and service and their summarised ratings). The
summated ratings is the sum of ratings of the food, decor and service. (Data extracted from Zagar
Survey 2013 New York City Restaurants and Zagar Survey 2012 - 2013 Long Island Restaurants.
by Levine et al., 2013). The Tourism Minister would like you to write a report on the profile on city
and surburban restaurants. When writing your report take into consideration the following.

(a) Do separate analysis for the urban and surburban restaurants by

(i) Construct the five number summaries for the variables cost of a meal, food, decor , service
and the summated ratings

(ii) Construct the box plot of each variable and comment on the shape of the distribution.

(iii) Compute the descriptive statistics for each variable and interpret them.

(iv) Construct a scatter plot of the cost of meal versus summated rating.

(v) Compute and interpret the correlation coefficient of the cost of a meal and summated
rating.

(vi) Assuming a linear cost relationship, use the least-squares method to estimate the
regression equation of predicting cost per meal using summated ratings.
 218

(vii) Interpret the slope of the regression line (b1 ).

(viii) Predict the cost per person for restaurant with a summated rating of 50.

(b) Using what you have obtained compare and contrast the urban and surburban restaurants.

(c) Repeat the procedures in (a) for the whole sample of 100 restaurants.
 219 STA1506/1

3.8 Learning outcomes

Use the following learning outcomes as a checklist after you have completed this study unit to
evaluate the knowledge you have acquired.

After studying study unit 3, you should know (and understand!) the following issues:

report writing

criteria of producing a professional report

interpretation of results

writing a managerial report.

 220

References
Acaps. (2016). Data cleaning, downloaded from: [Link]
/files/acaps_technical_brief_data_cleaning_april_2016_0.pdf

Anderson D. R., Sweeney D. J., Williams T.A., Freeman J., and E. Shoesmith, (2017), Statistics for
business and Economics, Fourth edition, Cengage learning, EMEA ISBN: 978-1-437-2656-7.

Berk K. N. and Carey P. (2004), Data Analysis with Microsoft Excel Updated for Windows XP, First
Edition, Thomson Learning, Inc. ISBN 0 -534-40714-5

Brase C. H. and Brase C. P. (2015). Understandable statistics: Concepts and methods, 11th Edition,
CENGAGE Learning, ISBN: 9781285460918.

Bryman A. and Bell E. (2015). Business Research Methods, Oxford University Press, ISBN
0199668647.

Buglear J. (2005), Quantitative Methods for Business The A-Z of QM, First Edition, Elsevier
Butterworth-Heinemann ISBN:0 7506 5898 3

Charmaz C. (2006). Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis,
Sage publications, ISBN 9780761973539.

Cooper D. and Schindler P. (2014). Business research methods, 12th edition, McGraw-Hill, ISBN
9780077774431.

Davis G., Pecar B. and L. Santana (2014), Business Statistics using Excel: A first course for South
African Students, Oxford University Press Southern Africa (Pty) Ltd

Deming W. E. (1990). Sample design in business research, John Wiley & Sons, ISBN
9780471523703.

Duignan J. (2014), Quantitative Methods for Business Research Using Microsoft Excel, First Edition,
Cengage learning, EMEA ISBN: 978-1-4080-6482-5.

Francis A. and B. Mousley (2014), Business Mathematics and Statistics, 7th Edition, CENGAGE
Learning EMEA ISBN-13: 978-1-4080-8315-4.

Galetto M. (2016). What is data management? Downloaded from [Link]

data-analysis/

Gravetter F. J. and Wallnau L.B. (2017). Statistics for The Behavioral Sciences, 10th Edition,
Cengage Learning, ISBN 9781305856424.

Hair J. F. Jr, Black W. C., Babin B. J. and Anderson, R. E. (2019). Multivariate Data Analysis, 8th
edition, Cengage Learning EMEA, ISBN 9781473756540.

Hair J. F., Celsi M., Money A., Samouel P. and Page M. (2016). Essentials of Business Research
Methods, Third Edition, Routledge Taylor and Francis Group, ISBN 9780765646132.
 221 STA1506/1

Heiman G. (2015). Behavioral Statistics STAT, Student Edition, Cengage Learning, ISBN
9781285458141.

Jackson S. L (2014). Statistics Plain and Simple, Third Edition, Cengage learning, ISBN
9781133955757.

Jones J. & Hidiroglou M. A. (2013). Capturing, coding and cleaning survey data. In:
Designing and conducting business surveys, Snijkers G., Haraldsen G., Jones J. and Willimack
D. Wiley, 9781118447918. Downloaded from: [Link]
_Capturing_Coding_and_Cleaning_Survey_Data

Keller G. (2014). BSTAT2, Student Edition, Cengage learning, ISBN 9781285447681.

Keller G. (2018), Statistics for Management and Economics, 11th Edition, Cengage learning, ISBN
9781337296946.

Levine D.M. Krehbiel T.C and M.L. Berenson (2013), Business Statistics A first Course, Sixth Edition,
Pearson Educated Limited, USA.

Levine D.M. Szabat K.A and D.F. Stephan (2016), Business Statistics A first Course, Seventh Edition,
Global Edition, Pearson Educated Limited, USA.

Marks E. and Lozano B. (2010). Cloud Computing. New Jersey, NJ: Wiley.

Muchengetwa, S. (2005). Business Statistics. Revised ed. Harare: Zimbabwe Open University.

Muchengetwa, S. (2007). Undergraduate Statistics. Revised ed. Harare: Women’s University in

Africa.

Neuman W. L. (2014). Social research methods: Qualitative and quantitative approaches, Seventh
Edition, Edinburgh Gate: Pearson Education Limited.

Obrutsky S. L. (2016). Cloud storage: Advantages, disadvantages and enterprise solutions for
business. Downloaded from [Link]
_Advantages_Disadvantages_and_Enterprise_Solutions_for_Business

Ott R. L. and Longnecker M. T. (2016). An Introduction to Statistical Methods and Data Analysis, 7th
edition, Cengage Learning, ISBN 9781305465527.

Quinlan C., Babin B. J., Zikmund W. G., Carr J. and Griffin M. (2019). Business research methods,
2nd edition, Cengage Learning, ISBN 9781473760356.

Rubin A. and Babbie E. (2016). Essential Research Methods for Social Work, Fourth Edition,
Cengage Learning Empowerment Series, ISBN 9781305633827.

Russom P. (2013). Managing big data. Downloaded from [Link]

/sites/default/files/Managing%20Big%20Data%[Link]

Saunders M., Lewis P. and Thornhill A. (2016). Research methods for business studies, Seventh
Edition, Edinburgh Gate: Pearson Education Limited, ISBN 9781292016627.
 222

Schoenbach V. J. (2002). Data analysis and interpretation. Downloaded from:

[Link]

Wangikar V. C. and Deshmukh R. R. (2011). Data Cleaning: Current Approaches and Issues.
Downloaded from: [Link]
_Approaches_and_Issues

Wegner T. (2012), Applied Business Statistics Method and Excel-based Applications, Third Edition,
Juta & Company Ltd, ISBN 978-0-70217-774-3

Wegner T. (2012), Applied Business Statistics Method and Excel-based Applications Solutions
Manual, Third Edition, Juta & Company Ltd, ISBN 978-0-70218-865-7

Williams T. A., Sweeney D. J., and D. R. Anderson, (2012), Essentials of Contemporary Business
Statistics, Fifth International Edition, South-Western Cengage learning. ISBN: 13: 978-1-133-18765-
3

Zikmund W. G., Babin B. J., Carr J. C. and Griffin, M. (2013). Business research methods, 9th edition,
Cengage Learning, ISBN 9781111826929.

Zins C. (2007). Conceptual Approaches for Defining Data, Information, and Knowledge, Journal Of
The American Society For Information Science And Technology, 58(4):479–493.