Quantitative Methods in Psychology

PSY103

University of Ibadan Distance Learning Centre

Ibadan Open and Distance Learning Course Series Development
Version ev1
Copyright

Copyright © 2007, 2013, 2023 by Distance Learning Centre, University of Ibadan,

Ibadan.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without the prior permission of the copyright
owner.

ISBN: 978-021-278-7

General Editor: Prof. Bayo Okunade ???

Page layout, instructional design & development by

EDUTECHportal, [Link]

University of Ibadan Distance Learning Centre

University of Ibadan,
Nigeria
Telex: 31128NG
Tel: +234 (80775935727)
E-mail: ssu@[Link]
Website: [Link]
Course Development Team
The University of Ibadan Distance Learning Centre wishes to thank those below for
their contribution to this Error! No text of specified style in document.:

Course Writer / Facilitator Benjamin Osayawe Ehigie, Ph.D

Content Editor Prof. Remi Raji-Oyelade

Production Editor Dr. Gloria O. Adedoja
Learning Design & Technologist Folajimi Olambo Fakoya
Managing Editor Ogunmefun Oladele Abiodun
General Editor Prof. Bayo Okunade

???????
 PSY103 Quantitative Methods in Psychology

Contents

About this course manual 1

How this course manual is structured .................................................................... 1

Course overview 3
Welcome to Quantitative Methods in Psychology PSY103 .................................. 3
Quantitative Methods in Psychology PSY103—is this course for you? ............... 3
Timeframe .............................................................................................................. 4
Study skills ............................................................................................................. 4
Need help? ..................................................... ????Error! Bookmark not defined.
Assessments .................................................. ????Error! Bookmark not defined.

Getting around this course manual 7

Margin Icons .......................................................................................................... 7

Study Session 1 8
The Meaning of Statistics....................................................................................... 8
Introduction ................................................................................................... 8
1.1 The Definition of Statistics...................................................................... 8
1.1.1 Layman‘s Perspective .................................................................. 9
1.1.2 Research Perspective ................................................................... 9
1.1.3 Grammatical Perspective ........................................................... 10
1.2 Types of Statistics ................................................................................. 10
1.2.1 Descriptive Statistics ................................................................. 11
1.2.2 Inferential Statistics ................................................................... 11
Study Session Summary ....................................................................................... 13
Assessment ........................................................................................................... 13

Study Session 2 14
Why we Study Statistics....................................................................................... 17
Introduction ................................................................................................. 17
Rationale for the Inclusion of Statistics into Psychology ........................... 17
Study Session Summary ....................................................................................... 19
Assessment ........................................................................................................... 19

Study Session 3 19
Common Terms used in Statistics ........................................................................ 20
Introduction ................................................................................................. 20
3.1 Variable ................................................................................................. 20
 Contents

3.1.1 Classification of Variables......................................................... 20

3.1.2 Measurement of Variables ......................................................... 22
3.2 Data........................................................................................................ 23
3.2.1 Types of Data ............................................................................. 23
3.3 Population .............................................................................................. 24
3.3.1 Parameter ........................................................................... 24
3.3.2 Sample................................................................................ 24
Study Session Summary ....................................................................................... 27
Assessment ........................................................................................................... 27

Study Session 4 29
Sampling............................................................................................................... 29
Introduction ................................................................................................. 29
4.1 What is Sampling? ................................................................................. 29
4.1.1 Important Concepts in Sampling ............................................... 29
4.2 Sampling Techniques ............................................................................ 30
4.2.1 Probability Sampling Techniques .............................................. 30
A. Simple Random Sampling ..................................................... 30
B. Stratified Random Sampling .................................................. 32
C. Cluster Sampling .................................................................... 33
D. Systematic Sampling .............................................................. 33
4.2.2 Non-probability Sampling ......................................................... 34
A. Accidental Sampling .............................................................. 34
B. Quota Sampling ...................................................................... 34
C. Purposive Sampling................................................................ 34
Study Session Summary ....................................................................................... 37
Assessment ........................................................................................................... 37

Study Session 5 39
Scales / Levels of Measurement ........................................................................... 39
Introduction ................................................................................................. 39
5.1 Nominal Scale ....................................................................................... 39
5.2 The Ordinal Scale .................................................................................. 40
5.3 Interval Scale ......................................................................................... 41
5.4 Ratio Scale ............................................................................................. 42
Study Session Summary ....................................................................................... 43
Assessment ........................................................................................................... 43

Study Session 6 45
Frequency Distribution......................................................................................... 45
Introduction ................................................................................................. 45
6.1 Regular Frequency Distribution ............................................................ 45
 PSY103 Quantitative Methods in Psychology

6.1.1 Steps in Constructing a Regular Frequency Distribution? ........ 47

6.2 Grouped Frequency Distributions ......................................................... 48
6.2.1 Steps in Constructing a Grouped Frequency Distribution ......... 50
6.3 The Exact Limits and Mid-Points of Class Intervals ............................ 52
6.3.1 Exact Limits and Mid-Points ..................................................... 52
Study Session Summary ....................................................................................... 54
Assessment ........................................................................................................... 54
Assignment ........................................................................................................... 54

Study Session 7 56
Graphic Representation of Frequency Distribution ............................................. 56
Introduction ................................................................................................. 56
7.1 The Frequency Polygon......................................................................... 56
8.2 The Histogram ....................................................................................... 57
7.3 The Bar Chart ........................................................................................ 59
Study Session Summary ....................................................................................... 61
Assessment ........................................................................................................... 61

Study Session 8 63
Diagrammatic Representation of Frequency Distribution ................................... 63
Introduction ................................................................................................. 63
8.1 The Pie Chart ......................................................................................... 63
8.2 The Pictogram ....................................................................................... 66
8.3 The Stem-and-Leaf Display .................................................................. 67
Study Session Summary ....................................................................................... 69
Assessment ........................................................................................................... 69

Study Session 9 71
Cumulative Distributions ..................................................................................... 71
Introduction ................................................................................................. 71
9.1 Constructing Cumulative Distribution .................................................. 71
9.1.1 Constructing a ―More Than‖ Cumulative Distribution ............. 71
9.1.2 Constructing a ―Less Than‖ Cumulative Frequency Distribution
............................................................................................................ 72
9.2 Graphical Presentations of Cumulative Distribution ............................ 73
9.2.1 Graphical Presentation of a ―More Than‖ Cumulative
Distribution. ........................................................................................ 73
9.2.2 Graphical Presentation of ―Less Than‖ Cumulative Distribution.
............................................................................................................ 74
 Contents

Study Session Summary ....................................................................................... 75

Assessment ........................................................................................................... 76

Study Session 10 77
Measures of Central Tendency:Mean .................................................................. 77
Introduction ................................................................................................. 77
10.1 The Arithmetic Mean .......................................................................... 78
10.1.1 Ungrouped Data ....................................................................... 78
10.1.2 Regular Frequency Distribution .............................................. 79
10.1.3 Grouped Frequency Distribution with Class Intervals ............ 80
..................................................................... Error! Bookmark not defined.
Study Session Summary ....................................................................................... 81
Assessment ........................................................................................................... 82

Study Session 11 82
Measures of Central Tendency: Median .............................................................. 83
Introduction ................................................................................................. 83
11.1 The Median of Ungrouped Data .......................................................... 83
11.2 The Median of Grouped Data .............................................................. 84
11.2.1 The Median of Grouped Data with Regular Frequency
Distribution ......................................................................................... 84
11.2.2 The Median of Grouped Data with Grouped Frequency
Distribution with Class Intervals ........................................................ 85
11.3 Determining the Median using the Cumulative Frequency (Or OGIVE
Curve) .......................................................................................................... 88
Study Session Summary ....................................................................................... 89
Assessment ........................................................................................................... 89

Study Session 12 90
Measures of Central Tendency: The Mode .......................................................... 90
Introduction ................................................................................................. 90
12.1 Mode of Grouped Data with Intervals ................................................. 90
12.1.1 The Crude Mode ...................................................................... 90
12.1.2 The Interpolated Mode ............................................................ 91
12.2 Determining the Mode using the Histogram ....................................... 93
12.3 The Characteristics and Use of Measures of Central Tendency ......... 93
Study Session Summary ....................................................................................... 95
Assessment ........................................................................................................... 95

Study Session 13 96
Measures of Variability ........................................................................................ 96
Introduction ................................................................................................. 96
 PSY103 Quantitative Methods in Psychology

13.1 The Range ............................................................................................ 97

13.2 The Mean Absolute Deviation (MAD) ............................................... 99
13.3 The Standard Deviation (S.D.) .......................................................... 101
13.3.1 Calculation of S.D. from Ungrouped Data ............................ 103
13.3.2 Calculation of the MAD and S.D. from Grouped Data ......... 104
Study Session Summary ..................................................................................... 105
Assessment ......................................................................................................... 106

Study Session 14 107

Transformed Scores ........................................................................................... 107
Introduction ............................................................................................... 107
14.1 Percentiles.......................................................................................... 107
14.1.1 Computational Procedure: Computing the Corresponding
Percentile Rank from a given Raw Score ........................................ 108
14.1.2 Computational Procedure: Computing the Corresponding Raw
Score from a Given Percentile .......................................................... 110
14.2 ―Z and T‖ Scores ............................................................................... 112
14.2.1 Standard Scores (Z Scores).................................................... 112
14.2.2 T Scores ................................................................................. 113
Study Session Summary ..................................................................................... 114
Assessment ......................................................................................................... 114

References 115

Feedback on SAQs 116

 PSY103 Quantitative Methods in Psychology

Error! No text of specified style in

About this

document.
Quantitative Methods in Psychology PSY103 has been
produced by University of Ibadan Distance Learning Centre.
All Psychology Error! No text of specified style in document.s
produced by University of Ibadan Distance Learning Centre
are structured in the same way, as outlined below.

How this Error!

No text of specified
style in document. is structured
The course overview
The course overview gives you a general introduction to the
course. Information contained in the course overview will
help you determine:
 If the course is suitable for you.
 What you will already need to know.
 What you can expect from the course.
 How much time you will need to invest to complete the
course.
The overview also provides guidance on:
 Study skills.
 Where to get help.
 Course assessments and assignments.
 Activity icons.
 Study sessions.
We strongly recommend that you read the overview carefully
before starting your study.
The course content
The course is broken down into study sessions. Each study
session comprises:
 An introduction to the study session content.
 Learning outcomes.
 Content of study sessions with learning activities.
 A study session summary.
 Assessments and/or assignment, as applicable.
About this Error! No text of specified style in document. The Meaning of Statistics

Your comments
After completing this course, Quantitative Methods in
Psychology, we would appreciate it if you would take a few
moments to give us your feedback on any aspect of this
course. Your feedback might include comments on:
 Course content and structure.
 Course reading materials and resources.
 Course assessments.
 Course assignments.
 Course duration.
 Course support (assigned tutors, technical help, etc).
 Your general experience with the course provision as a
distance learning student.
You might forward your comments to
coursereview@[Link] or post same on comment pad of
your course website at UI Mobile Class. Your constructive
feedback will help us to improve and enhance this course.

2
 PSY103 Quantitative Methods in Psychology

Course overview

Welcome to Quantitative Methods in Psychology

PSY103

Psychology is a science and, therefore, adopts scientific

procedures in its study, including the computation aspects. For
every student desiring to study psychology, therefore, the
knowledge of statistics is essential. This course - PSY103 -
will expose you to the quantitative skills needed for scientific
research.
This course manual supplements and complements PSY103
UI Mobile Class Activities as an online course. The UI
Mobile Class is a virtual platform that facilitates classroom
interaction at a distance where you can discuss / interact with
your tutor and peers while you are at home or office from
your internet-enabled computer. You will also use this
platform to submit your assignments, receive tutor feedback
and course news with updates.

Quantitative Methods in Psychology PSY103—is

this course for you?

PSY103 is a compulsory course for undergraduate students of

psychology. The course attempts to equip you with
quantitative skills needed for scientific research.
Course outcomes
Upon a successful completion of Quantitative Methods in
Psychology PSY103 you will be able to:
Course overview The Meaning of Statistics

 explain the concept of statistics.

 justify why statistics is essential in the study of
psychology.
Outcomes  differentiate among the types of measurements used for
data collection in psychological research.
 collect and summarize data; and present facts in
psychological research.
 expose data to various forms of descriptive statistical
analysis, graphs and diagrammatic presentation.

Timeframe

This is a one semester course.

45 hours of formal study time is required.

How long?

Study skills
As an adult learner your approach to learning will be different
to that from your school days: you will choose what you want
to study, you will have professional and/or personal
motivation for doing so and you will most likely be fitting
your study activities around other professional or domestic
responsibilities.
Essentially you will be taking control of your learning
environment. As a consequence, you will need to consider
performance issues related to time management, goal setting,
stress management, etc. Perhaps you will also need to
reacquaint yourself in areas such as essay planning, coping
with exams and using the web as a learning resource.
Your most significant considerations will be time and space
i.e. the time you dedicate to your learning and the
environment in which you engage in that learning.
We recommend that you take time now—before starting your
self-study—to familiarize yourself with these issues. There are
a number of excellent web links and resources on the ―Self-

4
 PSY103 Quantitative Methods in Psychology

Study Skills‖ at your course website.

Need help?

As earlier noted, this course manual complements and

supplements PSY103at UI Mobile Class as an online course.

Help You may contact any of the following units for information,
learning resources and library services.

Distance Learning Centre Head Office ???????

(DLC) Morohundiya Complex,
Ibadan-Ilorin Expressway,
University of Ibadan, Nigeria Idi-Ose, Ibadan.
Tel: (+234) 08077593551 – 55 ????????????
(Student Support Officers)
Email: ssu@[Link]

Information Centre Lagos Office

Speedwriting House, No.
20 Awolowo Road, Bodija, 16 Ajanaku Street, Off
Ibadan. Salvation Bus Stop, Awuse
Estate, Opebi, Ikeja, Lagos.

For technical issues (computer problems, web access, and

etcetera), please send mail to webmaster@[Link].

Academic Support

A course facilitator is commissioned for this course. You have

also been assigned an academic advisor to provide learning
support. The contacts of your course facilitator and academic
Help advisor for this course are available at
onlineacademicsupport@[Link]
Course overview The Meaning of Statistics

Activities

This manual features ―Activities,‖ which may present material

that is NOT extensively covered in the Study Sessions. When
completing these activities, you will demonstrate your
understanding of basic material (by answering questions)
Activities before you learn more advanced concepts. You will be
provided with answers to every activity question. Therefore,
your emphasis when working the activities should be on
understanding your answers. It is more important that you
understand why every answer is correct.

Assessments

There are three basic forms of assessment in this course: in-

text questions (ITQs) and self assessment questions (SAQs),
and tutor marked assessment (TMAs). This manual is
Assessments essentially filled with ITQs and SAQs. Feedbacks to the ITQs
are placed immediately after the questions, while the
feedbacks to SAQs are at the back of manual. You will
receive your TMAs as part of online class activities at the UI
Mobile Class. Feedbacks to TMAs will be provided by your
tutor in not more than 2 weeks expected duration.
Schedule dates for submitting assignments and engaging in
course / class activities is available on the course website.
Kindly visit your course website often for updates.

Bibliography

For those interested in learning more on this subject, we

provide you with a list of additional resources at the end of
this Error! No text of specified style in document.; these may be
books, articles or websites.
Readings

6
 PSY103 Quantitative Methods in Psychology

Getting around this Error! No text of specified style in

document.
Margin Icons

While working through this Error! No text of specified style in

document. you will notice the frequent use of margin icons. These
icons serve to ―signpost‖ a particular piece of text, a new task or
change in activity; they have been included to help you to find
your way around this Error! No text of specified style in document..
A complete icon set is shown below. We suggest that you
familiarize yourself with the icons and their meaning before
starting your study.

Activity Assessment Assignment Case study

Discussion Help Outcomes Reflection

Study skills Summary Time Tip

Study Session 1 The Meaning of Statistics

Study Session 1

The Meaning of Statistics

Introduction
Statistics is perceived and interpreted differently by different
professionals. This study session will therefore expose you to
alternative definitions and explanations of the concept of
statistics.

When you have studied this session, you should be able to:
 explain the concept of statistics, in grammatical terms,
and as used by scientific researchers. (SAQ 1.1)
Learning
Outcomes  explain the two basic types of statistics. (SAQ1.2)

1.1 The Definition of Statistics

Statistics is a branch of mathematics that has to do with the collection,
compilation, analysis, display, and interpretation of numerical data. It is
applicable to a wide variety of academic disciplines, including the
physical, social sciences, and the humanities; it is also used for making
informed decisions in all areas of research, business, marketing and
government.

o ITQ Statistics is the same as mathematics

 Yes
 No

Feedback on ITQ question

 The correct answer is B because statistics is only a branch of

mathematics

The word ‗statistics‘ can be used in three different perspectives:

1. Layman‘s perspective
 PSY103 Quantitative Methods in Psychology

2. Research perspective
3. Grammar perspective
1.1.1 Layman’s Perspective
As used in everyday language, statistics implies a collection of
numerical data. Whenever figures or numerical data are used to
represent observations, it is called statistics. This may be
expressed in terms of the number of students in a school, the
number of employees in a company, the number of children given
birth to in a year, the number of accidents recorded in a year and
so forth.
o ITQ In a lay-person‘s understanding, statistics
means:
 A. Addition, division, and subtraction of data.
 B. A collection of numerical data.
Feedback on ITQs answers
 The correct answer was B
 If you have chosen A (addition, division and
subtraction of data), then you have defined based on
mathematical experience in primary school. In fact
the question asks you to define statistics as a
collection of numerical data.
1.1.2 Research Perspective

 Statistics may also refer to the scientific methods for

collecting, organizing, describing, analyzing, presenting,
and interpreting data in line with the purpose for which it is
required to serve. Psychologists rely heavily on statistics to
help assess the meaning of the measurements they make.
Sometimes the measurements involve individuals who complete
psychological tests; at other times, the measurements involve
statistics that describe general properties of groups of people or
animals. In psychological testing, the psychologist may interpret
test results in light of norms, or the typical results, provided
from previous testing. Valid conclusions and reasonable
decisions are subsequently based on such analysis. Hence,
statistical results help researchers to make inferences about
occurrences in the environment.
o ITQ From the empirical research‘s point of view,
statistics means:
A. Organization of data for presentation purposes.
B. A collection of numerical data.
Study Session 1 The Meaning of Statistics

C. Purposeful use of scientific methods to collect,

organize, analyze and interpret data.
Feedback on ITQs answers
 If you have chosen A, then you have defined based
on part of this definitive point of view.
 If you have chosen B, then you have defined based
on a definitive perspective of layperson‘s view of
statistics. In fact, the question asks you to define
statistics as the scientific methods for collecting,
organizing, describing, analyzing, presenting and
interpreting data in line with the purpose for which it
is required to serve.
 The answer is C.
1.1.3 Grammatical Perspective
The word statistics can also be used as the plural of statistic (singular),
which refers to the result of applying a statistical algorithm to a set of
data, as in admission statistics, sales statistics, employment statistics,
accident statistics, etc.
. Statistic is used as a summary measure of a set of data, after
adequate analysis has been made. The mean or average, for
instance, is a statistic when used as a summary measure of
students‘ performance. When other summary measures like
mode, median, standard deviation and variance are considered
collectively, the word ‗statistics‘ is most adequate.
o ITQ In grammatical terms, the word ―statistic‖
implies:
A. A fact from a study of data.
B. A collection of numerical data.
Feedback on ITQs answers
 The correct answer is A. In grammar, statistic is the
summary of measured data after adequate analysis
has been made.
 Option B (a collection of numerical data) is a
definition based on a lay-person‘s perspective of
statistics. In fact the question asks you to define
statistics as a summary of a measured set of data,
after adequate analysis has been made.

1.2 Types of Statistics

In research parlance, psychologists use two types of statistics,
descriptive and inferential. These two types of statistics are considered
part of applied statistics. The main difference is that descriptive
 PSY103 Quantitative Methods in Psychology

statistics describe a data set as it is, while inferential statistics attempt

to make predictions, which goes beyond the values in the data set.

1.2.1 Descriptive Statistics

Statistical methods that are used to summarize or describe a collection
of data are called descriptive statistics. These are statistical methods
involving data collection that are modeled in a way that account
for randomness and uncertainty in the observations made in a research.
Descriptive statistics simply give a general picture of the scores in a
given group. This type of statistics summarizes or describes the
characteristics of a set of data in a clear and convenient way.
There are four kinds of descriptive statistics: measures of
frequency, measures of central tendency (also called measures of
center), measures of dispersion (also called measures of
variability or spread), and measures of position. Measures of
frequency are concerned with how many items there are in data sets.
These statistics include frequency, or counts, and relative frequency or
proportions. Measures of central tendency describe the value(s) which
the data set seems to be clustered about, while measures of dispersion
describe how spread out the data is. By considering the two together,
one can determine a "typical" value for the data set. They can also
know how far away from that typical value a data point is likely to be.
Measures of the position include percentile rank and quartile
rank (which is itself a subset of percentile rank).
Descriptive statistics mark the development of the discipline
called statistics. They include all calculations, which reduce data
to smaller sizes in which results of such calculations could as well
be presented in tabular form and diagrams or charts.
1.2.2 Inferential Statistics
Inferential statistics are used by psychologists to draw inferences, or
conclusions, from the data collected in a research. Data are analyzed
and interpreted in such a way that conclusion about the population
is made, based on data received on the sample. Inferential
statistical methods are devised to measure the degree of
confidence on how close an estimated sample statistic is to the
actual population parameter. This degree of confidence is
expressed in terms of probability.

The most common inferential statistical tests include the student's t-test
and the Analysis of Variance (ANOVA) or F-test; these statistics are
used to determine whether the average differences across comparison
groups are due to the effects of the independent variable under study.
Another commonly used inferential statistic is the correlation
coefficient, which describes the strength of the relationship between
Study Session 1 The Meaning of Statistics

two variables. For example, to determine if there is a positive

correlation between a student's score in PSY 103 and PSY 319, a
correlation statistic is used.

Descriptive statistics are considered as the preliminary

stage for inferential statistics. Results of inferential
statistics are what can be used in decision-making.
Examples of descriptive statistics are the measures of
central tendency (i.e. median, mode, and median),
measures of variability (e.g. standard deviation,
Tip variance etc.), frequency distributions, and transformed
scores. Inferential statistics include the t-test, analysis
of variance (ANOVA), chi-square test, etc.

o ITQ Look at the diagram below and fill the empty

boxes A and B with the most appropriate options.

Feedback on ITQs answers

 If you have considered inferential statistics as option
A then you are correct. Researchers use inferential
statistics if they are using their results to estimate a
population parameter
 Option B is in fact descriptive statistics. Researchers
use descriptive statistics if they are using the results
to describe the data they actually collected.
 PSY103 Quantitative Methods in Psychology

Study Session Summary

In this Study Session, the concept of statistics has been

approached from three perspectives. First, from a lay-person‘s
perspective that statistics is a collection of numerical data.
Summary Second, from a researcher‘s view that statistics is a scientific
method of collecting, organizing, describing, analyzing,
presenting and interpreting data. Third, in grammatical terms,
statistics is a plural form of statistic.

You also learnt about types of statistics which are: descriptive and
inferential statistics. The relationship between these types of
statistics was also highlighted.

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions. Write your answers in your Study
Assessment Diary and discuss them with your Tutor at the next Study
Support Meeting. You can check your answers with the Notes
on the Self-Assessment Questions at the end of this Manual.

SAQ 1.1 (tests Learning Outcome 1.1)

The definitive stands of statistics can be explained from
three points of views (layperson‘s view, grammatical view
and research view). What are the differences between these
perspectives?
SAQ 1.2 (tests Learning Outcome 1.2)
A clinical psychologist conducted a study in which she
gave some of her clients a new depression treatment. What
form of statistics would she use if she wanted to describe
the average depression score of only those clients who got
the treatment?
Study Session 1 The Meaning of Statistics
 PSY103 Quantitative Methods in Psychology

Study Session 2

History of Statistics
Introduction:
This Study Session is concerned with the history of Statistics. Most
times when statististics is taught, especially outside the scope of its
domain, not much is said about its history. A body of knowledge is
better appreciated when its history is known. It is on this basis that this
Study Session is carved out for knowledge

Learning Outcomes:
At the end of this Study Session, the learner would be able to achieve:
(a) Additional knowledge of the definition of statistics from
historilcal background
(b) Gain knowledge on the contribution of some individuals
in the origin of statistics
(c) Appreciate areas of historical background in the use of
statistics

2.1 Historical Definition of Statistics

The word statistics is derived from the modern Latin term statisticum
collegium ("council of state") and the Italian word statista ("statesman"
or "politician"), while the German word Statistik, was used to designate
the analysis of data about the state, also called the "science of state"
(then called political arithmetic in English).

2.2 Historical Background of Statistics

In the early 19th century, statistics acquired the meaning of the
collection and classification of data generally. It was considered as a
body of knowledge that is responsible of data collection and
organization. Statistics was introduced into English by Sir John
Sinclair. The original principal purpose of Statistik was to generate data
to be used by governmental and administrative bodies for the purpose
of decision making. The collection of data about states and localities
continued through national and international statistical services. For
instance, censuses provide regular information about the population.

During the 20th century, the field led to the creation of precise
instruments for public health concerns, especially in the area of
epidemiology, biostatistics, and the like, and economic and social
purposes like unemployment rate, econometry, etc. All these
necessitated substantial advances in statistical practices. After World
War I, the necessity for Western welfare states to develop a specific
knowledge of their "population" promoted statistical concerns.
Philosophers like Michel Foucault argued that these developments
Assessment: The Meaning of Statistics

constituted a form of "biopower", a term which has since been used by

many other authors.

Today the use of statistics has broadened far beyond its origins as a
service to a state or government. Individuals and organizations use
statistics to understand data and make informed decisions throughout
the natural and social sciences, medicine, business, and other areas.
Statistics is generally regarded not as a subfield of mathematics but as a
distinct, allied, field. Many universities maintain separate mathematics
and statistics departments. Statistics is also taught in departments as
diverse as psychology, education, and public health.

Assessment:
a) Apart from the three definitional perspectives of
statistics, provide an historical definitional perspective
b) Mention the names of some founding fathers of statistics
c) Mention areas of human endeavours where statistics
could be applied.
 PSY103 Quantitative Methods in Psychology

Study Session 3

Why we Study Statistics

Introduction
This Study Session aims to provide explanations on why students
of psychology and other related disciplines require the knowledge
of statistics.

When you have studied this session, you should be able to:
 justify the inclusion of statistics in the study of
Learning psychology. (SAQ2.1)
Outcomes

Rationale for the Inclusion of Statistics into Psychology

Quantification translation In the present scientific age, almost all branches of
to figures.
knowledge; arts or the sciences are involved in one type of
statistics or another . People in the behavioural sciences (like
Statistics a scientific psychology, sociology, economics, political science, and
method of collecting, education) require the knowledge of statistics for the
organizing, describing,
analyzing, presenting and following reasons:
interpreting data. 1. To understand professional literature like journals,
magazines and other periodicals the knowledge of
statistics is inevitable. Research reports published in
scientific journals and books are based on statistical
analyses. You will always find it difficult to understand
many pertinent articles in these sources without the
knowledge of statistics. Similarly, you will find it
difficult to plan your own research to meet scientific
standards without the knowledge of statistics.
2. For a scientific research to be taken serious, it must be
statistical. Inferences are drawn about a population based on
data obtained and analyzed from a sample. Thus, ignorance
of statistics will limit your comprehension of inferences
drawn in research reports, and you will not be skilled to
make genuine inferences from your research.
Study Session 3 Why we Study Statistics

3. As scientists, we require the knowledge of statistics to

progress in any quantitative scientific pursuit. To carry out
any significant behavioural science research, we have to
engage in statistical analysis. In conducting successful
research, we need to design the statistical analysis to test our
hypotheses, prior to the collection of our data. By so doing,
much information will be contained in our data and the
objective of the research will be achieved. Even when a
computer is to be used for analysis, a computer programmer
needs to be guided on the adequate statistics needed to test
hypotheses, which conform to the data-collected. When data
are wrongly collected, due to ignorance of the knowledge of
statistics, the goal of the research will be defeated.
4. People acquainted with statistics accept that research,
especially in the behavioural sciences, is mainly educated
guesses. In other words, there are no perfect right or wrong
answers in any research, and no single research has a
conclusive statement on any issue. The knowledge of
statistics thus helps us to evaluate the strengths and
weaknesses inherent in any research, in terms of the
techniques adopted by the researcher in collecting
information and drawing inferences.
5. The knowledge of statistics also helps us to be aware of our
limitations as human beings in conducting our personal
research. Sampling, research designs and statistical analysis
are devices employed to check these inherent defects, and
we are thereby warned of the dangers of invalid, biased or
over-generalized conclusions. Thus, with the knowledge of
statistics, we are aided to make less subjective conclusions.
6. The knowledge of statistics is indispensable to the
absorption of most written information around us, for
example in newspapers. In actual fact, data become more
meaningful when treated statistically. For instance,
comparisons and relationships are easily made when figures
or numerals are used, and they are much clearer than verbal
descriptions.
 PSY103 Quantitative Methods in Psychology

Study Session Summary

In this Study session you learned that statistics enables us to

progress in scientific pursuit because data become more
meaningful when treated statistically. The knowledge of
Summary statistics helps us to be aware of our limitations as human
beings in conducting our personal research. The knowledge of
statistics also helps us to evaluate the strengths and
weaknesses inherent in any research.

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions. Write your answers in your Study
Assessment Diary and discuss them with your Tutor at the next Study
Support Meeting. You can check your answers with the Notes
on the Self-Assessment Questions at the end of this Manual
SAQ 2.1 (tests Learning Outcome 3.1)
Identify at least four reasons why it is necessary for you to
study statistics?

So far, which of the provided points in SAQ3.1 will you

consider as most cogent and why? Or do you have a better
Discus reason or different perspective?
sion Post your response on UI Mobile Class forum for discussion.
Study Session 4 Common Terms used in Statistics

Study Session 4

Common Terms used in Statistics

Introduction
The teaching of statistics involves the understanding of concepts
that are used in scientific research. This study session will
therefore focus on scientific concepts that are used in statistics.

When you have studied this session, you should be able to:
 explain the meanings of concepts such as variables,
Learning data, population, parameter, and sample. (SAQ4.1)
Outcomes

4.1 Variable
Variables are any measurable feature, which has the potential of
taking different values when it is quantified. It includes such
human features like weight, sex, intelligence quotient (I.Q), etc.
These human characteristics can assume different values when
different individuals are evaluated, or when the same individuals
are evaluated on different occasions.
4.1.1 Classification of Variables
Variables in research are classified into three:
Independent Variables (IV)
Independent variables are those that are easily manipulated by
researchers so as to determine their effects on another set of
variables called the dependent variables. In fact, the independent
variables are the variables which the researcher determines its
variations to examine how the outcome variable is affected. In
behavioral research, the independent variable usually consists
of the two (or more) treatment conditions to which participants
are exposed. The independent variable consists of the
antecedent conditions that were manipulated prior to observing
the dependent variable.
 PSY103 Quantitative Methods in Psychology

Dependent Variables (DV)

Dependent variables change in values as a result of changes in the
independent variables. It is the values obtained on the dependent
variables that constitute the data needed for statistical analysis in a
research. The dependent variable constitutes the center of the
research problem, and therefore the focus of a research. It is what
attracts the interest of a researcher in a study.
o ITQ The IV (independent variable) in a study is
the
A. variable expected to change the outcome variable.
B. a outcome variable.
Feedback on ITQs answers
 A is correct.
 If you have chosen B, you are wrong. DV (dependent
variable) is the outcome variable that is used to
compare the effects of the different IV levels.
Extraneous Variables
These are also referred to as secondary, intervening, nuisance,
confounding, or concomitant variables. They are uncontrolled
variables that are unintentionally allowed to vary
systematically with the independent [Link] are
variables that have similar effects on the dependent variables,
like the independent variables, but they are not the
independent variables of concern.

Extraneous variables have to be checked in every research to

prevent their contaminating the dependent variable and,
thereby, providing alternative explanations for any change
observed. An example is a research designed to examine the
effect of noise on academic performance of students. The IV
Extraneous variables can is noise while the DV is academic performance. A possible
be likened to weeds in a extraneous variable is the students‘ level of intelligence, class
farmland. An okra of study, age, and the like. To determine the true effects of an
vegetable found in a maize
farm is regarded as a weed independent variable on a dependent variable, the effects of
even though it is edible; the extraneous variables have to be controlled, else we may
because it is not the wrongly be attributing an effect to the IV whereas it is the
originally planted crop.
extraneous variable that accounted for the change in the DV.
A Constant

A constant is a characteristic or condition that does not

vary, but is the same for every individual. In any
psychological research, apart from the IV and the DV,
Study Session 4 Common Terms used in Statistics

every other variable is expected to be held constant; hence

the extraneous variables should be held constant for all the
research participants so that the efficacy of the IV can be
determind. That is, any change in the DV can be attributed
to the change in the IV.

4.1.2 Measurement of Variables

Variables can be measured in two forms: discrete and
continuous measurements.
A discrete measurement of variables

This can take up only finite values of raw scores with no

fraction or decimals. A discrete variable consists of separate,
indivisible categories. No values can exist between two
neighboring categories. It could be countable data, such as
the number of students who registered for a course. It could
be 200 students but not 200.17 students, for example. A
discrete variable could also be the score on a psychological
variable like intelligence quotient (IQ), aggression,
satisfaction, etc. Data generated from the conversion of
qualitative data to quantitative data contain discrete values.
For instance, the values of 1 and 2 could be used to represent
male and female, respectively, in data coding. Similarly, 1, 2,
3, and 4 could be used to represent ethnic groupings like
Yoruba, Igbo, Hausa, and Others. Discrete values are in the
form of whole numbers such as 1, 20, 70, 200 etc..

A continuous measurement of variables

For a continuous variable, there are an infinite number of

possible values that fall between any two observed values.
A continuous variable is divisible into an infinite number of
fractional [Link] form of measurement can take up any
values within a given range. Thus, they accommodate
fractions or decimals. For instance, between the range of 1 and
2, we may have 1.33, 1.52, 1.75 to 2. Data on continuous
measurement normally come from interval or ratio scales.
These include measurements like height, weight, money, etc.,
which could be expressed in decimals or fractions like 5.2
metres, 70.3 kilograms or N7.40.

o ITQ The DV (dependent variable) in a study is

the:
 PSY103 Quantitative Methods in Psychology

A. variable expected to change the outcome variable.

B. an outcome variable.
Feedback on ITQs answers
 A is not correct. In fact a variable that is expected to
change the outcome variable is the manipulated
variable which is otherwise known as IV
(independent variable).
 B is correct.

4.2 Data
Data are information in Collections of numbers or measurements used to represent or
raw or unorganized form
(such as numbers or
quantify observations during research before transformation to
symbols).. other statistical forms is referred to as raw data. For instance,
students‘ scores in an examination can be used to represent
academic performance, and could be taken as data for the
research on effect of noise on academic performance.

Data is in a plural form, so we may say "the data are", not

"the data is". A data set is a collection of measurements or
observations. A datum (singular) is a single measurement or
observation and is commonly called a score or raw score.

4.2.1 Types of Data

Data can be grouped into two: quantitative or numerical data and
qualitative or non-numerical data.
Quantitative data arise from measurable characteristics of
objects such as height, weight, length, width, IQ scores, age,
anxiety level, income, and the like. On the other hand, qualitative
data arise from non-numerical characteristics of objects, such as
sex, religion, occupation, marital status, colour and some Yes or
No responses.
For the purpose of distinction, the numerical characteristics of
objects are referred to as variables, while the non-numerical
characteristics are called attributes. However, non-numerical data
could be converted to numerical data by counting the number of
cases falling in a given attribute category. This may vary from one
to infinity. By so doing, such values can also be called variables,
just like the numerical data. An example is sex as an attribute,
which may be expressed in terms of the number of males and the
Study Session 4 Common Terms used in Statistics

number of females. Sex could thereby be converted to a numerical

value like 1 and 2 respectively; this is now called a variable.

4.3 Population
This is also referred to as universe. It is ideally defined as a large
group of people, animals, objects, responses, or measurements
that are alike in at least one respect. Thus, we could have a
population of Yoruba students in the University of Ibadan, a
population of all ‗white rats‘ of a given genetic strain, the
population of undergraduate students in the University of Nigeria,
and the population of all federal government workers in Ibadan.
4.3.1 Parameter
This refers to the describable features of a population. They are
the features or characteristics that a collection of people or objects
possess that allow them to be called a population. For example, a
population of undergraduate students has the feature of grouping
all undergraduates, and thereby excluding the postgraduate
students. We may have the male population in Ibadan also,
comprising all male persons in Ibadan.
4.3.2 Sample
Representative Sample: A This is a subset of a population, usually drawn from the
subset of a statistical
population that accurately
population randomly or otherwise, for the purpose of
reflects the members of the generalization to the population. In the behavioural sciences
entire population. and education, it is not often easy to study the entire
population of interest. Consequently, only a part of the
population is taken for study and this part is representative of
the entire members in the population under study. The part
selected for a study is the sample. When the sample is selected
in scientific ways, all the characteristics of the population are
believed to be reflected in the chosen sample. Such sample is
then referred to as a representative sample of the population.
In a class of 60 students, 10 of them could be selected as a
sample if the selection is done randomly. We will discuss the
concept of sampling in the next Study Session.

4.3.3 Sampling Error

Sampling error is the discrepancy or amount of error, that

exists between a sample statistic and the corresponding
population parameter. A sample is presumed to represent a
population, hence any statistic obtained from a sample could be
used to generalize to the population. Statistical error exists
 PSY103 Quantitative Methods in Psychology

where the sample statistic differs from the population statistic;

this difference is termed sampling error.
Study Session 4 Common Terms used in Statistics

So far, we have discussed some statistical terminologies. Now

work through the following activity, referring to this Study
Session. To complete the activity, you need a small collection
of different types of books in your office or home (about 20
Activity 3.1 will do)
Allow 15 minutes
1. What data do you want to collect?
2. If you collected data initially as raw data, how would you
record this data?
3. Make a tally of your data – use the table below. Add lines
as necessary.

Major Colour of Front Tally Frequency

cover

4. On the table above, as indicated in the third column, record

the frequencies of the colours (answer in same table).
5. Is it necessary to group your data? Explain your answer.
6. Can you find the range for this data? Explain your answer.
7. What form of data collection did you use?
8. Did you use random selection to find your data?
9. What is the population of the data that you have collected?
10. Did you need to choose information from a sample when
you collected your data?
NOTE
You may refer to Statistics (pages 1-3), an additional reading
material in the bibliographic section for further help. This
material is free and available for downloading at your course
website (UI Mobile Class).
 PSY103 Quantitative Methods in Psychology

Once you have collected and recorded data in a table like you
have in questions 3 and 4 of the activity above, you have a
frequency table of your data. If you look at the frequency
table, you can start to answer descriptive and interpretive
Activity 3.2 questions about your data such as the questions in the
Allow 5 minutes following activity.
1. Which colour was the most common in your collection of
books?
2. Which colour was the least common in your collection of
books?
3. Do you think that another collection of books would have
the same numbers of different colours as you have found?

Study Session Summary

In this Study Session, we defined and provided explanations

on some statistical concepts. Anything that can change in
value or form is referred to as a variable; it could exist as
Summary independent, dependent or extraneous variable. Measurement
of variables could be expressed in discrete or whole number
form, as well as in continuous form or measures with decimals
or fractions. A collection of objects with similar features is
called a population. The features which the members of a
population possess that characterize their being called a
population are referred to as parameter. A subset that is drawn
from a population is a sample. Numbers or a measurement
used to represent or quantify observations during research is
called data, which could be qualitative or quantitative.

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions. Write your answers in your Study
Assessment Diary and discuss them with your Tutor at the next Study
Support Meeting. You can check your answers with the Notes
on the Self-Assessment Questions at the end of this Manual
SAQ 4.1 (tests Learning Outcome 4.1)
Study Session 4 Common Terms used in Statistics

Carefully study the scenario in the short passage below, and fill in
the blank spaces.
Mr. Biodun, a researcher studying depression gave a new
treatment to a sample of 100 people with depression out of
a____A____, a group of all things that share a set of
characteristics. In this case, the ―things‖ are people, and the
characteristic they all share is depression.
Mr. Biodun wants to know what the mean depression score for the
population would be if all people with depression were treated
with the new depression treatment. In other words, he wants to
know the____B___, the value that would be obtained if the entire
population were actually studied. Of course, Mr. Biodun don‘t
have the resources to study every person with depression in the
world, so he must instead study a____C___, a subset of the
population that is intended to represent the population. In most
cases, the best way to get a sample that accurately represents the
population is by taking a ___D___ from the population, when
each individual in the population has the same chance of being
selected for the sample.
So, he then uses the sample statistic value as an estimate of the
population parameter value. When researchers use a sample
statistic to infer the value of a population parameter it is called
___E___.
 PSY103 Quantitative Methods in Psychology

Study Session 5

Sampling

Introduction
This Study Session will expose you to the meaning of sampling,
and the various techniques of conducting sampling, specifically in
social science research setting .

When you have studied this session, you should be able to:
 use sampling technique. (SAQ 4.1)
Learning  differentiate between probability and non-probability
Outcomes sampling techniques. (SAQ 4.1 and 4.2)

5.1 What is Sampling?

Sampling is the process of taking or selecting any portion of a

population or universe, scientifically, as representative of that
population or universe. A representative sample is one which has
all the characteristics of the population under study. A selection
technique in which every member of the population has equal
chance of being selected is called random sampling. Random
selection, or random sampling, is a process for obtaining a
sample from a population that requires that every individual in
the population has the same chance of being selected for the
sample. A sample obtained by random selection is called a
random sample.

On the other hand, a selection technique in which the members of

the population do not have equal chances of being selected is
called, biased sampling; each member has a higher or lower
probability of being selected than others.

5.1.1 Important Concepts in Sampling

Sample Frame: By sampling frame, we refer to a list of items in
the population from which the sample is drawn. For instance, the
Study Session 5 Sampling

names of all undergraduate students in the University of Ibadan

could stand as a sampling frame if the population is defined as
undergraduate students of the university. To select the sample
randomly, we need a complete sampling frame.
Sample Size: It is the total number of items in the population that
is actually selected for a research. The sample size in the example
given may be 200 undergraduate students of the University of
Ibadan. The sample size desirable for any given research is
determined by the sampling frame, the reliability of results
expected, and costs of managing the sample. Generally, however,
the greater the sample size, the higher its reliability and accuracy,
but the higher its costs. This is because a larger sample size
increases the representativeness of the population characteristics,
but it could be more difficult to manage, in terms of conducting
the research. Some formulars have been devised to scientifically
calculate the sample size adequate for a population size; one of
these is the Yamane‘s formula.

5.2 Sampling Techniques

These are the methods used to select a sample for a research.
There are two main techniques for selecting samples; these are
probability and non-probability sampling.
Probability Sampling: In this case, some forms of random
sampling are used in one or more of their stages.
Non-probability Sampling: Non-probability techniques do not
use random sampling techniques.
5.2.1 Probability Sampling Techniques
A. Simple Random Sampling
By this technique, all members of interest have equal probabilities
or chances of being selected. A common method adopted is the
use of table of random numbers. The table of random numbers is
generated by the computer and contains decimal digits of 0 to 9.
In usage, the researcher should first prepare the sampling frame
by listing all members of the population in alphabetical order, and
number them serially. The sample size is decided before entering
the table of random numbers.
To start the selection, the researcher enters the table at any point,
without bias. To ensure objectivity, the researcher may close his
or her eyes and just touch any number with the tip of a pencil. The
random number digits can be read sideways (horizontally), or up
and down (vertically) from this starting point. The number of
 PSY103 Quantitative Methods in Psychology

random digits to select depends on the members of the population

in the sampling frame. If they are between 1 and 99 then two-digit
random numbers are selected at a time, from the table. The serial
numbering of the members of the population in the sampling
frame would range from 01 to 99, to reflect what would be
observed on the table of random numbers. A frame having
members up to 999 requires the use of three-digit random
numbers. Thus, the serial numbers should range from 001 to 999.
A selected random number is taken as the serial number of the
corresponding member of the sample frame in the population, and
by inference that member is included in the sample as a
participant. This exercise continues until the total number of
participants desired for the research is exhausted. When a number
greater than any in the frame is selected, it is discarded and the
process continues until the sample size required is obtained.
However, rather than using this manual method of selection, the
computer could be used to do the random selection of numbers.
Another method of simple random sampling is ballot. By this
method, members of the population are also arranged
alphabetically and numbered serially. Numbers are then written
on pieces of papers to cover all the numbers in the population.
Each number that is written on a slip of paper is folded, and kept
in a container. After all the papers are folded and placed in the
container, the papers are shuffled after which selection of papers
from the container is made, to cover the sample size. Each
selected number is taken as the serial number of the
corresponding member of the population; the member is thus
included in the sample.
Coin flipping, coin tossing, or heads or tails is another method of
random selection. It is the practice of throwing a coin in the air and
checking which side is showing when it lands, in order to choose
between two alternatives, head or tail. When you toss a coin there are
two possible outcomes: 'head' or 'tail'. Each of these outcomes is
equally likely, since there is 1 'head' and 1 'tail' on the coin. So when
you flip a coin, there's a 50/50 chance of getting heads. If you flip it
once and get heads, the second time you flip the coin, there's a 50/50
chance of getting heads.

Coin flipping is a form of selection which inherently has two possible

outcomes. It is sometimes used to resolve a dispute between two
parties, or used to assign sample group into sub-groups. It could be
used to assign samples to experimental and control group. After a
sample is divided into two, they could be assigned head or tail. When
the coin is flipped, which ever that turns up will determine which group
is assigned the experimental group while the other is assigned the
Study Session 5 Sampling

control group. The toss of a coin has been a method used to determine
random outcomes for centuries. It is still used in some research studies
as a method of randomization, although it has largely been discredited
as a valid randomization method. In football matches, it is used to
randomly determine what team starts the kick.

B. Stratified Random Sampling

In stratified random sampling, the population is divided into
subgroups, called strata. Each stratum is defined by unique
characteristics of each subgroup in the population. In a population
we may have our strata as males and females, blacks and whites,
junior employees and senior employees, etc. Within each stratum,
simple random sampling is used to select participants who will be
included in the sample, hence, it is called stratified random
sampling.
This sampling technique is used mostly when a researcher notices
that some peculiar features of the population have to be
adequately represented in the sample, which may not be achieved
if the simple random sampling technique only is used. The simple
random sampling may not reflect the heterogeneity of some
populations. By dividing a population into distinct subgroups or
strata, with each stratum differing from others, internal
homogeneity within each stratum is ensured. Employing the
simple random selection technique in each stratum presents a
sample which reflects the heterogeneity of the population. Such
sample is, therefore, a better representative of the population.
To improve on the heterogeneity of the population while adopting
the stratified random sampling, the researcher may decide to
select equal number of participants from each stratum or the
selection may reflect the ratio in which strata exist in the
population. For instance, regardless of the number of males and
females in the sample frame, a researcher may decide to select
just 20 males and 20 females as the sample. On the other hand, if
he/she had 300 males and 100 females in the sample frame, this
gives a ratio of 3:1. In selecting participants for a sample size of
40, using the simple random sampling in each stratum, he/she
could select ¾ x 40 i.e. 30 males and ¼ x 40 i.e. 10 females. The
40 participants needed will, therefore, mean 40 males from the
300 available and 10 females from the 100 available.
However, the stratified random sampling is not adequate where a
sampling frame does not exist or cannot be easily compiled. For
instance, having a sampling frame for the population of Nigerian
 PSY103 Quantitative Methods in Psychology

youths may be rather difficult. Also when a sample frame exists

but members of the population are so scattered that one may not
be able to clearly spell out a stratum as distinct from another, then
the stratified random sampling is not adequate.
C. Cluster Sampling

Cluster sampling is a method of probability sampling that is often used

to study large populations, particularly those that are widely
geographically dispersed. The researcher creates multiple clusters of
people from a population where they are indicative of homogeneous
characteristics and have an equal chance of being a part of the sample.
Random groups are thereafter selected with a simple random or
systematic random sampling technique for data collection and data
analysis. There are two ways to classify this sampling technique. The
first way is based on the number of stages followed to obtain the cluster
sample, and the second way is the representation of the groups in the
entire cluster analysis.

In most cases, sampling by clusters happens over multiple stages. A

stage is considered to be the step taken to get to the desired sample.
This can be divided into single-stage, two-stage, and multiple stages.
An example of single-stage cluster sampling is a case of the Ministry of
Education proposing to create a sample of girls across five villages in a
Local Government Area (LGA) to provide education for them. Using
single-stage sampling, the Ministry randomly selects villages (clusters)
in a LGA to form a sample and extend help to the girls deprived of
education in those villages. In the two-stage cluster sampling technique,
instead of selecting all the elements of a cluster, only a sample is
selected from each group by implementing systematic or simple
random sampling. An example of two-stage cluster sampling is in the
case above. The Ministry may randomly select towns from the LGA,
and then select random samples from each of these towns (clusters).
For the multiple-stage cluster sampling, the multiple-stage
sampling technique is used. Here the researcher randomly selects states
in Nigeria, from the selected states LGAs are selected randomly from
each state, from the selected LGAs towns are randomly selected then
persons are randomly selected from the selected towns. From towns, ,
streets, family, individual household, and individuals in a
household may be subsequently selected.
D. Systematic Sampling
Systematic sampling entails the random selection of the first
person and successive persons at a particular interval, called the
nth interval. It starts by the alphabetical arrangement of all
members of the population and serial numbering of these. By this,
the total numbering of the sample frame is determined. The
Study Session 5 Sampling

researcher then decides on the sample size for his/her research. To

determine the nth term for random selection, divide the population
size by the sample size desired. Thus, if the population size is 100
and the sample size desired is 10, then the nth term becomes
100/10 i.e. 10. In selecting participants for the sample, every 10th
person in the sample frame is selected. The researcher enters into
a ballot to decide on where to commence the pick, between 1 and
10. If 10 was picked, t then the order of persons selection will be
the 10th, 20th, 30th, 40th, 50th, - - - 100th in the frame. This will
finally give us a total of 10 participants as the sample size. If 6
was picked in the ballot, the researcher will select the 6th, 16th,
26th, 36th, - - -96th persons as the sample for the research.
5.2.2 Non-probability Sampling
A. Accidental Sampling
It is the weakest form of sampling but the most frequently used in
survey research. One simply takes available samples at hand,
rather than embark on the laborious process of obtaining a
probability sample. A good example is the distribution of research
questionnaire indiscriminately to people as they are found. Also
are interviews of persons on the street by news reporters.
This technique is used when both time and money are limited. It
is quick and easy to conduct accidental sampling. However, there
is limitation of generalizing the result of the sample to the entire
population; the results and interpretation are limited to the sample.
B. Quota Sampling
Quota sampling is the most systematic and scientific of the three
non-probability sampling techniques. It involves dividing the
population into a number of segments, just like the stratified
random sampling. However, unlike the stratified random
sampling, in this case, the researcher arbitrarily selects a quota
sample from each segment. Thus, selection is not random so quota
sampling does not ensure the selection of a representative sample.
Despite this shortcoming, quota sampling remains an attractive
alternative to stratified sampling.
C. Purposive Sampling
Purposive sampling is also known as judgmental, selective, or
subjective sampling. It is a form of non-probability sampling in which
researchers rely on their own judgment when choosing members of the
population to participate in a research. In this method, researchers are
required to have prior knowledge about the purpose of their research so
that they can properly choose and approach eligible participants for the
research. Researchers use purposive sampling when they want to access
a particular subset of people, as all participants of a survey are selected
 PSY103 Quantitative Methods in Psychology

because they fit a particular profile. The researcher sets some criteria
based on the needs of the research, and selects from the population
those who fit into the set criteria. A researcher might, for example,
want to select his sample for a research; entrepreneurs who have a
minimum size of 10 employees. While the findings from purposive
sampling do not always have to be statistically representative of the
greater population of interest, they are qualitatively generalizable. The
more prior information researchers have about their particular
population of interest, the better the sample that they are going to
select.

D. Convenience Sampling
Convenience sampling is a non-probability sampling method where
participants are selected for inclusion in the sample because they are
the easiest for the researcher to access. This can be due to geographical
proximity, availability at a given time, or willingness to participate in
the research. Convenience sampling involves using respondents who
would be ―convenient‖ for the researcher to access. The researcher
leverage individuals that can be identified and approached with as little
effort as possible. Although there is no pattern whatsoever in acquiring
these respondents, but they may be recruited by merely asking people
who are present in the street, in a public building, or in a workplace, for
example, to participate in the research.

The concept is often confused with ―random sampling‖ because of the

notion that people are being stopped ―at random‖ (in other words,
haphazardly). However, whereas the correct definition of random
sampling (using random numbers to pick potential respondents or
participants from a sampling frame) generally results in a statistically
balanced selection of the population, a convenience sample has an
extremely high degree of bias.

E. Snow Ball Technique

Snowball sampling or chain-referral sampling is defined as a non-

probability sampling technique in which the samples have uncommon
traits. By this we mean that the sampling method is extensively used
where a population is unknown and rare and it is tough to choose
participants and assemble them as samples for research. The sampling
technique operates in a way, in which existing participants provide
referrals to recruit samples required for the research study. The
Snowball sampling is a form of selection by recruitment, whereby
research participants are asked to assist researchers in identifying other
potential participants; a participant nominates another person with the
same trait that is being studied in the research. The researcher then
observes the nominated persons and continues in the same way that
they nominate others until the required number of participants is
gained.
Study Session 5 Sampling

If you are studying the level of quality of life of HIV/AIDs

victims among the members of the population, for example, you will
find it extremely difficult to collect primary data sources unless a
contacted victim agrees to have a direct conversation with you and
provides the contact details of the others they know. The sampling
method involves a primary data source nominating other potential data
sources that will be able to participate in the research. Thus, snowball
sampling method is purely based on referrals and that is how a
researcher is able to generate a sample, hence this method is also called
the chain-referral sampling method. The sampling technique can go on
and on, just like a snowball increasing in size (in this case the sample
size) till the time a researcher has enough data to analyze, to draw
conclusive results that can help the researcher make informed
decisions.

o ITQ Which of the following sampling techniques

is more objective than the others?
A. Cluster sampling
B. Simple random sampling
C. Purposive sampling
D. Accidental sampling

Feedback on ITQs answers

 The correct answer is A because cluster sampling
involves random sampling within specialized group
clusters.
 If you chose B, you would be wrong because the
simple random sampling may not reflect the
heterogeneity of some populations.
 If you chose C, or D, you would be wrong.
Purposive, and accidental sampling techniques are
non-probability sampling, and do not allow for
scientific objectivity. They do not provide all
members of the sample frame equal probability or
chance of being included in the sample size.
 PSY103 Quantitative Methods in Psychology

Study Session Summary

In this Study session you learned that sampling is the
process of taking a representation from a population as
research participants. Random sampling was explained as
Summary
a case where every member of a population has equal
chances of being included in the sample. This was
differentiated from biased sampling where members of the
population do not have equal chances of being selected. A
sample frame was differentiated from sample size; the
former being the total units in the population, while the
latter is the number of participants selected in a study. Two
broad sampling techniques were differentiated; probability
and non-probability sampling techniques. While the
former adopts some forms of random selection, the latter
does not. Types of probability sampling techniques
presented include simple random, stratified, systematic,
and cluster sampling; while non-probability techniques
include accidental, quota, and purposive sampling.

Assessment
Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
Assessment answering these questions. Write your answers in your Study
Diary and discuss them with your Tutor at the next Study
Support Meeting. You can check your answers with the Notes
on the Self-Assessment Questions at the end of this Manual
SAQ 5.1 (tests Learning Outcome 5.1)
A researcher desires to select 200 participants for a research to
examine the relationship between sex and academic performance
at the University of Ibadan. She however discovers that the ratio
of males : females in the institution‘s sample frame is 3:2. What
number of males : females should be selected from this population
for the study?
A. 150:50
B. 80:120
C. 120 : 80
D. 50:150
SAQ 5.2 (tests Learning Outcome 5.1 and 5.2)
Assuming that a population size is known to be 600 and the above
researcher wishes to systematically sample 20 participants from
that population. Determine the nth term of the random selection
for the sample.
A. 20
B. 30
Study Session 5 Sampling

C. 600
D. 400
Find the first and last individuals that would be included in
the sample in SAQ 5.2. Post your answer on Study Session
4 Assignment Page at the UI Mobile Class.
Assignment
 PSY103 Quantitative Methods in Psychology

Study Session 6

Scales / Levels of Measurement

Introduction
Measurement scale is very pivotal in determining what statistical
analysis would be appropriate during an analysis. This Study
Session is therefore designed to acquaint you on various forms in
which data can be expressed in research.
When you have studied this session, you should be able to:
 define , use correctly and differentiate all of the key
words printed in bold. (SAQ 6.1)
Learning  Nominal scale
Outcomes
 Ordinal scale
 Interval scale
 Ratio scale
To put data of a variable into statistical use, they are recorded in
some systematic ways. The procedure of describing the variables
is called scaling, and the descriptions which emerge are called
scales. The scales on which variables are measured have different
properties.
Scales of measurement are also referred to as levels of
measurement. This is because the numbers used to represent some
variables have more meaning than those used to represent other
variables.
Four scales are conventionally used in converting observations
into numerical data. These are nominal, ordinal, interval and ratio
scales.

6.1 Nominal Scale

This is the simplest and weakest level of measurement. By this
method, observations of variables are classified into defined
groups or categories, and each group or category is given a name
or label for identification purposes. As a rule, each observation
must fall under only one category,. All observations that have the
Study Session 6 Scales / Levels of Measurement

features of a defined category are grouped into that category and

given the appropriate label for the category. Observations that are
not of the same features are grouped into different categories and
given different labels. Labels used to describe the categories could
be names like males or females, for sex variable. They could as
well be numerical numbers like ‗1‘ for males and ‗2‘ for females.
The numbers used here are, however, used as labels merely to
classify observations into different categories, or used to describe
different categories. They do not imply that one category is
numerically related to any other, or numerically higher than the
other. Thus, the fact that females are represented with ‗2‘ in the
example above and males with ‗1‘, does not imply that females
are superior to males; it could be numbered the other way round.
Another example is the numbers used to represent footballers on
the football field, without any number indicating level of
performance. In fact, such numbers used in nominal scaling do
not imply order of importance. Invariably, this scale does not
permit any arithmetic computation. Examples of variables in the
nominal scale form are sex, religion, ethnic group, vehicle
number, hair colour, name of organization, marital status, state of
origin, etc.

6.2 The Ordinal Scale

While nominal scales only classify, ordinal scales classify and
order the classes in order of superiority. Ordinal scales, thus,
indicate the order of observations on a particular attribute but do
not show their separations on the scale. In other words, ordinal
scales permit discussion of ―more than‖ or ―less than‖, but the
extent to which an observation is more than, or less than another
is not evident from the scale. In this case too, numbers could be
used in ordering or ranking observations on the degree of
importance or superiority, but the numbers do have some
quantitative meanings, unlike the nominal scales where numbers
are mere labels.
An example of ordinal scale is the positioning of students on their
examination performance. Numbers could be used to indicate the
relative performance of students in terms of 1st, 2nd, 3rd and so
forth. It is an ordinal scale because the student occupying the first
position performed better than the others, the second student
performed better than others, except the first student. However,
the numbers do have quantitative meanings. But we cannot say
ideally that the actual difference in performance between the 1st
and the 2nd student is equivalent to the difference in performance
 PSY103 Quantitative Methods in Psychology

between the 2nd and 3rd student. Although ordinal scales permit
the use of numbers with quantitative meaning, the numerals
employed are non-quantitative in that arithmetic computations of
addition, subtraction, multiplication and division are not
computable, just like the nominal scales. Variables expressed in
ordinal terms include job status, class levels at school, level of
education, etc.
o ITQ Religious grouping is what type of
measurement?.
A. Nominal scale.
B. Ordinal scale.
Feedback on ITQs answers
 The correct answer is A, because religious grouping
is just a descriptive categorization.
 If you chose B, you would be wrong because ordinal
scales are not mere categories, they also show
relative size.

6.3 Interval Scale

This is the highest scale or level of measurement usually attained
in the behavioural sciences. Interval scales have all the qualities
of ordinal scales and an added quality. Not only do they represent
the ordering of observations on the characteristic being measured,
they also represent the relative separation of items in the scale.
Equal intervals between any two pairs of observations are equal.
For instance, the distance or interval between 9 and 10 in the
interval scale is the same as that between 112 and 113 or between
any two adjacent observations.
With an interval scale, we can say precisely how much bigger,
smaller, or better an observation or event is, compared with
another. Thus, with nominal scale, we can say that today is hot
while yesterday was cold. With ordinal scale, we can say today is
colder than yesterday. In terms of interval scale, we can say that
today is 15o colder than yesterday. Interval scales permit the use
of numerical values, which are quantitative, hence the operations
of arithmetic computations such as adding, subtracting,
multiplying and dividing are allowed. We can therefore say that
the average of 9 and 10 is 9.5 without violating the properties of
the attribute being measured. To achieve this, measurements
must be based on some agreed units such as degrees, seconds,
number of responses and so forth. It is the unit that permits us to
equate intervals between points on the scale. Examples of
Study Session 6 Scales / Levels of Measurement

variables expressed in interval measurement scale are intelligence

quotient (IQ), anxiety, personality, aggression, etc

6.4 Ratio Scale

This scale has all the qualities of the interval scale with some
additional qualities. That is, only ratio scale permits the making
of statements that relate to ratio of numbers in the scale. For
instance, 6cm is to 3cm as 2cm is to 1cm so they both have ratio
2:1. We can then meaningfully say that one is twice as good as
the other. Ratio scale is mostly adequate for the measurement of
physical objects or variables like height, weight, length, and so
forth. Variables measured in the behavioural sciences are mostly
in interval form. So if a student had 80% in a course and another
40% in the same course, we may not be right to say that the
former is twice as intelligent as the latter on the course. This is
because performance in examination in this form is measured in
interval scale. But we can convincingly say that a father is twice
as tall as his son because height is in ratio scale.
The ratio scale also employs a true or absolute zero mark, whereas
the interval scale permits an arbitrary zero mark. With ratio scale
a length of 0cm or time of 0.00 seconds implies absence of the
variables being measured; in other words, zero is real. But with
interval scale, if a student scores 0% in statistics, for instance, it
does not necessarily imply that the student does not have any
knowledge of statistics. In statistics, however, there is no clear-
cut difference in the treatment of the interval and the ratio scales.
All ratio scales are interval scales; all interval scales are
ordinal scales, and all ordinal scales are nominal scales.
But all nominal scales are not ordinal scales; all ordinal
scales are not interval scales, and all interval scales are not
Tip ratio scales.
o ITQ A measuring scale that permits you to
determine whether a data is greater than another but
does not imply that an individual can score nothing
may be referred to as which one of the following scale
types?
A. Nominal scale.
B. Ordinal scale.
C. Interval scale.
D. Ratio scale.
Feedback on ITQs answers
 PSY103 Quantitative Methods in Psychology

 The correct answer is C because interval scales tells

you the extent of difference between two individuals;
and does not confer absolute zero on the individual
that scores zero.
 Option A is a mere categorization scale. Although
option B can enable you to determine qualities of
superiority, it is also a mere categorisation scale.
Option D provides an absolute zero

Study Session Summary

In this Study session you learned that the process of

describing variables for statistical use is called scaling. When
variables are classified into defined groups or categories and
Summary each group or category is given a name or label for
identification purposes, it is called a nominal scale. Ordinal
scales exist where observations on a particular attribute are
expressed in order of importance. Measures of behaviour are
expressed in interval forms, while physical measures are in
ratio forms; both allow for equal distances between any two
pairs of observations

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions.
Assessment

SAQ 6.1 (tests Learning Outcome 6.1)

Fill the empty columns in the table below with appropriate terms

Scale of Uses of the Scale Example

Measurement
Study Session 6 Scales / Levels of Measurement

Pets: 5 dogs, 12
cats, 7 fish, 2
hamsters
RANK some
things as having
more of
something than
others (but NOT
QUANTIFY how
much of it they
have)
Interval
QUANTIFY how
much of
something there
is and a score of
zero means the
absence of the
thing being
measured
 PSY103 Quantitative Methods in Psychology

Study Session 7

Frequency Distribution

Introduction
In trying to understand, interpret, predict, and modify behaviour,
psychologists engage in various forms of research that involve
data collection. Such data represent observable behaviours that
are recorded in numerical forms. This study session will expose
you to how to summarize data into manageable forms, using
frequency distributions.
When you have studied this session, you should be able to:
 Demonstrate how data can be put in a simple
frequency distribution pattern. (SAQ7.1)
Learning  construct class intervals, and enhance the use of tally
Outcomes marks with subsequent frequency values. (SAQ7.2)
 determine exact limits and mid-point values in
grouped frequency distribution. (SAQ7.2)

7.1 Regular Frequency Distribution

Often times when research works are conducted by economists,
educationists, psychologists, political scientists or sociologists,
the dependent variable of interest is measured and represented
with numerical numbers. Each participant‘s score on the
dependent variable is thus recorded, leaving the research with a
mass of unordered data for the whole research participants. These
data are of little or no meaning until they have been organized and
classified in a systematic way that will allow the introduction of
statistical analysis like the average, standard deviation, and
variance.
After raw data are collected, they are placed in a summarized form,
called frequency distribution. A frequency distribution, often the first
organizational step, is an ordered arrangement of all observations in a
research, which shows the number of occurrences in each category.
Suppose that a training manager wants to evaluate how well his
trainees have performed at the completion of a training exercise in
accounting. He gave his trainees a test and their scores fall
Study Session 7 Frequency Distribution

between 0 and 10. Look at the scores for the group of 50 trainees
in Table 6.1.

Table 7..1 Number of Problems solved correctly by 50 Trainees

5 8 8 7 10 8 5 4 6 9
8 7 3 9 5 3 1 5 6 6
6 4 4 9 9 8 6 7 5 5
9 7 8 5 6 2 5 5 4 3
8 6 5 7 5 5 4 4 10 2
Frequency Distribution As you can observe, the data in Table 7.1 are unorganized or
An organized
tabulation/graphical
unordered and are, therefore, confusing and difficult to
representation of the comprehend. In statistical terms, they are called raw or
number of occurrence in unclassified data. Items or raw data become more
each category on the scale
of measurement.
comprehensible when classified. The resulting table obtained,
after classifying raw data, is called a frequency distribution.
When the number of different scores in the raw data is not too
large, the regular frequency distributions are an excellent way
of summarizing a set of data. The scores in Table 7.1 range
between 1 and 10, and it gives the number of different scores .
This is not too large, so regular frequency distribution can be
used to summarize the raw data, as shown in Table 7.2.

Table 7.2 Regular Frequency Distribution for Data in Table 7.1

Scores Frequency
Tally Marks
(X) (f)
10 || 2
9 |||| 5
8 |||| || 7
7 |||| 5
6 |||| || 7
5 |||| |||| || 12
4 |||| | 6
3 ||| 3
2 || 2
 PSY103 Quantitative Methods in Psychology

1 | 1
0 0
Σ =N=
50
7.1.1 Steps in Constructing a Regular Frequency
1 List every raw score value in the first column of a table and
denote this by the symbol [X], with the highest score at the
top and the lowest score at the bottom. Thus, 10 is at the top
while 0 is at the bottom.
2 Run through the raw data either horizontally or vertically,
recording each observed score against identical score in the
first column with a tally mark. Show the tally marks in the
second column. After every four tallies, the fifth tally is
used to cross the previous four (it is a rule). It is aimed at
easing the recording of frequencies.
3 When the tally is completed, the tally marks for each score
in the first column are added up to find the frequency (f) and
recorded in the third column. The frequency of each score
shows the number of times a given score was obtained. You
can easily read from Table 7.2 that two students had a score
of 10, five students had a score of 9, seven students had a
score of 8, etc. Cross check with Table 7.1
4 Add up the frequencies and place the total at the bottom of
the frequency column. The Greek sign Σ (Sigma) means the
―sum of‖ and Σf means ―the sum of frequencies‖ is used to
show this. It means the total number of scores in the sample.
Σf and N must be equal, they are therefore used to check
error in tallying. Thus, if Σf and N are not equal, the tallying
exercise should be repeated. Even when they are equal, it is
still advisable to repeat the exercise to check for error of
placing one or more tally marks wrongly. If the first tallying
was done vertically, the second may be done horizontally to
check for such error.
Study Session 7 Frequency Distribution

You have learnt that a frequency distribution is a convenient

way to organize a set of data. You have also been exposed to
the steps of constructing a frequency distribution table.
Activity 7.1 You will now prepare a frequency distribution for "the heights
of 24 students" under this guided activity.
Allow 15 minutes
The heights in inches of 24 students are as follows:
66, 68, 65, 67, 64, 68, 64, 66, 64, 69, 69, 64, 67, 63, 63, 68,
67, 65, 69, 65, 67, 66, 69, 67
Prepare a frequency distribution
Solution: Step 1: Make three columns.

Step 2: List the heights (in order of size) in the first column,
the tally in the second column, and the frequency in the third
column.

Heights Tally Frequency

63 |||| 2
64 || 4
65 ||| Fill the
66 remaining
columns
67 accordingly
68
69

Post your findings on Activity 7.1 Page at the UI Mobile

Class.

7.2 Grouped Frequency Distributions

When data obtained in a research are relatively more, the regular
frequency distribution might not be adequate in summarizing the
data; rather, the grouped frequency distribution becomes
necessary. This calls for the grouping of raw data into class
intervals and the number of scores in each class interval is then
 PSY103 Quantitative Methods in Psychology

recorded as frequency. An interval in which scores are grouped is

called a class interval.
Suppose a production manager is interested in the production
efficiency of employees in his department. He selected 50
members of his department and recorded their performances in
terms of number of cartons packaged in one hour, as presented in
Table 7.3.
Table 7.3 Performance Data of Employees in a Packaging Firm
You can see that the scores
in Table 7.3 range between 63 53 62 40 54 58 61 68 50 67
19 and 75. Therefore, the
grouped frequency 48 46 74 26 24 19 23 27 32 62
distribution should be
constructed to summarize 36 20 68 45 35 43 41 45 33 30
the data as shown in Table
7.4. 47 39 36 51 22 66 40 70 20 34
72 55 60 57 56 75 44 61 50 58

Table 7.4 Grouped Frequency Distribution of Performance for

Employees in a Packaging Firm
Class Tally Marks Frequency
Intervals (f)
73-77 || 2
68-72 ||| 3
63-67 |||| 4
58-62 ||||| 5
53-57 ||||| 5
48-52 ||||| || 7
43-47 ||||| | 6
38-42 |||| 4
33-37 |||| 4
28-32 ||| 3
23-27 |||| 4
18-22 ||| 3
Σ f = N =
50
Study Session 7 Frequency Distribution

7.2.1 Steps in Constructing a Grouped Frequency Distribution

1. Calculate the range of the raw scores. This is obtained
by subtracting the lowest value from the highest value. In
Table 7.3 the highest value is 75, while the lowest is 19.
Therefore, the range is 75 - 19, which gives 56.
2. Determine the number of class intervals and the class
interval size or class width of each interval. Points to
note here are:
a. The number of class intervals must not be less than 10
or above 20. It is, however, better to have intervals
between 10 and 15 for convenience.
b. Certain interval sizes or width are preferred; these are
2, 3,5 or a multiple of 5 like 10, 15, 20 etc.
c. All intervals should be the same size. This means that,
the range must be divided into a number of equal
intervals.
This formula could be used

+1
Where C.I. = Class interval
H = Highest score
L = Lowest Score
I = Class Interval size chosen
H –L = Range
For the case under study we have:

The higher the interval size chosen, the fewer the

number of class intervals. The lower the interval size
chosen, the more the number of class intervals that
Tip will emerge.
 PSY103 Quantitative Methods in Psychology

3. Arranged classes in a decreasing order of sequence with

the lowest measure placed at the bottom of the table
and the highest measure at the top, according to
accepted rule. The first class interval from below is
therefore 18-22; this covers the performance scores (units
packaged) of 18, 20, 21 and 22, which makes the class
interval of 5. The second interval of 23-27 covers the
scores of 23, 24, 25, 26 and 27. The lowest score in each
interval is called the lower limit while the highest score is
called the upper limit. Thus, 18, 23, 28, etc are lower
limits while 22, 27, 32 etc are respectively upper limits in
the example in Table 7.4.
As guidelines note the following:
a. Ensure that the lowest score in the raw data is neither
the lower limit nor the upper limit of the lowest class
interval, or else the number of class intervals emerging
may exceed or be less than that expected from your
calculations.
a. Classes must be mutually exclusive; that is they must
not over lap. This means that the upper limit of a class
interval must not be the same as the lower limit of the
next following interval. For instance, 18-22, 22-26, 26-
30 are over-lapping class intervals.
b. Classes must also be collectively exhaustive, that is,
should cover all observations. To ensure this, the
lowest class interval should be able to accommodate the
lowest raw score, while the top-most class interval
should be able to accommodate the highest score.
Thus, the class interval 18-22 accommodates 19, which
is the lowest score, while 73-77 accommodates 75 the
highest score, in our example.
c. Make the lowest score in each following interval a
multiple of the interval size. Thus, we have 18, 23, 28
etc. with 5 as interval size.
d. Gaps between classes must be avoided as much as
possible.
4. Run through the raw data either horizontally or vertically
and score each observation with a tally mark against its
class until all observations are covered.
5. The tally marks in each class interval are added up to find
the frequency for the interval. The frequency opposite a
given class interval indicates the number of cases with
Study Session 7 Frequency Distribution

scores in that interval. Thus, grouped frequency

distributions lose information, since the exact value of each
score is not provided.
6. Add the frequencies and place the total at the bottom of the
frequency column.
7. Always give your tables suitable headings.

7.3 The Exact Limits and Mid-Points of Class Intervals

In grouped frequency distributions, class intervals do not show
how values in continuous measurement forms should be
presented. This is because class intervals make provisions only for
discrete measures. This Study Session would show how to correct
this anomaly using class intervals. Exact limits of class intervals
are as well used in grouped frequency distribution. Within a class
interval, a set of values are grouped together, thereby making all
scores in a class interval to lose identity. To represent all the
scores in a class interval, therefore, the mid-point value is
determined.
7.3.1 Exact Limits and Mid-Points
The limits (lower and upper) of the intervals in Table 7.4 do not
show exactly where each interval begins and ends. However, in
statistical situations, it is very important to think in terms of exact
limits. The interval 18-22 contains the scores 18, 19, 20, 21 and
22. A measure of 18, in essence, means anything from 17.5 to
18.5, which could be 17.5, 17.6, 17.7 ……… 18.5. Thus, an
interval of 18-22 has exact limits of 17.5 to 22.5. An interval of
23-27 has exact limits of 22.5 to 27.5. Exact limits are, therefore,
established for each interval by subtracting .5 from the lower limit
and adding .5 to the upper limit. These are then referred to as
lower exact limits and upper exact limits respectively. This
principle holds, no matter the size of the interval. Exact limits are
also called real limits. So, we can refer to these limits
respectively as lower real limit and upper real limit of class
intervals.
It will easily be observed that each interval begins exactly where
the one below it ends, which is as it should be. The interval with
its exact limits should read 17.5 up to 22.5, which means it
includes 22, 22.1, 22.2, 22.3, 22.4 but not 22.5. This is the
interpretation, which is different from how it is read ordinarily.

Table 7.5 Exact Limits and Mid-Points of the Distribution in Table

7.4
 PSY103 Quantitative Methods in Psychology

Intervals Exact Limits Mid-Points Frequencies

(x) (f)

73-77 72.5 - 77.5 75 2

68-72 67.5 - 72.5 70 3

63-67 62.5 - 67.5 65 4

58-62 57.5 - 62.5 60 5

53-57 52.5 - 57.5 55 5

48-52 47.5 - 52.5 50 7

43-47 42.5 - 47.5 45 6

38-42 37.5 - 42.5 40 4

33-37 32.5 - 37.5 35 4

28-32 27.5 - 32.5 30 3

23-27 22.5 - 27.5 25 4

18-22 17.5 - 22.5 20 3

The mid-point of a class interval is the most central score in the

interval. This is calculated easily by adding the lower and upper
limits of each interval and dividing the sum by 2. The exact limits
could as well be added and divided by 2; they give the same mid-
point. The mid-point is represented by small letter (x).
Study Session 7 Frequency Distribution

Study Session Summary

In this Study Session, we have demonstrated how raw data

can be presented in regular frequency distribution form. This
involves the use of tally marks that are later converted / added
Summary up to find the frequency. You have also learnt that the
grouped frequency distribution is used when there are more
than 15 values of scores. It calls for the grouping of raw data
into class intervals and the number of scores in each class
interval is recorded as frequency.
You also learned that exact limits are calculated by subtracting .5
from the lower limit to obtain the lower exact limit and adding .5
to the upper limit to have the upper exact limit. The mid-point
value is obtained by dividing the result of the summation of the
lower and upper limits by 2.

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions.
Assessment
SAQ 7.1 (tests Learning Outcome 7.1)
A security unit of a psychological centre measured the speed
of 25 cars that entered into its premises. The resulting speeds
were:
29, 23, 30, 30, 27, 24, 30, 25, 23, 28, 25, 24, 28, 30, 23, 30,
27, 25, 29, 24, 23, 26, 30, 28, 25
Prepare a frequency distribution for this data.

Assignment

Below is a set of intelligence quotient (IQ) scores of 50

undergraduate students.
IQ Scores of 50 Undergraduate Students

Assignment 109 99 10 12 94 106 97 110 115 106

1 6
104 11 84 11 105 122 110 92 112 99
2 6
 PSY103 Quantitative Methods in Psychology

125 10 95 10 122 102 119 120 103 103

7 3
108 98 11 11 114 109 109 118 105 111
8 4
107 11 12 12 88 121 112 128 117 107
1 7 2
Task
1. Using a class interval size of 5, and with the lower limit of the
lowest class interval as 80, develop a grouped frequency
distribution for the IQ Scores.
2. How many class intervals emerged?
3. What is the range of the raw scores?
Send in your answers to your tutor by submitting at Study
Session 7 Assignment Panel at UI Mobile Class.
Study Session 8 Graphic Representation of Frequency Distribution

Study Session 8

Graphic Representation of
Frequency Distribution

Introduction
So far, Study Session 7 has shown how data could be collected
and organized statistically. As earlier explained, statistics also
involve the presentation and interpretation of data. This Study
Session will show how facts of frequency distribution can be
presented clearly in some graphic ways. Such presentations help
to interpret the statistical data, by translating numerical
information, which are sometimes difficult to comprehend into
pictorial forms that are more readily understandable. Graphic
presentations, thus give better picture of a frequency distribution
by setting out its general contour and showing more precisely the
number of cases in each interval.

When you have studied this session, you should be able to:
 present frequency distributions in polygon, histogram,
Learning bar chart, and relative bar chart forms. (SAQ 8.1)
Outcomes

8.1 The Frequency Polygon

Polygon means many-sided figure. Simple frequency polygons
are particularly adequate for continuous data where scores can
emerge between score values. Generally, scores are always
treated as continuous, whether or not fractions occur. The steps
involved in drawing a frequency polygon are as follow:
1. Note the size of the range of the distribution and rule a
base line of suitable length to cover the number of class
intervals. This line is referred to as the x-axis or abscissa.
2. Clearly mark the exact limits of the class intervals on the
X-axis. However, in drawing the diagram, always allow
two extra intervals, one at each end of the axis so as to
allow you to bring down the diagram to the base line.
Using the frequency distribution in Table 7.4, we have 12
intervals; adding two extra intervals make it 14 intervals.
 PSY103 Quantitative Methods in Psychology

The base line is therefore divided into 14 equally spaced

intervals, beginning with 12.5 and ending with 82.5 as
shown in Figure 8.1. The intervals 12.5 – 17.5 and 77.6 –
82.5 are the two intervals added.
3. At the left-hand side of the base line, draw a perpendicular
line, called Y-axis or ordinate, and mark off on this axis
equally spaced units to represent the frequencies. Note that
the height of the diagram should not exceed the width. The
values on the Y-axis start with 0.
4. At the mid-point of each interval on the X-axis, go up on
the Y-axis until a position is reached corresponding to the
frequency for that interval and place a dot here.
5. Connect all dots with straight lines to give the frequency
polygon.
6. The polygon looks hanging. To bring the polygon to the
base line, extend the straight lines to the mid-points of the
two extra intervals. Thus, the frequency to be recorded at
the mid-points of the added intervals is a zero. The
resulting frequency polygon provides a pictorial illustration
of the frequency distribution.
Fig 781 Frequency Polygon of Employee Performance in Table 7.4
(Exact Limits and Mid-Points of the Distribution in Table
7.5)

8.2 The Histogram

The histogram graph is sometimes called a column diagram. It
looks like a bar chart except that there are no spaces between the
―bars‖ of the histogram, as can be seen in Figure 8.2
Fig 8.2 A Histogram plotted from the Distribution of Employees
Study Session 8 Graphic Representation of Frequency Distribution

Performance in Table 7.4 (Exact Limits and Mid-Points of

the Distribution in Table 7.5)

Please correct the values on the horizontal line (i.e. X-axis)

*****
12.5 17.5 22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 62.5 67.5
72.5 77.5 82.5 *****

The procedure in marking out the histogram is identical to that

involved in plotting the frequency polygon. However, there are
some exceptions.
1. There may be no need to include the two extra intervals as in
frequency polygon.
2. After the points representing the frequencies are located as
before at the mid-points of the intervals, these points are not
joined by straight lines. Rather, a horizontal line is drawn
through each point and extended to the exact limits of each
interval, after which two vertical lines are drawn to the lower
exact limit and upper exact limit respectively, on the X-axis.
These then complete the rectangles or columns; thus, the
histogram is called a column diagram.
A Comparison of the Frequency Polygon and the
Histogram
The frequency polygon and the histogram show basically
the same thing. Both enable us to see graphically how
scores are distributed, whether symmetrically or
Tip asymmetrically (piled up at the low or high end). Both
enable us also to see at a glance the number of cases falling
in each interval, but the histogram is more exact in this than
the polygon. This is because in the histogram, each
 PSY103 Quantitative Methods in Psychology

measurement occupies exactly the same area size, and each

rectangle is directly proportional to the number of measures
within that interval.
The frequency polygon is often preferred to the histogram
when there is need to give a clearer picture of the contour or
shape of the distribution. It is also easy to compare two
frequency distributions with the frequency polygon.

8.3 The Bar Chart

The bar chart is like the histogram in shape but with spaces
between the ―bars‖. This is because bar charts are used in
displaying frequency distribution of qualitative or attribute data.
Such data have distinct points that cannot be merged, as the case
with histogram. Assume the birth and the death rates in a country
for five consecutive years are shown on a table as we have it in
Table 8.1:
Table 8.1 Birth and Death Rates in a Country for Five Years
Year Birth Death
1989 20,000 10,000
1990 25,000 15,000
1991 37,000 20,000
1992 43,000 32,000
1993 56,000 45,000
To represent the birth rate graphically on a bar chart, the attribute
(year) is shown on the x-axis, while the frequency (number of
births) is plotted on the y-axis as shown in Figure 8.3
Study Session 8 Graphic Representation of Frequency Distribution

Fig 7.3 Bar Chart on the Number of Births

Bar charts can also be used for graphical comparisons between

two or more classes of attribute data. Look at Fig 8.4 for graphical
comparison between the birth and the death rates in Table 8.1.
Fig 8.4 Comparative Bar Chart on the Number of Births and
Deaths

o ITQ What type of graph is used for discrete data

or qualitative data?
A. bar graph.
B. histogram.
 PSY103 Quantitative Methods in Psychology

Feedback on ITQs answers

 The correct answer is A. Option B (histogram) is
used when the data are continuous.

Study Session Summary

In this Study Session, you have seen that there are three basic
types of graphs that we use for most data: (1) frequency
polygon, (2) bar chart(graphs), and (3) histograms. The simple
Summary frequency polygons are adequate for continuous data where
scores can emerge between score values. The names of the
last two are a bit misleading because both are created using
bars. The only difference between a bar graph and a histogram
is that in a bar graph the bars do not touch while the bars do
touch in a histogram. In general, bar graphs are used when the
data are discrete or qualitative. The space between the bars of
a bar graph emphasize that there are no possible values
between any two categories. For example, when graphing the
number of children in a family, a bar graph is appropriate
because there is no possible value between any two categories
(e.g., 1 and 2 children). When the data are continuous, we use
a histogram. The bars touch in a histogram to indicate that
there are possible values between any two categories. For
example, if we were graphing time to complete a test, the bars
would touch to indicate that there are possible values between
any two times (e.g., 27 and 28 minutes).

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions.
Self Assessment
SAQ 8.1 (tests Learning Outcome 8.1)
Draw a bar graph of frequency distribution car speed in
SAQ7.1
Study Session 8 Graphic Representation of Frequency Distribution

Now that you have completed this study session, attempt the
following assignment (a tutor marked assessment).
The table below shows the population of undergraduate
students in the Faculty of the Social Sciences at the University
Assignment of Ibadan for the 2005/2006 academic session.

Departments Levels of Study

1 2 3 4
Psychology 100 150 180 60
Sociology 120 180 90 50
Political 80 110 160 100
Science
Geography 110 165 200 140
Economics 90 140 170 200
Task
1. Present the data on the table in a suitable graph.
2. Graphically present the statistics of the students in the
faculty in terms of their departments.
3. Graphically present the statistics of the student population
in the faculty in terms of their levels of study.
 PSY103 Quantitative Methods in Psychology

Study Session 9

Diagrammatic Representation of
Frequency Distribution

Introduction
Statistics involve, among others, the presentation and
interpretation of data in some diagrammatic ways. This Study
Session will discuss the representation of frequency distribution
using pie chart, pictogram and stem-and-leaf display.

When you have studied this session, you should be able to:
 present raw data in the form of pie chart, pictogram,
Learning and stem-and-leaf display. (SAQ 9.1)
Outcomes

9.1 The Pie Chart

It is a circular diagram in which classes or attributes are
represented by sectors. Each sector represents a class or attribute
at a proportion equal to the percentage or relative frequency of the
class or attribute. The relative frequency of an attribute is the
ratio of the frequency of that attribute to the total number of
occurrences of observations.
Using the example of the number of employees in the various
departments of a company , the steps involved in drawing a pie
chart are provided after the table.
Table 9.1 Distribution of Employees in a Company
Department Student
Population
Marketing 153
Accounts 60
Production 102
Personnel 85
Study Session 9 Diagrammatic Representation of Frequency Distribution

Works 98
Total 498

1. Find the relative frequency of percentage of the employees

in each department using the formula:

*** change department in the denominator to company***

For marketing, R.F is

For Accounts, it is

For Production, it is

For Personnel, it is

For Works, it is

Look at Table 9.2.

Table 9.2 Relative Frequency for a Pie Chart Distribution of the

Employees in a Company in Table 9.1
Department Student Relative
Population Frequencies
Marketing 153 30.7%
Accounts 60 12.0%
Production 102 20.5%
Personnel 85 17.1%
Works 98 19.7%
Total 498 100%
 PSY103 Quantitative Methods in Psychology

Note that the sum of relative frequency must be 100.

2. Calculate the angle of the sector for each department, using

the formula

Angle of sector

For Marketing it is

For Accounts, it is

For Production, is

For Personnel, it is

For Works, it is
Note that the angles must sum up to 360o. See Table 9.3
Table 9.3 Pie Chart Distribution of the Employees in a Company in
Table 9.1
Department Student Relative Angle of
Population Frequencies Sector
Marketing 153 30.7% 110.60o
Accounts 60 12.0% 43.37o
Production 102 20.5% 73.74o
Personnel 85 17.1% 61.45o
Works 98 19.7% 70.84o
Total 498 100% 360o

3. Draw the Pie Chart: This can be done neatly with a pair of
compass and a protector. Using a pair of compass, measure
any length and draw a circle, showing the circumference.
From the center of the circle, draw a straight line to the
Study Session 9 Diagrammatic Representation of Frequency Distribution

circumstance (i.e. radius). At the radius, use your protector

to measure the various degrees calculated for each
department, and show the corresponding percentage in the
division. The resultant diagram is a pie chart which you can
see in Fig 9.1.
Fig 8.1 Pie Chart of Employees in a Company in Table 8.3

9.2 The Pictogram

The pictogram is also called the pictograph or ideograph. It is a
graphic way of presenting data with emphasis on the aesthetic
aspect of the display and not accuracy. The symbol or picture
chosen for any set of data must have relationship or resemblance
with the concept being portrayed. For example, Look at Fig 9.2,
it shows the pictogram of deaths recorded in Table 8.1
 PSY103 Quantitative Methods in Psychology

Fig 8.2 Pictogram of the Total Number of Deaths in a Country

8.3 The Stem-and-Leaf Display

This is a simple, useful but uncommon technique for summarizing
a set of data by combining the features of the frequency
distribution and the histogram. The class intervals form the
―stems‖ while specific values of the raw data within each interval
form the ―leaves‖. An example using scores in Table 8.1 and
illustrated in Table 9.2 is shown in Table 9.4.
Study Session 9 Diagrammatic Representation of Frequency Distribution

Table 9.4 Stem-and-Leaf Display for data in Table 7.3

Stems (Intervals) Leaves Frequency
(Observations) (f)
73-77 4 5 2
68-72 2 8 8 3
63-67 3 6 6 7 4
58-62 1 2 2 8 8 5
53-57 3 4 5 6 7 5
48-52 0 0 0 1 1 8 9 7
43-47 3 4 5 5 6 7 8
38-42 0 0 1 9 4
33-37 45 6 6 4
28-32 0 2 2 3
23-27 3 4 6 7 4
18-22 0 0 9 3
Each leaf is made up of the number of scores falling within a
given interval with each score represented mainly by the last digit
of the score, if they are two-digit scores. The scores in each leaf
are then ordered from low to high. With the class interval of 23-
27 in Table 9.4 for instance, reading the raw data in Table 7.3
horizontally, the scores falling in this interval are 26, 24, 23 and
27. The last digits for these are 6, 4, 3 and 7 respectively; these
form the leaf for the interval 23-27. When ordered from low to
high, the leaf stands as 3, 4, 6 and 7. Do the same for other
intervals.
The stem-and-leaf display shows the same graphic representation
as the histogram because each observation takes up one space.
Also, it is very easy to recall each score in the raw data from the
display. For instance, it can be read from the display that the
scores within class interval 18-22 are 20, 22 and 19 since the leaf
is made up of 0, 2 and 9. The 0 is for 20, 2 is for 22, and 9 is for
19 (remember the interval is 18-22). For data with three or more
digits, you may choose to increase, correspondingly, the number
of digits to use for the leaves.
 PSY103 Quantitative Methods in Psychology

Study Session Summary

In this Study Session, you have learnt that the pie chart is a
circular diagram in which classes or attributes are represented
by sectors. A pictogram is a graphic way of presenting data
Summary with emphasis on the aesthetic aspect of the display and not
accuracy. The stem-and-leaf display is a technique for
summarizing a set of data by combining the features of the
frequency distribution and the histogram.

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions. Write your answers in your Study
Self Assessment Diary and discuss them with your Tutor at the next Study
Support Meeting. You can check your answers with the Notes
on the Self-Assessment Questions at the end of this Manual
SAQ 9.1 (tests Learning Outcome 9.1)
The table below shows the monthly spending of Mrs. Halima
ITEM AMOUNT SPENT (N:K)

Food 7000

Transportation 1250

Accommodation 6000

Savings 1500

Miscellaneous 1000.00

Represent the data in a pie chart

Study Session 9 Diagrammatic Representation of Frequency Distribution

In an election that was conducted in a Local Council Ward,

among the 2, 500 votes recorded, 500 voted for party A; 230
voted for party B; 800 voted for party C; 180 voted for party
D, and the others voted for party E. Present the result of the
election in an appropriate diagram.
Assignment
List any three variables that could be presented in a
pictographic form. Demonstrate how each could be presented
using self-generated data.
Post your answer to your tutor on Study Session 9 Assignment
page at UI Mobile Class.
 PSY103 Quantitative Methods in Psychology

Study Session 10

Cumulative Distributions

Introduction
In the simple frequency distribution, you can easily determine the
number of cases falling within a class interval by adding up
frequencies in that interval. But when additional information is
required, like the number of cases occurring above or below a
given interval, then cumulative frequency distribution becomes
necessary. This Study Session will expose you to how to use the
cumulative frequency distribution which is an alternative method
that is used when information is required in form of ―more than‖
or ―less than‖.
When you have studied this session, you should be able to:
 infer from frequency distributions, the number of cases
falling at, and above or less than, a particular point.
Learning  plot graphs showing ―more than‖ and ―less than‖
Outcomes cumulative frequency distributions.

9101 Constructing Cumulative Distribution

Let us take a look at the information in Table 7.3. Can you state
the number of employees whose performance scores are 38 and
above or those who are 62 and below. You most likely observe
that this cannot be easily decided. To achieve such a task, the
cumulative frequency distribution is required, by cumulating the
frequencies to see how many employees are ―more than‖ or ―less
than‖ a certain level.
10.1.1 Constructing a ―More Than‖ Cumulative Distribution
The cumulating starts from the highest class interval and
descends. The highest class interval in Table 10.1 is 73-77 with a
frequency of 2. This explains that two employees scored 73 and
above, thus 2 appears in the column of ―more than‖ cumulative
for the interval 73-77. For the interval 68-72, the frequency is 3
and it is interpreted that three employees scored between 68 and
72. The number of employees whose scores are 68 and above will
then be 2+3=5, so 5 appears as the ―more than‖ cumulative
Study Session 10 Cumulative Distributions

frequency for the interval 68-72. For the interval 63-67, it is

2+3+4=9.
The successive addition continues until the class intervals are
exhausted. It should be noted that the last cumulative frequency
must be equal to Σf or N, which is the total number of
observations; otherwise, a mistake must have been done. In our
example, Σf or N = 50, this tallies with what is on the Table 10.1.
It implies that 50 employees scored 18 or more.

Table 10.1 ―More than‖ and ―Less than‖ Cumulative Distribution of

Performance for Employees in a Packaging Firm
Class Frequency More than Less than
Intervals (f) Cumulative Cumulative

73-77 2 2 50
68-72 3 5 48
63-67 4 9 45
58-62 5 14 41
53-57 5 19 36
48-52 7 26 31
43-47 6 32 24
38-42 4 36 18
33-37 4 40 14
28-32 3 43 10
23-27 4 47 7
18-22 3 50 3
Σ f = N = 50

10.1.2 Constructing a ―Less Than‖ Cumulative Frequency

Distribution
It is very much similar to the ―more than‖ cumulative frequency
distribution, except that we now begin to cumulate the frequencies
from the bottom class interval. As shown In Table 10.1, there are
three employees whose performance scores are between 18 and
22 or less; 3 therefore appears at the bottom of the ―less than‖
 PSY103 Quantitative Methods in Psychology

cumulative frequency column for the interval 18-22. The

following interval shows that 4 employees had performance
scores range between 23-27. The total number of employees
whose performance scores are 27 or less is calculated as 3+4=7.
Thus, the ―less than‖ cumulative frequency at the interval 23-27 is
7. This procedure also continues until all the intervals are
exhausted. Again, the ―less than‖ cumulative frequency for the
top most interval must be equal to Σf or N. In this case, it is 50,
and it is interpreted that 50 employees have performance scores of
77 or less.

10.2 Graphical Presentations of Cumulative Distribution

When distributions are placed in cumulative forms, they could
also be presented graphically. The procedure involved in
constructing the cumulative distributions graphically is what this
section is designed to achieve.
10.2.1 Graphical Presentation of a ―More Than‖ Cumulative
Distribution.
Cumulative frequencies are plotted on the Y-axis, while the
corresponding performance scores are on the X-axis. The method
of plotting is quite similar to that of plotting the simple frequency
distribution. However, while the frequency polygon type of graph
is adopted, the histogram type of graph is not applicable here.
When plotting the frequency polygon, the frequency in each
interval is taken at the mid-point of the interval but in the ―more
than‖ cumulative frequency curve, each cumulative frequency is
plotted at the lower exact limit of the interval. For the ―more
than‖ cumulative frequency curve of the data in Table 10.1, take a
look at Figure 10.1. Also two extra intervals are not added in this
case.
Study Session 10 Cumulative Distributions

Fig 10.1 ―More than‖ Cumulative Distribution Curve of Employees’

Performance

***Please the title for all Figures are supposed to be below the
figure not above. Only Tables should have their titles above.
Please take not for all corrections

When all the cumulative frequencies have been plotted, the points
are joined by straight lines. To bring the curve down to the base
line, the cumulative frequency of 0 is plotted at the upper exact
limit of the interval 73-77, which is 77.5. This means that there
are no employees with performance scores of ―more than‖ 77.5.
10.2.2 Graphical Presentation of ―Less Than‖ Cumulative
Distribution.
In the ―less than‖ cumulative frequency, the graph is plotted
between the cumulated frequencies and the upper exact limit of
the interval. The graph is brought down to the base by placing the
cumulative frequency of 0 at the lower exact limit of the bottom
interval. Thus, there is no employee who performed less than
17.5.
From the graph in Figure 10.2, you can now easily read the
number of employees who performed less than any given score.
So also, you can read the number of employees who performed
more than any given score, with the ―more than‖ cumulative
distribution.
 PSY103 Quantitative Methods in Psychology

Fig 10.2 ―Less than‖ Cumulative Distribution Curve of Employees

Performance Score.

Study Session Summary

In this Study session you learned that the ―more than‖

cumulative distribution shows how many cases fall at, and
above a given point, while the ―less than‖ distribution shows
Summary how many cases fall at, and below a given point. We also
noted that the method of plotting cumulative distribution
graphs is quite similar to that of plotting the simple frequency
distribution.
However, when plotting the ―more than‖ cumulative frequency
curve, each cumulative frequency is plotted at the lower exact
limit of the interval. To bring the curve down to the base line, the
cumulative frequency of 0 is plotted at the upper exact limit of the
highest class interval. In the ―less than‖ cumulative frequency, the
graph is plotted between the cumulated frequencies and the upper
exact limit of the interval. The graph is brought down to the base
by placing the cumulative frequency of 0 at the lower exact limit
of the bottom interval.
Study Session 10 Cumulative Distributions

Assessment

Now that you have completed this study session, you can
assess how well you have achieved its Learning Outcomes by
answering these questions. Write your answers in your Study
Self Assessment Diary and discuss them with your Tutor at the next Study
Support Meeting. You can check your answers with the Notes
on the Self-Assessment Questions at the end of this Manual
SAQ 10.1 (tests Learning Outcome 10.1)
Cumulative frequency distribution is that which helps to calculate
the number of people falling at a particular score:
(a) True
(b) False.

With the skills acquired in this Study Session, answer the

following questions using Table 10.1 as a guide:
1. How many employees had score of 43 and
above?
Assignment 2. How many had score of 32 and below?
3. How many had scores between 32 and 43?
Post your answers on Study Session 10 assignment page at UI
Mobile Class.
 PSY103 Quantitative Methods in Psychology

Study Session 11

Measures of Central Tendency:

Mean

Introduction
The frequency distribution with its graphic representations
primarily shows the organization of data into a manageable form.
They are the preliminary steps toward quantitative treatment of
data. The method of statistically treating the distribution gives
numerical descriptions which are the basic features of the
distribution. These further enable concise and definite comparison
of the features of one distribution with others. This study session
will attempt to expose you to measures of central tendency, which
is one of the types of numerical descriptions of frequency
distributions.

When you have studied this session, you should be able to:
 compute and interpret the arithmetic mean or average
score for both ungrouped and grouped data.
Learning
Outcomes 

A measure of central tendency is a location or representative

point on a scale, a central point where data are concentrated,
centered, packed or clustered with other scores scattering on
either side. Such measure serves the following purposes:
1. It is a measure that represents all measures in a sample,
which could be utilized in comparing one distribution
with another.
2. A sample is a representative of a particular population. A
measure of central tendency, therefore, describes
indirectly, with some amount of accuracy, the population
from which the sample has been drawn. Thus, we can
estimate a measure of central tendency for a population
Study Session 11 Measures of Central Tendency: Mean

from the knowledge of the sample measure of central

tendency. Generalizations and predictions are therefore
made beyond that of the sample to the population. This
makes scientific research in the behavioural sciences and
education possible.
There are basically three measures of central tendency; these are:
the mean, the median and the mode.
Three types of mean are used in statistics. These are:
1. the arithmetic mean;
2. the geometric mean; and
3. the harmonic mean
The arithmetic mean is the most frequently used, while the other
two have a more restricted use. In this case we shall limit our
learning to the arithmetic mean

11.1 The Arithmetic Mean

11.1.1 Ungrouped Data
Another name for arithmetic mean is average. To calculate the
arithmetic mean for ungrouped data, add up all the measurements
and divide by the number of measurements added. The formula
for the calculation is given as:

???

Where: = Arithmetic mean

Σ = ―the sum of‖
X = an individual score
ΣX = sum of X scores
N = the number of measurements
Given that the scores (in percent) obtained by a student in seven
subjects are 43, 59, 61, 73, 55, 64, and 49. The average
performance of the students on the seven subjects is calculated
thus:

ΣX = 404
N=7
 PSY103 Quantitative Methods in Psychology

11.1.2 Regular Frequency Distribution

With scores in the form of a regular frequency distribution, the
mean is computed, using the formula:

Where and Σ remain the same

Fx = x score multiplied by the frequency of that
score
Σfx = sum of fx values
N = total number of scores i.e. Σf
Assume that a company recorded the number of accidents
experienced per worker in a year, as shown in Table 11.1????? in
a regular frequency distribution form; calculate the average
number of accidents recorded in the company for the year.

Table 11.1 Regular Frequency Distribution of Accidents recorded in

a Company
No. of No. of Fx
Accidents Employees
(X) (f)
9 2 18
8 2 16
7 6 42
6 5 30
5 3 15
4 4 16
3 5 15
2 4 8
1 0 0
Σ f = N = 31 Σfx = 160
Σ fx = 18 + 16 + 42 + - - - - - + 0 = 160
Study Session 11 Measures of Central Tendency: Mean

N = total number of employees = 31

The average number of accidents recorded in the company for the

year is therefore 5.16, which may be approximated to 5 since
number of accidents can only be expressed in a discrete
measurement form.
11.1.3 Grouped Frequency Distribution with Class Intervals
The formula used for this frequency distribution is the same as
that used for the regular frequency distribution. In fact, both
distributions are grouped frequency distributions. While the
former is without class intervals, the latter is with class intervals.
For the grouped frequency distribution with class intervals:

Where and Σ remain the same

X = mid-point of class interval
F = number of cases in an interval
Fx = multiply each x score by corresponding f
value
Σfx = sum of fx values
N = total number of scores i.e. Σf
Look at Table 10.2. Two employees scored between 90 and 99.
Since the individual scores of the employees is lost in the course
of grouping, we assume that two employees scored 94.5; this
being the mid-point for the interval 90-99. This, however, is not a
true representation of all intervals. Nonetheless, the error in any
interval is small and in the final calculation of the mean, most of
the small errors tend to cancel each other out, bringing out a final
result that is essentially correct.
Table 11.2 Grouped Frequency Distribution of Performance Scores
for 100 Employees
Class Interval Mid- F FX
points
90-99 94.5 2 189
80-89 84.5 4 338
 PSY103 Quantitative Methods in Psychology

70-79 74.5 3 223.5

60-69 64.5 15 967.5
50-59 54.5 24 1308
40-49 44.5 21 934.5
30-39 34.5 14 483
20-29 24.5 7 171.5
10-19 14.5 5 72.5
0-9 4.5 5 22.5
N=10 Σfx =
0 4710

Study Session Summary

In this Study Session, we have mentioned three types of mean;

arithmetic, geometric, and harmonic but narrowed down on
the arithmetic mean. We stated that the arithmetic mean or
Summary average is calculated for ungrouped data by adding all the
values together and dividing by the number of cases. For the
grouped frequency data with class intervals, you have to
multiply the mid-point value of each class interval with
corresponding frequency value, and add all the multiples, and
then divide the outcome by the total number of cases grouped.
Study Session 11 Measures of Central Tendency: Mean

Assessment

1.

Assignment
The Table 11.3 are data of time spent by students reading for an
examination. Study the Table closely and answer the questions
that followTable 11.3: Showing Raw Data of Time Spent by
Students Reading for an Examination

6 2 2 6 5 7 4 9 8 5
4 9 5 6 7 5 3 5 5 6
3 7 8 4 3 7 6 7 6 6
8 6 5 4 10 5 3 9 4 8
(a) Using the date in Table 11.3, compute the average time spent
by the students reading(b) Using a grouped frequency
distribution, compute the average time spent by the students
reading
Post your answers on Study Session 11 assignment page at UI
Mobile Class.
 PSY103 Quantitative Methods in Psychology

Study Session 12

Measures of Central Tendency:

Median

Introduction
This study session will expose you to how scores are arranged.
The median is the mid-point of a set of scores. It is the point on a
continuum of scores above which fall exactly one half of the cases
and below which fall the other half. When scores are arranged in
order of magnitude, either in ascending or descending order, the
median has the same number of scores above and below it.
When you have studied this session, you should be able to:
 locate the median position of a set of ungrouped data,
as well as calculate the median value of grouped
Learning frequency distribution, with and without class intervals.
Outcomes  determine the median value of a distribution, using the
distribution curve, called OGIVE.

12.1 The Median of Ungrouped Data

The calculation of the median of ungrouped data takes two forms,
as determined by the number of observations, N. When N is an
odd number, the median is determined as the ½(N+1)th
observation. When N is even, the median is the average of the two
most central scores. For example, with scores like 9, 11, 8, 7, 15,
13, 8, 9, 9. N is 9, which is an odd number, therefore, the median
will be calculated as Md = ½(9 + 1)th value, i.e. ½(10)th value =
5th value. To decide on the 5th value, we have to re-arrange the
values in order of magnitude. This gives us: 7, 8, 8, 9, (9), 9, 11,
13, 15.
The 5th value is 9, therefore, the median for the set of scores is 9.
For another set of scores like 9, 11, 8, 7, 15, 13, 8, 9 the number
of observations is 8, which is even. The median will be taken as
the mean of the two middle values, that is, the mean of the 4th and
5th observations. Arranging the scores in order of magnitude, we
Study Session 12 Measures of Central Tendency: Median

have 7, 8, 8, (9, 9), 11, 13, 15. The 4th and 5th observations are 9
and 9 respectively. Md is ½ (9 + 9) = ½ (18) = 9.

12.2 The Median of Grouped Data

When the data are grouped in a regular frequency distribution or
in a grouped frequency distribution with intervals, the calculation
of the median takes a different pattern. The computational
formula is given as:

Md
Where LRL = lower real limit of median interval
N/2 = one half of the total number of cases
SFB = sum of all the frequencies up to, but not
including the median interval
F = the frequency within the median interval
I = class interval size
12.2.1 The Median of Grouped Data with Regular Frequency
Distribution
The computational procedures shall be explained, using the data
in Table 11.1, which is now shown in Table 12.1 for the median
calculation.
Table 12.1 Regular Frequency Distribution of Accidents recorded in
a Company
No. of No. of Employees
Accidents (f)
(X)
9 2
8 2
7 6
6 5 5 cases down here
5 3 Median interval
4 4 13 cases up here
3 5
 PSY103 Quantitative Methods in Psychology

2 4
1 0
Σ f = N = 31

1. Determine the median interval. The task of calculating the

median is to determine the point on the scale on either side
of which lie half the cases. As the total number of cases, N,
in this example is 31, then 15.5 cases will lie on either side
of the median, i.e. N/2. However, there is a problem
locating this in a frequency distribution where the
individual identity of the items has been lost while
grouping.
From the bottom of the frequency column, add up the
frequencies until the interval having the 15.5th cases is
reached. This is called the median interval or critical
interval. In the table, the median interval is 5. The sum of
the frequencies before the median interval (i.e. ―less than‖
cumulative frequency) is 13, and the frequency of the
median interval is 3. Therefore, the 15.5th case must lie
within that interval; thus, it is regarded as the median
interval.
2. Determine the lower real limit (LRL) of the median
interval, which is 4.5
3. Calculate half the total number of cases N/2 = 31/2 = 15.5
4. Sum the frequencies below the median interval, SFB = 13.
5. Determine the frequency within the median interval, f = 3
6. Determine the class width (I) = 1
7. Calculate the median (Md) = 5.33.
12.2.2 The Median of Grouped Data with Grouped Frequency
The steps involved in the computation are similar to those of the
regular frequency distribution. Using the data in Table 11.2, we
will present the calculation of the median. The median interval is
the class interval, having the N/2, i.e 100/2th case. This makes the
50th case the median point. The corresponding interval, 40-49, is
the median interval. Follow the steps outlined before:
LRL =39.5
N =100
SFB = 31
Study Session 12 Measures of Central Tendency: Median

F =21
I =10
Theoretical Explanations
In Table 11.2, the median interval is the interval 40 – 49 since the
50th case resides within the interval. Remember, the 50th case is
the median point for the distribution as N/2 i.e. 100/2 is 50.
Thirty-one cases are counted up to the top of the interval 30-39
(less than cumulative frequency). The next interval 40-49, which
is the median interval contains 21 cases, suggesting that the 50th
case lies somewhere within this interval because adding 21 to 31
gives 52; this is greater than the 50th case desired. The problem
now is how to determine where the 50th case lies within the
median interval?
For the sake of interpolation, we assume that the 21 cases to the
median interval are evenly distributed over the whole range of the
interval with exact limits as 39.5 and 49.5. Actually, we require
19 cases of the 21 cases in the median interval to be added to the
sum of the preceding cases (i.e. 31) to make the desired 50 cases.
Thus, we need a fraction of 19/21 of the median interval. Since
the class width of the median interval is 10, it implies we desire
19/21 of 10, which is 9.05 units of the interval to correspond with
the 50th case desired. Adding the 9.05 to the exact lower limit of
the interval (39.5), we now have 39.5 + 9.05, which is 48.6 as the
median value. So 48.6 corresponds with the 50th case in the
distribution. You can as well observe that 48.6 falls within the
median interval.
Table 12.2 Calculation of the Median of Performance Scores for 100
Employees
Class Interval F Cf
90-99 2 100
80-89 4 98
70-79 3 94
60-69 15 91
50-59 24 76 48 cases down
here
40-49 21 52 Median
Interval
 PSY103 Quantitative Methods in Psychology

30-39 14 31 31 cases up here

20-29 7 17
10-19 5 10
0-9 5 5
N=10
0

As the median is the dividing point of a distribution into two

halves, it could as well be determined from the top-end of a
frequency distribution, that is, by working down from the top.
Starting at the top of the frequency column (―more than‖
cumulative frequency), 48 cases are count down to the interval
50-59. We need 2 more cases to make up the 50th case desired.
This implies going 2/21 of the way down into the median interval.
Again, the class width is 10, therefore we need to go down 2/21 of
10, which is 0.95. Since we are working from the top intervals, we
now subtract 0.95 from the exact upper limit of the median
interval (49.5). We now have 49.5 - 0.95, which is 48.6 as the
median value. This is exactly what we obtained before when
calculating from the bottom interval.
Thus, working from the top class intervals, the formula for
median becomes

Where URL = upper real limit of median interval

N = total number of cases
N/2 = one half of the total number of cases
Study Session 12 Measures of Central Tendency: Median

SFA = sum of all the frequencies above the

median interval but not including the
median interval
F = frequency within the median interval
I = class interval size
For the data in Table 12.2
URL = 49.5
N = 100
SFA = 48
F = 21
I =10

12.3 Determining the Median using the Cumulative Frequency (Or OGIVE Curve)
To find the median of grouped data, the cumulative frequency
distribution curve can as well be used. From the curve, we can
easily read off the median, having determined the median point as
N/2 on the Y-axis for the cumulative frequency.
Figure 12.1 is the OGIVE of data in Table 12.2, employing the
―less than‖ cumulative frequency distribution. The median is the
dividing point of the desired point, that is N/2, which is 100/2 =
50. Read off 50 on the Y-axis to meet the curve at N. From N,
draw another dotted line perpendicular to the Y-axis to meet the
X-axis at P. The corresponding value of P on the X-axis is the
median of the data.

Fig 12.1 OGIVE showing Median Scores

 PSY103 Quantitative Methods in Psychology

Study Session Summary

In this Study session you learned that the median is the

dividing point of a distribution into two halves. In ungrouped
data, it is determined by arranging the data in ascending or
Summary descending order and then locating the most central value. For
grouped data with class intervals, it involves a longer process.

Assessment

Now that you have completed this study session, attempt the
following assignment (a tutor marked assessment).
1. Ten employees in an organization reported the years
they have spent on the job as follows:
Assignment
14, 20, 16, 14, 12, 10, 18, 16, 11, and 12
What is the median year of the employees on the job?
Study Session 13 Measures of Central Tendency: The Mode

Study Session 13

Measures of Central Tendency:

The Mode

Introduction
This Study Session aims to provide explanations on the mode,
which is the most frequent score in a distribution of scores. The
mode is the value with the highest frequency.

When you have studied this session, you should be able to:
 compute the modal value for ungrouped and grouped
frequency distributions with and without class
Learning
Outcomes intervals.

Given a set of scores like these in Table 13.1, the mode is 5. The
highest frequency is 15, while the corresponding score is 5,
indicating that the score 5 occurred 15 times in the set of data.
Thus, 5 is the mode, being the most frequent score.
Table 13.1 Determining the Mode of a Distribution
Score 0 1 2 4 5 7 8 9

Frequency 7 3 11 7 15 6 4 5

13.1 Mode of Grouped Data with Intervals

When data are grouped in the class interval form, the crude mode
or interpolated mode may be determined. The interpolated mode
is, however, more accurate than the crude mode.
13.1.1 The Crude Mode
The crude mode is simply the mid-point value of the class interval
having the highest frequency. In Table 12.2, for instance, the class
interval with the highest frequency is 50-59. The mid-point is 54.5
therefore the crude mode of the distribution is 54.5. In some cases
a distribution may have two class intervals with the same
frequency as the highest. Care must be taken in deciding what the
mode is.
The following rules should be noted:
 PSY103 Quantitative Methods in Psychology

1. If there are more than one class intervals separating the two
class intervals, like distribution A in Table 13.2, then we
say that the distribution has two modes or it is bimodal.
The modes will be the two mid-points of the respective
class intervals. Thus, we have 9.5 and 3.5 as the modes for
distribution A in Table 13.2.
2. If the two intervals are separated by only one intervening
interval, like distribution B in Table 13.2, then it is
possible that the distribution is actually unimodal. It is
more certain to be unimodal when the intervening interval
has a relatively high frequency. In such a case, the crude
mode is not adequate as it will not be possible to decide
what the crude mode is.
3. When the two class intervals with the highest frequency
occupy adjacent positions like distribution C in Table 13.2,
then the crude mode is decided as the dividing point
between the two intervals. The dividing point is simply the
upper real limit of the lower class interval, which is
equivalent to the lower real limit of the higher of the two
class intervals. For distribution C in Table 13.2, the crude
mode is 6.5. Such distribution is invariably unimodal.
Table 13.2 Illustrations of when the Crude Mode is Adequate
Distributions (Frequency)
Class Interval
A B C
11 – 12 2 2 2
9 – 10 7 7 6
7–8 6 6 7
5–6 2 7 7
3–4 7 2 2
1–2 3 3 3
Bimodal Unimodal

13.1.2 The Interpolated Mode

With the limitation in the use of the crude mode, the interpolated
mode is favoured. The crude mode is ideal for small sample of
less than 100. When the sample is large, and the distribution is
skewed, the mid-point value of crude mode is not sufficiently
accurate. This calls for an interpolation within the modal interval
to obtain a more accurate estimate. The formula for calculating
the interpolated mode is given as:
Study Session 13 Measures of Central Tendency: The Mode

Where LRL = lower real limit

d1 = difference between the frequency of the
modal interval and frequency of the
preceding interval
d2 = difference between the frequency of the
modal interval and the frequency of the next
following interval
I = class interval size
An example of calculating the interpolated mode will be applied
to the data in Table 12.2. Some extracts from the Table are shown
below:
60 – 69 15
50 – 59 24 Modal Interval
40 – 49 21
The modal interval is the interval with the highest frequency, that
is, 50 – 59. The preceding interval is 40 – 49, the following
interval is 60 – 69.
LRL = 49.5
d1 = 24 – 21 = 3
d2 = 24 – 15 = 9
I = 10

= 49.5 + 3/12 x 10
= 49.5 + 2.5 = 52
The crude mode is 54.5 while the interpolated mode is 52, which
is a ―pull away‖ towards the preceding interval. The reason is that
the preceding interval has a frequency of 21, while the following
interval has a frequency of 15; hence, the pull towards the
preceding interval. If the frequency of the following interval was
larger, then the pull would have been directed towards the
following interval, making the interpolated mode above 54.5 but
less than 59. Thus, the interpolation of the mode within the modal
class interval improves the estimate of the mode by allowing the
 PSY103 Quantitative Methods in Psychology

adjoining frequencies to add their weight in arriving at a final

estimate.

13.2 Determining the Mode using the Histogram

The mode of grouped data can easily be read off from histogram.
An illustration is made from the data in Table 12.2, and the
histogram is displayed in Figure 13.1.

Fig 13.1 Showing how to determine Mode using the Histogram

To determine the mode from a histogram, draw two diagonals to

join the corners of the modal class, labelled as Mo and Np
respectively in Figure 13.1. At the point where the diagonals
intersect, draw a perpendicular line through to meet the x-axis at
Mo. The value of Mo on the x-axis is read as the mode, which is
52.5 in this case.

13.3 The Characteristics and Use of Measures of Central Tendency

The mean, median, and mode, already referred to as the measures
of central tendency are commonly used in statistical
computations. There are, however, situations in which one is more
adequate than the others. Knowledge of the characteristics and
uses of each of the measures will invariably expose when to use
any of the measures.
Study Session 13 Measures of Central Tendency: The Mode

The Mean
1. The arithmetic mean is often used when the distribution is
reasonably symmetrical. This is because the mean takes
into account all measures in a distribution. The extremely
small values and the extremely large ones, as well as those
near the centre of the distribution, influence the mean.
When, however, there are extreme values on one side of
the distribution that are not balanced by extreme values on
the other side, the mean is greatly influenced. This results
in a misleading picture of the location of the measure of
central tendency. Consequently, the mean is not adequate
for a distribution that is highly skewed.
2. For a normal distribution, the sum of differences (or
deviations) of all scores from the mean is zero. That is, X –
Mean = 0. This implies that the mean balances the sums of
the positive and negative deviations. This will be expanded
later.
3. The mean is the most reliable and stable of the three
measures of central tendency. This is because when
different samples are drawn from the same population and
their means computed on the same measure, there will be
little or no difference in the means. Also, errors of
measurement tend to neutralize one another around the
arithmetic mean. If any error exists in the use of a mean,
the error is considerably smaller than the error of a single
measure in the distribution.
4. The mean is useful in further statistical computation, while
other measures of central tendency are usually not. For the
characteristics of the mean, it is used in many of the
procedures of inferential statistics.
The Median
1. The median is a positional measure of central tendency. Its
position is determined directly by the number of values in a
distribution, not the magnitude of the values. The
magnitude of values only indirectly affects the median by
determining the serial ordering before decision on what the
median is.
2. Though all measures contribute to the calculation of the
median, just like the mean, the median is less affected by
the extreme of the distribution than is the arithmetic mean.
For example, the median of the scores 3, 5, 7, 9, and 9 is 7
while the arithmetic mean is 6.6. If the extreme values are
changed and we now have 4, 6, 7, 12, and 13, the median is
 PSY103 Quantitative Methods in Psychology

till 7 but the arithmetic mean is now 8.4. If the scores are 4,
6, 7, 9, 9, 12, and 13 then the median becomes 9. Thus, the
median is mainly influenced by the number, rather than the
size of the extreme values in a distribution.
3. The median is an easy measure of central tendency to
compute.
The Mode
1. It is the quickest estimate of central tendency available.
2. It is the most typical measure in that it tells the most
occurring in a set of scores.

Study Session Summary

In this Study Session, we have discussed that the mode is the

most occurring value in a set of values. We have also looked
at the importance of the different measures of central
Summary tendency.

Assessment

1. What is the modal family size of the data in

Assignment 11.
2. Calculate the mode of the values in Assignment 12.1.
Forward your findings to your tutor by submitting answers
Assignment
at Assignment 13 page on UI Mobile Class.
Study Session 14 Measures of Variability

Study Session 14

Measures of Variability

Introduction
A measure of central tendency tells much about a distribution but
does not give a complete picture of the distribution. When two
distributions are to be compared, if it is based solely on averages,
quite wrong conclusions may be drawn. Other descriptive
statistical measurements need to accompany averages to amplify
the description of a set of data, hence this Study Session focus on
measures of variability are essential.
When you have studied this session, you should be able to:
 differentiate between measures of central tendency and
measures of dispersion or variability.
Learning  compute and interpret measures of dispersion like
Outcomes range, mean absolute deviation, standard deviation, and
variance . (SAQ 14.1)

A measure of variability tells how spread out or scattered scores

in a distribution are. The measures of variability are needful when
comparing two distributions, for instance, two distributions of
statistical data may be symmetrical and may have the same
means, medians and modes; yet they may differ in the distribution
of the individual scores about the measure of central tendency.
The measures of variability, thus, explain and illustrate the
methods of describing the dispersion and skewness of
distributions by the use of single numbers. For example, consider
the three distributions below:
A B C
8 10 20
8 9 9
8 8 8
8 7 3
3 3 3
 PSY103 Quantitative Methods in Psychology

3 3 2
3 3 1
3 2 1
3 2 0
The arithmetic mean of each of the three distributions is the same,
i.e. 5, while the median and mode is 3 and it is the same for the
three distributions. If we are to compare the three distributions on
the basis of the measures of central tendency, then we may
conclude that the distributions are the same. This may be rather
erroneous because a close observation of the scores in each
distribution shows that they are dispersed differently. For
instance, the lowest score in distribution A is 3 while the largest is
8 and the scores do not vary much from one another. For
distribution B, the lowest is 2 while the largest is 10 and the
scores are more spread out than distribution A. Distribution C has
0 as the lowest score and 20 as the largest. The scores in
distribution C are more dispersed than the other two distributions.
A measure of dispersion or variability is needed for a more
accurate comparison to be made among the distributions. Thus,
the central locations of two distributions may be the same but the
variability may differ. Also, the variability may be the same but
the central locations may differ. Locations and variability are
actually independent. While variability refers to the distance
between each score and any other and is measured with the
measures of variability, the measures of central tendency show
locations in a set of data. A measure of variability show,
numerically, the degree to which scores spread around the
average. It is a measure which tells whether scores cluster closely
around the mean or whether they scatter widely.
Measures of variability are useful in many areas of life, although
it is not frequently reported because it is not familiar to many as
the measures of central tendency. A measure of variability tells us
how representative the average is. If the numerical value of the
measure is small, it means that the individual scores are close to
the average. If the variation score is large, the mean can be used
with less assurance because the scores are rather far from the
mean.

14.1 The Range

The range is the simplest and quickest method of summarising the
variability of a distribution. It is simply defined as the difference
Study Session 14 Measures of Variability

between the highest value, H, and the lowest value, L. This can be
symbolized as:
R=H–L
Where R = range
H = highest value
L = lowest value
For grouped data, the range is calculated by subtracting the lower
limit of the bottom interval from the upper limit of the top
interval. Thus,
R = Ult – Llb
Where R = range
Ult = upper limit of top interval
llb = lower limit of bottom interval
The range is, however, the most unreliable of the measures of
variability. This is because it is determined by only two measures,
with all the other individual values in between having no effect in
the calculation. There is often a gap between those extreme values
used in determining the range and the next highest or lowest value
in a distribution. If these extreme cases are absent from the
distribution then there would be a significant difference in the
calculated range. For instance, consider these two distributions.
Distribution A: 20 16 16 15 15 15 15 14 14 10
Distribution B: 20 19 19 18 17 16 15 14 11 10
The range for the two distributions is 20 - 10, which is 10. An
observation of the values in distribution A shows that the scores
are relatively clustered near the middle of the distribution while
those of distribution B are more spread out. The range is therefore
a weak measure of variability of distribution A. For instance,
when the two extreme scores are eliminated from the two
distributions, the range for distribution A becomes 16 - 14, which
is 2; this is rather far from the earlier range of 10. For distribution
B, the range becomes 19 - 11, which is 8; it is still close to the
initial range of 10. Since this type of change occurs frequently,
the range is considered as a crude measure of variability, just like
the mode, and it is avoided as much as possible in the behavioural
sciences.
Generally, the range is more unreliable with a small sample size
than a large sample size. This is because with a small sample size
the extreme values may be marked by one case. But when the
 PSY103 Quantitative Methods in Psychology

sample size is large, there may be more than one case of each of
the extreme values. Thus, the probability of the range being
altered as a result of missing scores is higher for a smaller sample
than for a larger sample.

14.2 The Mean Absolute Deviation (MAD)

The mean deviation is also referred to as the mean variation. It is
a measure of dispersion which considers the deviation of each
measure in a distribution from a measure of central tendency. The
deviation from the mean, as it is always called, is obtained by
subtracting each value in the distribution (X) from the mean value
(X), of the distribution. If a particular measure is greater than the
average, it will have a positive deviation, with a positive sign (+),
and it is said to have a positive deviation. If the score is less than
the mean value, it has a negative sign (-), and it is said to have a
negative deviation. A measure that is identical with the measure
has a zero deviation. If all the deviations from the mean were
calculated for all the measures in a distribution, and the average of
the sum computed, the resultant value is the mean of deviations
from the mean. If the measures in the distribution were widely
dispersed or scattered, then their deviations would be relatively
large and this would be reflected by the mean of the deviations. If
the measures were clustered or concentrated about the mean of the
distribution, then the mean of the deviations would be small.
Thus, the mean deviation is an excellent measure of dispersion.
In calculating the mean deviation, however, the signs of the
deviations of each measure from the mean are usually ignored.
This is because in a perfectly symmetrical distribution, the
algebraic sum of deviations from the median or mode would be
zero, and such is the case in any distribution; the difference
between the positive and negative deviations (i.e. algebraic sum
of deviations from the mean) must always be zero. Recall that it is
the mean of this sum of deviations that is calculated to obtain the
mean deviation.
Consequently, the deviation of a measure from the mean is
meaningless in this wise, rather it is the absolute deviation from
the mean that is meaningful. By absolute deviation from the
mean, the signs of the deviations are ignored, while the magnitude
of the deviation, that is, the actual value of deviations from the
mean are emphasized. This is referred to as mean absolute
deviation (MAD). Thus, all deviations are treated as positive.
Some examples are now considered for ungrouped and grouped
data.
Study Session 14 Measures of Variability

Table 14.1 shows the raw scores of trainees on a 20-point test.

The interpretation of the mean absolute deviation measure of
variability is that, on the average, the individual scores in the
distribution vary from the mean score of 16.2 by approximately
1.6. Thus, scores within the limits of 1 MAD below the mean and
1 MAD above it will range between 14.6 and 17.8, while 2
MAD range between 13 and 19.4. If you subtract 1.6 from 16.2 it
gives 14.6, while if you add 1.6 to 16.2 you will have 17.8 (i.e. for
1 MAD). For 2 MAD multiply 1.6 by 2, then subtract and
add respectively to 16.2 to have 13 and 19.4 respectively.
Table 14.1 Calculation of Median Absolute Deviation and Standard Deviation
from Ungrouped Data
Col. 1 Col. 2 Col. 3 Col. 4 Col. 5
Score Deviation Absolute
( )2 2
Deviation
(X)
( )
( )

17 0.8 0.8 0.64 289

11 -5.2 5.2 27.04 121
16 -0.2 0.2 0.04 256
17 0.8 0.8 0.64 289
15 -1.2 1.2 1.44 225
18 1.8 1.8 3.24 324
20 3.8 3.8 14.44 400
15 -1.2 1.2 1.44 225
16 -0.2 0.2 0.04 256
17 0.8 0.8 0.64 289

Σ X = 162 ΣX2= 2674

Σ( ) =0 Σ( ) Σ( )2 =
N = 10 = 16.0 49.60

= 16.2

Where MAD = Mean Absolute Deviation

= Individual Score

= an average of the individual scores

 PSY103 Quantitative Methods in Psychology

= the absolute deviation of each score

14.3 The Standard Deviation (S.D.)

This is statistically the most reliable measure of dispersion,
denoted by S for sample and δ for population. It varies less than
any other measure of variability, from sample to sample; the
samples being chosen at random from the same population. The
standard deviation and variance are the most frequently used
measures of dispersion or variability because they have certain
properties that make them preferred to the other measures.
Variability, in actual sense, refers to the extent to which scores
differ from one to another. In statistical terms, it means the
difference between each score and every other score in a
distribution. Since the mean is the most frequently used measure
of central tendency, and best represents all scores in a distribution,
it is more reasonable to define the deviation of a single score from
the other scores as its distance from the mean. This is symbolized
by:

Deviation score =
Where X = individual score in the distribution

= the average of the distribution

When the deviation of a single score from each of the other scores
in a distribution is determined; that is, by computing the
difference between the score and each of the other scores, and the
average of these deviation scores is calculated, it will not be much
different (if at all different) from the deviation of the score from
the mean of the distribution (i.e ).
Thus, the deviation or distance of a single score from the mean is
a representation of all possible deviations or distances of the score
from each of the other scores in the distribution.
When a score is extremely different from the mean, the numerical
deviation score for that score will be large. But scores that are
close to the mean will have small numerical deviation scores. The
word ―numerical‖ is emphasized because it is actually the
magnitude of the deviation that is of interest. The deviation scores
will normally have some positive and some negative values
which, when the deviation scores are added (i. e. Σ(X – ), it is
Study Session 14 Measures of Variability

always equal to zero. Thus, if we are to calculate the average

variability by averaging the deviation scores, that is, by summing
the deviation scores and dividing by N, i.e.

to replace X
Then we will always have a zero score, which will make nonsense
of the whole exercise. To overcome this problem, we have to take
the absolute value of each deviation by ignoring the sign, and then
utilizing its numerical value for the computation of the mean
absolute deviation (MAD). Though the MAD is a descriptive
measure of variability, but it is usually rejected for its inadequacy
in further statistical analysis. This is because absolute values
which are incorporated in MAD are unsuitable for use in
inferential statistical analysis.
The measure which takes care of this problem without using
absolute values, and is useful for further statistical analysis is that
which requires the squaring of each of the deviations before
taking the average of the squared deviations. By squaring the
deviation, the negative signs are taken care of and all the squared
deviations become positive. The sum of the squared deviations
from the mean, that is, Σ(X – )2 is called the sum of squares and
is symbolized by SS. The average of the sum of squares is a
measure of variability, called variance, which is symbolized by δ2.

This is, however, the formula for the variance of a population.

When the data of a sample are to be used in estimating the
variance of the population from which the sample was drawn,
then the sample variance is computed with some slight changes in
the formula:

Whereas the population variance is represented with a square of

the Greek letter sigma (δ2), the sample variance is represented by
; although both are often used interchangeably. For the
population variance, the sum of squared deviations is divided by
N, while that of the sample is divided by N – 1. This is to enable
to be unbiased estimate of the population variance. N – 1 is the
number of deviation about the mean that are free to vary. This
 PSY103 Quantitative Methods in Psychology

number is called the degree of freedom. For reasonably large

samples, however, the actual value of variance is affected a little,
whether we divide by N – 1 or N.
In each of the formulas above, the steps involved are:
1. Subtract the mean from each score.
2. Square each result to eliminate the minus signs
3. Sum the squares
4. Divide the sum by N if population variance, or N – 1 if
sample variance.
The computation of S.D. differs from the M.D. in two ways. First,
in computing S.D. the deviation of each measure from the mean is
squared. Second, the deviations are always taken from the
arithmetic mean, whereas that of the M.D. may be taken from the
median. Also, having squared the deviation these squares are then
summed, and the sum is divided by N. Then the square root is
taken, to give S.D.
Be informed that variance is a square of the original units in
which the mean deviations are based; this is aimed at eliminating
the negative signs that would have yielded a total of zero if the
deviations from the mean were added. The average of these
squared deviations will certainly not produce an average
variability that is the most ideal measure of variability. To
accomplish this, we have to take the positive square root of the
variance, bearing in mind that the variance is the average of
squared deviations. This yields a measure of variability called, the
standard deviation, symbolized by δ for population standard
deviation or S for sample standard deviation. The formulas are:
14.3.1 Calculation of S.D. from Ungrouped Data
Data from Table 14.1

 This formula is used for manual calculation and is called

the definition formula
Study Session 14 Measures of Variability

 This formula is used with a scientific calculator, and is

called the computational formula
Population

 This formula is used for manual calculation, called

definition formula

 This formula is used with a scientific calculator, called

computational formula.
The average deviation of scores from the mean is 2.35 and 2.23
for sample and population respectively, in the distribution.
14.3.2 Calculation of the MAD and S.D. from Grouped Data

Where M.D., f, , and N remain the same

X = the mid-point of each class interval

These formulae are called definition formula. They are used for
manual calculation.

These formulae are called computational formula. They are used

with a scientific calculator.
 PSY103 Quantitative Methods in Psychology

Computation of M.D. and S.D. from a Distribution of English Exam Marks

Table 14.2
(Mean = 59.5)
Col.1 Col.2 Col.3 Col. Col.5 Col.6 Col.7 Col.8 Col.9 Col.10
4
Scores Mid- f X2 fX2
X– (X– )2 f(X- ) f(X- )2
Point fX
(X)

79-83 81 1 81 6561 6561 21.5 462.25 21.5 462.25

74-78 76 3 228 5776 17328 16.5 372.25 49.5 816.75
69-73 71 3 213 5041 15123 11.5 132.25 34.5 396.75
64-68 66 4 264 4356 17424 6.5 42.25 26.0 169.00
59-63 61 7 427 3721 26047 1.5 2.25 10.5 15.75
54-58 56 4 224 3136 12544 3.5 12.25 14.0 49.00
49-53 51 1 51 2601 2601 8.5 72.25 8.5 72.25
44-48 46 4 184 2116 8464 13.5 182.25 54.0 729.00
39-43 41 1 41 1681 1681 18.5 342.25 18.5 342.25
34-38 36 2 72 1296 2592 23.5 552.25 47.0 1104.50

Study Session Summary

In this Study Session, you discovered that the measures of

variability tell us the degree to which scores vary from one
another, and from the mean. The measures include the range,
Summary mean absolute deviation, standard deviation, and variance.

The range is the difference between the highest value and the
lowest value in a distribution. The mean absolute deviation is the
deviation of each measure in a distribution from a measure of
central tendency. The standard deviation is the positive square
root of the average sum of square deviations from the mean, while
the square of this value is the variance. Thus, the average of the
sum of squares is variance.
Study Session 14 Measures of Variability

Assessment

SAQ 14.1
Given the following frequency distribution, find the standard
Assessment deviation of the data.

X 6 7 8 9

F 2 3 3 2

1. The ages of a sample of students is given as:

17, 12, 20, 22, 19, 19, 23, 24, 18, 19, 20 and 21.
Calculate the range, mean absolute deviation, standard
deviation, and variance of these ages and interpret each
Assignment
result.
2. The data below is the accident record of a sample of factory
drivers in one year. It shows the number of accidents each
driver had in the year.

Number of 9 8 7 6 5 4 3 2
Accidents

Drivers involved 2 2 6 5 3 4 5 4

Calculate and interpret the dispersions in the accident data.

Send your findings to your tutor for feedback at Study Session 14
Assignment Page at UI Mobile Class
 PSY103 Quantitative Methods in Psychology

Study Session 15

Transformed Scores

Introduction
If you are informed that you have obtained 75% in PSY 103
examination, you will definitely need additional information in
order to determine how well you performed, compared with other
students who took the examination. If the examination was
generally easy for most students and there were many high scores,
then your score of 75% may be an average or even below average
score. If, on the other hand, the examination was a little bit
difficult for the whole class, then your score may be among the
highest or even the highest. One way of getting this additional
information is to transform the original score to show at a glance
how well you performed in comparison to other students in the
course. Thus, this Study Session addresses the different ways of
transforming scores to allow for relative comparison of an
individual‘s scores across different tasks, as well as relative
comparisons of individuals‘ scores on the same task.

When you have studied this session, you should be able to:
 compute transformed scores.
 interpret percentile, standard score (Z score), and T
Learning score.
Outcomes

15.1 Percentiles
The percentile rank of a score is a single number which gives the
percentage of cases in the specific reference group, scoring at or
below that score. If, for example, your raw score in an
examination is 75%, and it corresponds to a percentile rank of 65,
it means that 65% of the class obtained equal or lower scores than
you did, while 35% of the class received higher scores. A
percentile is, thus, the score at or below which a given percent of
the case lie. The percentile shows directly how an individual score
compares to the scores of a specific group. It is important to keep
in mind that a percentile rank cannot be correctly interpreted,
unless the reference group is taken into consideration.
Study Session 15 Transformed Scores

15.1.1 Computational Procedure: Computing the

Corresponding Percentile Rank from a given Raw Score
To illustrate the computation of percentile ranks, reference will be
made to Table 15.1. Our task is to find the percentile rank
corresponding to the raw score of 43.
Table 15.1 Hypothetical Examination Scores for Students in PSY 103 Test:
Transferring a raw score of 43 to a percentile rank
Class Interval Frequency Cumulative Frequency (CF)
48 – 50 1 87
45 – 47 5 86 6
42 - 44 6 81
Critical Interval
39 – 41 8 75
36 – 38 9 67
33 – 35 14 58
30 – 32 12 44
27 – 29 9 32 75
24 – 26 7 23
21 – 23 5 16
18 – 20 5 11
15 – 17 6 6
F Percent (= f/N)
All higher intervals 6 6/87 = 6.9% (H %)
Critical interval (42 – 44) 6 6/87 = 6.9% (I %)
All lower intervals 75 75/87 = 86.2% (L %)

Step 1: Locate the class interval wherein the raw score falls; call
this the ―critical interval‖.
Step 2: Group the frequencies (F) into three categories,
embracing those with scores higher than the critical interval, those
corresponding to all scores in the interval, and those
corresponding to all scores lower than the critical interval. These
are 6, 6, and 75 respectively.
Step 3: Each frequency is then converted to a percent by dividing
by N; in this case 87. We will denote the percent of people
scoring in interval higher than the critical interval by H% (for
higher), the percent of people scoring in the critical interval by I%
 PSY103 Quantitative Methods in Psychology

(for in), and the percent of people scoring lower than the critical
interval by L% (for lower)
Step 4: It is clear at a glance that the score of 43 is better than at
least 86.2% of the scores, i.e., those below the critical interval.
Thus, the percentile rank must be at least 86.2%. It is also clear
that 6.9% of the scores are better than 43, i.e. the ones above the
critical interval. However, it is not evident whether the raw score
of 43 is higher than all the scores that fall within the critical
interval, or it is lower than, since the interval carries 6.9% of the
entire scores. The solution is to look at the score in comparison to
the size of the interval; the higher the score in relation to the
critical interval the more people in that interval is assumed to
have been outscored.
Step 5: To determine the standing of score 43 in the critical
interval, first ascertain the lower real limit of the interval, which is
41.5. Subtract this from the raw score of 43 (43 – 41.5) = 1.5.
Since the interval size is 3, this distance is expressed as a fraction,
and is equal to 1.5 points/3 points, = 0.5 of the interval. Thus,
score 43 is 0.5 of the 6.9% of students in the critical interval. This
gives a value of 3.45, which should be added to 86.2% (i.e. L %).
The final result is 89.65%; implying that 89.65% of those that
took the examination had 43 or less, which is the percentile rank.
The percentile rank is determined from the formula:

L% and I% remain the same

Score = raw score in question
LRL = Lower real limit of critical interval
H = interval size

The percentile rank is therefore 89.65%, indicating that

approximately 90% of the class received equal or lower scores
and only about 10% received higher scores.
Study Session 15 Transformed Scores

15.1.2 Computational Procedure: Computing the

Corresponding Raw Score from a Given Percentile
This is a reverse of the former, requiring finding the raw score
that corresponds to a specified percentile value. For example,
suppose the Study Session wishes to give a special teaching to the
bottom 32% of the class; what raw score should be used as the
cutting line? In this case the percentile (32%) is specified and the
raw score is needed. The steps to adopt are as follows:
Step 1: Convert the percentile to a case number by multiplying
the percentile by N. (i.e. .32 x 87 = 27.84). Thus, the score that
corresponds to the individual whose rank is 27.84 from the
bottom of the class (i.e. the person scoring at the 32nd percentile)
is the cutting line needed.
Step 2: Find the interval in which the case number computed in
step 1 falls. This is easily accomplished by starting at the bottom
of the cumulative frequency distribution and proceeding upward
until you find the first value equal to, or greater than the critical
case (27.8); the corresponding interval is the critical interval. This
is illustrated with data on Table 15.2.

Table 15.2 Hypothetical Examination Scores for Students in Psy 103

Test: Finding the raw score corresponding to the 32nd
percentile
Class Interval Frequency Cumulative Frequency
(CF)
48 – 50 1 87
45 – 47 5 86
42 - 44 6 81
39 – 41 8 75
36 – 38 9 67
33 – 35 14 58
30 – 32 12 44
27 – 29 9 32 first
of Z = 27.84
24 – 26 7 23
21 – 23 5 16
18 – 20 5 11
15 – 17 6 6
 PSY103 Quantitative Methods in Psychology

pN = .32 x 87 = 27.84

Step 3: The 27.8th case must have a score of at least 26.5; the
lower real limit of the interval in which it appears. However, the
critical value covers three score points (26.5 – 29.5); what point
value corresponds to the 32nd percentile?
There are 23 cases below the critical interval, so that the 27.84th
case needs 27.84 – 23, i.e. 4.84 cases up in the interval. The total
number of cases in the interval is 9, so this distance expressed as a
fraction is 4.84 cases /9 cases = .54. Therefore, in addition to the
lower real limit of 26.5, .54 of the three points included in the
critical interval must be added in order to determine the point
corresponding to the 27.84th case. The cutting score is therefore
equal to 26.5 + (.54 x 3) = 26.5 + 1.62 = 28.12.
A general formula to obtain this is:

Where
Score p = Score corresponding to the pth percentile
LRL – lower real limit of critical interval
P – specified percentile
N – total number of cases
SFB – sum of frequencies below critical interval
F – frequency within critical interval
H – interval size

Thus, people with scores 28 or less need special teaching, while

students with scores of 29 and above do not need.
Study Session 15 Transformed Scores

15.2 ―Z and T‖ Scores

These scores help us to derive a transformed score that shows at a
glance the relationship of an original raw score to the mean, using
the standard deviation of the reference group as the unit of
measurement. Suppose a student obtains a raw score of 70, 65,
and 55 in tests on PSY 101, PSY 102, and PSY 103 respectively,
it might seem as though the student‘s best score is in PSY 101 and
the poorest is in PSY 103. However, the raw scores of 70, 65, and
55 cannot be compared directly because they come from different
distributions with different means and different standard
deviations. Thus, the units of measurement are not the same from
test to test. You can overcome this by transforming the scores on
each test to a common scale with a specified mean and standard
deviation. This new scale would then enable the transformed
scores of different tests to be compared directly.
15.2.1 Standard Scores (Z Scores)
A procedure is to convert the original scores to new scores called
Z scores or standard scores, with a mean of 0 and a standard
deviation of 1. Standard scores have two major advantages. Since
the mean is zero, it can be said at a glance whether a given score
is above or below average; an above average score is positive and
a below average score is negative. Also, since the standard
deviation is 1, the numerical size of a standard score indicates
how many standard deviations above or below average the score
is; a score of one standard deviation above average (i.e. a standard
score of +1) would demarcate approximately the top 16% in a
normal distribution, while a score of two standard deviations
above average (a standard score of +2) would demarcate
approximately 21/2% in a normal distribution.
You can use this formula to convert a set of scores to standard
scores:

Where Z = standard score

X = raw data

= mean score

= standard deviation
Converting each of the original test scores to Z scores yield:
 PSY103 Quantitative Methods in Psychology

Psy101 Psy102 Psy103

X 75 65 55

80 60 50

10 15 5

The standard scores show at a glance that the student was half
standard deviation below the mean in PSY 101, less than half
standard deviation above the mean in PSY 102, and one standard
deviation above the mean in PSY 103. When raw scores are
transformed into Z scores, the shape of the distribution remains
the same. The Z scores give an accurate picture of the standing of
each score relative to the reference group, no matter where the
original scores are measured from or what scale is used. Standard
scores are used extensively in the behavioural sciences.
15.2.2 T Scores
Standard scores are a little bit difficult to explain to one not well
versed in statistics. Since behavioural scientists try to report test
scores to people who are not statistically sophisticated, several
alternatives to Z scores have been developed. One of such
alternative, called T scores, is defined as a set of scores with a
mean of 50 and a standard deviation of 10. The T scores are
obtained from this formula:
T = 10Z + 50
For example, the score of 55 in the example is converted to a Z
score of +1.00 by using the Z formula. Then T is equal to (10)
(+1.00) + 50 = 60.
Since the mean of T scores is 50 it can still be seen at a glance
whether a score is above average (it will be greater than 50) or
below average (it will be less than 50). Also one can tell how
many standard deviations above or below average a score is. For
example, a score of 40 is exactly one standard deviation below
average (i.e. a Z score of -1.00) since the standard deviation of T
scores is 10. But the T score of 60 confirms the Z score of +1.00
which is interpreted as one standard deviation above the mean.
Study Session 15 Transformed Scores

Study Session Summary

In this Study session you learned that transformed scores

allow for relative comparison of an individual‘s scores across
different tasks, as well as relative comparisons of individuals‘
Summary scores on the same task. You also learnt that the percentile
rank of a score is a single number which gives the percent of
cases in the specific reference group, scoring at or below that
score. Z score or standard score is a set of scores with a mean
of 0 and a standard deviation of 1. T score is a set of scores
with a mean of 50 and a standard deviation of 10.

Assessment

1. A psychologist administered three psychological tests

on 10 job applicants and had the following scores:
Applicant Tests
Assignment A B C
1 12 20 9
2 17 27 7
3 11 13 6
4 18 22 4
5 10 14 6
6 16 21 7
7 14 19 5
8 15 23 9
9 16 19 5
10 13 14 3
In what test did each applicant performed best?
 PSY103 Quantitative Methods in Psychology

References

Guilford, J.P. and Fruchter, B. (1986). Fundamental

statistics in Psychology and Education. London: MacGraw-Hill
Book Company.
Howitt, D. and Cramer, D. (1977). An Introduction to
Statistics in Psychology: A Complete Guide for Students.
London: Prentice Hall.
Welkowitz, J., Ewen, R.B. and Cohen, J. (2002).
Introductory Statistics for the Behavioural Sciences. USA: John
Wiley & Sons, Inc.
Feedback on SAQs Transformed Scores

Feedback on SAQs

SAQ1.1 A layperson‘s perspective views statistics ―as collection of data‖,

while empirical researcher‘s perspective differs on the bases of
―planning and reporting data in line with its specific objective at
hand‖. In grammatical terms however, statistics is ―a summary
provided after adequate analysis has been made on collected
data‖.
SAQ1.2 She would be using descriptive statistics.
SAQ 2.1 Your answers probably include the following:
1. Statistics will enable you to understand professional literature
like journals
2. Knowledge of statistics will maximize your comprehension of
inferences drawn in research reports, and you will be skilled to
make genuine inferences from your research.
3. As scientists, it will help you progress in any quantitative
scientific pursuit.
4. Statistics will help you to evaluate the strengths and
weaknesses inherent in any research, in terms of the
techniques adopted by you, the researcher in collecting
information and drawing inferences.
5. The knowledge of statistics will help you to make less
subjective conclusions.
6. It will enable you to easily make comparisons and
relationships via the use of figures which is much clearer than
verbal descriptions.
SAQ 3.1 Valid options to the exercise are as follows:
A. population
B. population parameter
C. sample / representative sample
D. random sample
E. inferential statistics
SAQ 4.1 The correct answer is C because the ratio of 3:2 means that if you
divide the population into groups of five members (3+2), each
group would be made up of 3 males and 2 females. Therefore, to
draw a proportionate representative sample of each of the sexes
from this population, the desired sample size of 200 similarly
needs to be divided into groups of fives (200/5). This gives you
40 groups, each of which is made up of 3 males and 2 females
similar to the population from which they have been drawn. Thus,
 PSY103 Quantitative Methods in Psychology

3 males x 40 groups = 120 males, and 2 females x 40 groups = 80

females.
If you chose A, you would be wrong because 3+2 is not equal to
4. (150:50 is a ratio of 3:1).
If you chose B, you would be wrong. Although 3+2 = 5, you may
have wrongly swapped the relative proportions of male : female
ratios
If you chose D, you would be wrong. Again, 3+2 is not equal to 4.
(50:150 is a ratio of 1:3).

SAQ 4.2 The correct answer is B because the nth term is obtained by
dividing the sample frame or population (600) by the desired
sample size (200), giving 30.
If you chose A, you would be wrong because you may have
assumed that the nth term was the sample size.
If you chose C, you would be wrong. You may have taken the
sample frame or population as the nth term.
If you chose D, you would be wrong. You may have subtracted
the sample size from the sample frame or population.

SAQ 5.1 We do not know what you have provided in your table, but it may
be filled like as shown below:

Scale of Uses of the Scale Example

Measurement
COUNT the
number of things Pets: 3 dogs, 12
Nominal
within different goats
categories
RANK some
things as having
more of Speed (measured
something than by place of finish
Ordinal
others (but NOT in a race): 1st,
QUANTIFY how 2nd, 3rd, etc.
much of it they
have)
QUANTIFY how Temperature: −2°
Interval much of F, 98° F, 57° F;
something there 0° F is not the
Feedback on SAQs Transformed Scores

is but a score of absence of heat

zero does not
mean the absence
of the thing being
measured
QUANTIFY how
much of Number of text
something there messages sent in
is and a score of a day: 0, 30, 15,
Ratio
zero means the etc.
absence of the
thing being
measured

SAQ 6.1 A Frequency Distribution Table

Speeds Tally Frequency

23 //// 4
24 /// 3
25 //// 4
26 / 1
27 // 2
28 /// 3
29 // 2
30 //// / 6
 PSY103 Quantitative Methods in Psychology

SAQ 7.1 Frequency Distribution Bar Graph (Car Speed)

SAQ 8.1 Pie chart showing the monthly spending of Mrs Halima

SAQ 9.1 The answer is (b). The cumulative frequency distribution enables
the calculation of the number of cases or people falling at, below,
or above given scores in the distribution.
SAQ 13.1 Variance formula is:

And the standard deviation is:

Feedback on SAQs Transformed Scores
 PSY103 Quantitative Methods in Psychology

Study Session 16
The Normal Distribution
Introduction
This study session presents the normal distribution and the skewed
distributions. It explains a normal distribution and its graphical
presentation. The empirical rule and parameter of the normal
distribution are presented, in terms of the mean and standard deviation.
The properties of the normal distribution are compared to those of the
skewed distributions and the locations of the three measures of central
tendency on the normal curve, positively skewed and negatively
skewed distributions are given

16.1 What is a Normal Distribution?

The normal distribution is also known as a Gaussian distribution is the

most significant continuous probability distribution. Sometimes it is
also called a bell curve. It is a type of continuous probability
distribution in which most data points cluster toward the middle of the
range, while the rest taper off symmetrically toward either extreme. It is
symmetric about the mean and indicates that values near the mean
occur more frequently than the values that are farther away from the
mean. The middle of the range is also known as the mean of the
distribution.

16.2 Normal Distribution Explained Graphically

Graphically, a normal distribution is a bell curve because of its flared
shape (see Fig 16.1). The precise shape can vary according to the
distribution of the values within the population. The population is the
entire set of data points that are part of the distribution. Regardless of
its exact shape, a normal distribution bell curve is always symmetrical
about the mean. A symmetrical distribution means that a vertical
dividing line drawn through the maximum/mean value will produce
two mirror images on either side of the line, in which half the
population is less than the mean and half is greater. However, the
reverse is not always true; that is, not all symmetrical distributions are
normal. In the bell curve, the peak is always in the middle, and
the mean, mode and median are all the same, coinciding at the middle
of the curve.
Feedback on SAQs Transformed Scores

Fig 16.1 The Normal Curve

16.2. 1 Basic examples of normal distribution: Height and

weight
Height is one simple example of values that follow a normal
distribution pattern. Most people are of average height; whatever that
may be for a given population. If the heights of these people are
represented in graphical format along with the heights of people who
are taller and shorter than the average, the distribution will always be a
normal distribution. This is because the people of average height will
be clustered near the middle, while those who are taller and shorter will
be farther away. Further, these latter groups will consist of very small
numbers of people. The number of people who are extremely tall or
extremely short will be even smaller, so they will be the farthest away
from the mean.

Similarly, weight can also follow a normal distribution if the average

weight of the population under consideration is known. Like height, the
weight outliers will be those who weigh more or less than the average.
The bigger the deviation from the average, the farther away those data
points will be on the distribution graph.

16.3 Normal distribution empirical rule

The empirical rule for normal distributions describes where most of the
data in a normal distribution will appear (see Fig 16.2), and it states the
following:
68.26% of the observations will appear within +/-1 standard deviation
of the mean;
95.44% of the observations will fall within +/-2 standard deviations;
and
99.7% of the observations will fall within +/-3 standard deviations.
All data points falling outside of three standard deviations (3σ) indicate
rare occurrences.
 PSY103 Quantitative Methods in Psychology

Fig 16. 2 Normal distribution empirical rule

16.4 Parameters of normal distribution

In a normal distribution, the mean, mode and median are the same;
there is no need to calculate them separately. These values represent the
distribution's highest point, or the peak. All other values in the
distribution then fall symmetrically around the mean. The width of the
mean is defined by the standard deviation, as seen in Fig 16.2. In fact,
only two parameters are required to describe a normal distribution: the
mean and the standard deviation.

(a) The mean

The mean is the central highest value of the bell curve. All other values
in the distribution either cluster around it or are at some distance away
from it. Changing the mean on a graph will shift the entire curve along
the x-axis, either toward the left or toward the right. However, its
symmetricity will still be maintained.

(b) The standard deviation

In general, standard deviation is a measure of variability in a
distribution. In a bell curve, it defines the width of the distribution and
shows how far away from the mean the other values fall. In addition, it
represents the typical distance between the average and the
observations.

Changing the standard deviation will change the distribution of values

around the mean. A smaller deviation will reduce the spread, i.e.
tightening the distribution, while a larger deviation will increase the
spread and produce a wider distribution. As the distribution gets wider,
it becomes more likely that values will be farther away from the mean.

16.5 Normal Distribution Standard Deviation

Feedback on SAQs Transformed Scores

Generally, the normal distribution has any positive standard deviation.

We know that the mean helps to determine the line of symmetry of a
graph, whereas the standard deviation helps to know how far the data
are spread out. If the standard deviation is smaller, the data are
somewhat close to each other and the graph becomes narrower. If the
standard deviation is larger, the data are dispersed more, and the graph
becomes wider. The standard deviations are used to subdivide the area
under the normal curve. Each subdivided section defines the percentage
of data, which falls into the specific region of a graph.
Using 1 standard deviation, the Empirical Rule states that,

 Approximately 68% of the data falls within one standard

deviation of the mean. (i.e., Between Mean- one Standard
Deviation and Mean + one standard deviation)
 Approximately 95% of the data falls within two standard
deviations of the mean. (i.e., Between Mean- two Standard
Deviation and Mean + two standard deviations)
 Approximately 99.7% of the data fall within three standard
deviations of the mean. (i.e., Between Mean- three Standard
Deviation and Mean + three standard deviations)
 Thus, the empirical rule is also called the 68 – 95 – 99.7 rule.

Fig 16.3 Normal distribution standard deviation

16.6 Normal Distribution Properties

Some of the important properties of the normal distribution are listed
below:
 PSY103 Quantitative Methods in Psychology

 In a normal distribution, the mean, median and mode are

equal.(i.e., Mean = Median= Mode).
 The total area under the curve should be equal to 1.
 The normally distributed curve should be symmetric at the
centre.
 There should be exactly half of the values are to the right of the
centre and exactly half of the values are to the left of the centre.
 The normal distribution should be defined by the mean and
standard deviation.
 The normal distribution curve must have only one peak. (i.e.,
Unimodal)
 The curve approaches the x-axis, but it never touches, and it
extends farther away from the mean.

16.7 Skewness and kurtosis in a normal distribution

Skewness represents a distribution's degree of symmetry. Since the
normal distribution is perfectly symmetric, it has a skewness of zero. In
other distributions with a skewness less than or greater than zero, the
left tail (left skewness) or the right tail (right skewness) will be longer,
respectively.

Kurtosis measures the thickness of each tail end of a distribution vis-à-

vis the tails of a normal distribution. For a normal distribution, kurtosis
is always equal to 3. In a distribution with kurtosis greater than 3, the
tail data will exceed the tails of the normal distribution, resulting in a
phenomenon called fat tails. Distributions with kurtosis less than 3
show tails that are skinnier than the tails of a normal distribution.

16.8 Skewed Distribution

A skewed distribution occurs when one tail is longer than the other.
Skewness defines the asymmetry of a distribution. Unlike the familiar
normal distribution with its bell-shaped curve, these distributions are
asymmetric. The two halves of the distribution are not mirror images
because the data are not distributed equally on both sides of the
distribution‘s peak.

16.8.1 How to Tell if a Distribution is Left Skewed or Right Skewed

Let‘s start by contrasting characteristics of the symmetrical normal

distribution with skewed distributions.
Feedback on SAQs Transformed Scores

The normal distribution has a central peak where most observations

occur, and the probability of events tapers off equally in both the
positive and negative directions on the X-axis. Both halves contain
equal numbers of observations. Unusual values are equally likely in
both tails.

However, that‘s not the case with asymmetrical distributions where

probabilities decrease more slowly in one direction relative to the other.
In other words, extreme values that fall further away from the peak are
more likely to occur in one tail than the other. That‘s why you‘ll hear
about left and right skewed distributions, also known as negatively and
positively skewed distributions.

Normal curve Right Skewed Left-Skewed

(Positively Skewed)
(Negatively Skewed)

Fig. 16.4 Symmetric and Asymmetric distributions

Right skewed distributions, also called positively skewed, occur when

the long tail is on the right side of the distribution. This condition
occurs because probabilities taper off more slowly for higher values.
Consequently, you‘ll find extreme values far from the peak on the high
end more frequently than on the low. For example, income and wealth
are classic examples of right skewed distributions. Most people earn a
modest amount, but some millionaires and billionaires extend the right
tail into very high values. Meanwhile, the left tail cannot be less than
zero. This situation creates a positive skew. Consequently, reports
frequently refer to median incomes because the mean overestimates the
most common values.

Left skewed distributions, also called negatively skewed, occur when

the long tail is on the left side of the distribution. This condition occurs
because probabilities taper off more slowly for lower values. Therefore,
 PSY103 Quantitative Methods in Psychology

you‘ll find extreme values far from the peak on the low side more
frequently than the high side.

The normal distribution has a central peak where most observations

occur, and the probability of events tapers off equally in both the
positive and negative directions on the X-axis. Both halves contain
equal numbers of observations. Unusual values are equally likely in
both tails.
However, that‘s not the case with asymmetrical distributions where
probabilities decrease more slowly in one direction relative to the other.
In other words, extreme values that fall further away from the peak are
more likely to occur in one tail than the other.

16.8.2 Skewed Distributions and the Mean, Median, and Mode

The mean, median, and mode are all equal in the normal distribution
and other symmetric distributions.

Because the mean over or underestimates the most frequently occurring

values in asymmetric distributions, analysts often use the median in
these cases. The median is a more robust statistic in the presence of
extreme values. For a positively skewed distribution, the median
position is at the middle point of the cureve and higher than the mode,
while the mean is higher than the media and mode. For a negatively
skewed distribution the median is at the middle of the distribution and
higher than the mean, while the mode is higher than the median and the
mean.

Questions

[Link] and thre characteristics of a normal distribution

2. Name any three characteristics of a positively skewed distribution

3. Name any three characteristics of a negatively skewed distribution

Feedback on SAQs Transformed Scores

4. Present a graph of a normal curve showing the locations of the mean, mode,
and median

5. Present a graph of a negatively skewed distribution showing the three

measures of central tendency