Study Manuals
Certificate in
INTRODUCTION TO
QUANTITATIVE
METHODS
A
B
E
S
T
U
D
Y
M
A
N
U
A
L
I
N
T
R
O
D
U
C
T
I
O
N
T
O
Q
U
A
N
T
I
T
A
T
I
V
E
M
E
T
H
O
D
S
5th Floor, CI Tower • St Georges Square • High Street • New Malden
Surrey KT3 4TE • United Kingdom
Tel: + 44(0)20 8329 2930 • Fax: + 44(0)20 8329 2945
E-mail: info@abeuk.com • www.abeuk.com
i
INTRODUCTION TO QUANTITATIVE METHODS
Contents
Unit Title Page
Introduction to the Manual iii
Syllabus v
Formulae, rules and tables provided with examination paper vii
1 Basic Numerical Concepts 1
Introduction 3
Number Systems 4
Numbers – Approximation and Integers 5
Arithmetic 7
Dealing with Negative Numbers 11
Fractions 13
Decimals 22
Percentages 26
Ratios 29
Further Key Concepts 31
2 Algebra, Equations and Formulae 39
Introduction 40
Some Introductory Definitions 40
Algebraic Notation 43
Solving Equations 50
Formulae 55
Introduction 64
Currency and Rates of Exchange 64
Discounts and Commission 68
Simple and Compound Interest 70
Depreciation 76
Pay and Taxation 79
4 Simultaneous, Quadratic, Exponential and Logarithmic Equations 89
Introduction 90
Simultaneous Equations 91
Formulating Problems as Simultaneous Equations 96
Logarithmic and Exponential Functions 106
ii
Unit Title Page
5 Introduction to Graphs 119
Introduction 120
Basic Principles of Graphs 121
Graphing the Functions of a Variable 124
Graphs of Equations 126
Using Graphs in Business Problems 147
6 Introduction to Statistics and Data Analysis 171
Introduction 172
Data for Statistics 173
Methods of Collecting Data 175
Classification and Tabulation of Data 177
Frequency Distributions 185
7 Graphical Representation of Information 193
Introduction 194
Graphical Representation of Frequency Distributions 194
Pictograms 204
Pie Charts 206
Bar Charts 207
Scatter Diagrams 209
General Rules for Graphical Presentation 210
8 Measures of Location 221
Introduction 222
The Arithmetic Mean 223
The Mode 232
The Median 236
9 Measures of Dispersion 249
Introduction 250
Range 251
The Quartile Deviation, Deciles and Percentiles 251
Standard Deviation 255
The Coefficient of Variation 262
Skewness 263
10 Probability 275
Introduction 276
Estimating Probabilities 277
Types of Event 279
The Two Laws of Probability 280
Conditional Probability 286
Tree Diagrams 288
Sample Space 294
Expected Value 295
The Normal Distribution 298
iii
Introduction to the Study Manual
Welcome to this study manual for Introduction to Quantitative Methods.
The manual has been specially written to assist you in your studies for the ABE Certificate in
Business Management and is designed to meet the learning outcomes specified for this
module in the syllabus. As such, it provides a thorough introduction to each subject area and
guides you through the various topics which you will need to understand. However, it is not
intended to "stand alone" as the only source of information in studying the module, and we
set out below some guidance on additional resources which you should use to help in
preparing for the examination.
The syllabus for the module is set out on the following pages and you should read this
carefully so that you understand the scope of the module and what you will be required to
know for the examination. Also included in the syllabus are details of the method of
assessment – the examination – and the books recommended as additional reading.
Following the syllabus, we also set out the additional material which will be supplied with the
examination paper for this module. This is provided here for reference only, to help you
understand the scope of the syllabus, and you will find the various formulae and rules given
there fully explained later in the manual.
The main study material then follows in the form of a number of study units as shown in the
contents. Each of these units is concerned with one topic area and takes you through all the
key elements of that area, step by step. You should work carefully through each study unit in
turn, tackling any questions or activities as they occur, and ensuring that you fully understand
everything that has been covered before moving on to the next unit. You will also find it very
you have completed the study unit.
 ABE website – www.abeuk.com. You should ensure that you refer to the Members
Area of the website from time to time for advice and guidance on studying and
preparing for the examination. We shall be publishing articles which provide general
guidance to all students and, where appropriate, also give specific information about
particular modules, including updates to the recommended reading and to the study
units themselves.
 Additional reading – It is important you do not rely solely on this manual to gain the
information needed for the examination on this module. You should, therefore, study
some other books to help develop your understanding of the topics under
consideration. The main books recommended to support this manual are included in
the syllabus which follows, but you should also refer to the ABE website for further
details of additional reading which may be published from time to time.
 Newspapers – You should get into the habit of reading a good quality newspaper on a
regular basis to ensure that you keep up to date with any developments which may be
relevant to the subjects in this module.
 Your college tutor – If you are studying through a college, you should use your tutors to
help with any areas of the syllabus with which you are having difficulty. That is what
they are there for! Do not be afraid to approach your tutor for this module to seek
clarification on any issue, as they will want you to succeed as much as you want to.
 Your own personal experience – The ABE examinations are not just about learning lots
of facts, concepts and ideas from the study manual and other books. They are also
about how these are applied in the real world and you should always think how the
topics under consideration relate to your own work and to the situation at your own
workplace and others with which you are familiar. Using your own experience in this
iv
way should help to develop your understanding by appreciating the practical
application and significance of what you read, and make your studies relevant to your
personal development at work. It should also provide you with examples which can be
And finally …
We hope you enjoy your studies and find them useful not just for preparing for the
examination, but also in understanding the modern world of business and in developing in
your own job. We wish you every success in your studies and in the examination for this
module.
September 2008
Unit Title: Introduction to Quantitative Methods Unit Code: IQM
Level: 3 Learning Hours: 100
Learning Outcomes and Indicative Content:

Candidates will be able to:
1. Demonstrate the rules of numeracy

1.1 Apply the four rules to whole numbers, fractions and decimals
1.2 Express numbers in standard form
1.3 Multiply and divide negative numbers

2. Apply calculations

2.1 Compare numbers using ratios, proportions and percentages;
2.2 Obtain values for simple financial transactions involving
purchases, wages, taxation, discounts
2
p g
purchases, wages, taxation, discounts
2.3 Determine values for simple and compound interest
depreciation
2.4 Convert foreign currency
2.5 Make calculations using a scientific calculator including roots
and powers; logarithms and exponential values
2.6 Evaluate terms involving a sequence of operations and use of
brackets
2.7 Interpret, transpose and evaluate formulae
2.8 Approximate data using rounding, significant figures

3. Use algebraic methods

3.1 Solve linear and simultaneous equations
3.2 Solve quadratic equations using factorisation and formulae
3.3 Solve equations using roots or logarithms
3.4 Determine the equation of a straight line through two points and
also when given one point and its gradient
3.5 Determine the gradient and intercepts on the x or y axes for a
straight line

4. Construct and use: graphs, charts and diagrams

4.1 Draw charts and diagrams derived from tabular data: eg bar
charts, pie charts, scatter diagrams
4.2 Plot graphs, applying general rules and principles of graphical
construction including axes, choice of scale and zero
4.3 Plot and interpret mathematical graphs for simple linear,
4.4 Identify points of importance on graphs eg maximum, minimum
and where they cut co-ordinate axes

Determine values for simple and compound interest, and for
depreciation of an asset using the straight line method and the
reducing balance method
2.3
v
vi
INTRODUCTION TO QUANTITATIVE METHODS
FORMULAE FOR BUSINESS MATHEMATICS AND STATISTICS
INTEREST
The formula for calculating compound interest:
where: A = Accrued amount
P = Original principal
r = Rate of interest (for a particular time period, usually annual)
n = Number of time periods.
DEPRECIATION
● Straight-line method:
Cost of asset
Annual depreciation = ––––––––––––
Useful life
(Cost of asset) – (Value at end of useful life)
or Annual depreciation = ––––––––––––––––––––––––––––––––––––
Useful life
● Reducing balance method:
D = B(1 – i )
n
where: D = Depreciated value at the end of the nth time period
B = Original value at beginning of time period
i = Depreciation rate (as a proportion)
n = Number of time periods (normally years)
STRAIGHT LINE
A linear function is one for which, when the relationship is plotted on a graph, a straight line is
obtained.
The expression of a linear function, and hence the formula of a straight line, takes the following
form:
y = mx + c
Note that: c = the y intercept (the point where the line crosses the y axis)
m = the gradient (or slope) of the line
A P
r
n
= +

1
100
vii
A quadratic equation of the form ax
2
+ bx + c = 0 can be solved using the following formula:
RULES FOR LOGARITHMS
1. log(p × q) = log p + log q
2.
3. log p
n
= n log p
4. If y = ax
n
then n = (log y – log a) ÷ log x
PROBABILITY
● Probability rules:
Probability limits: 0 ≤ P(A) ≤ 1
Total probability rule: ΣP = 1 (for all outcomes)
For complementary events: P(A) + P(A

) = 1
For two mutually exclusive events: P(A and B) = 0
For independent events: P(A) = P(A| B) and/or P(B) = P(B| A)
● Multiplication rules:
For independent events: P(A and B) = P(A)P(B)
For dependent events: P(A and B) = P(A)P(B| A)
For mutually exclusive events: P(A or B) = P(A) + P(B)
For non-mutually exclusive events: P(A or B) = P(A) + P(B) – P(A and B)
● Conditional rules:
P(A and B) P(A and B)
P(A| B) = –––––––––– and P(B| A) = –––––––––
P(B) P(A)
Expected value of variables x with associated probabilities P(x) is E(x) = ΣxP(x)
log log log
p
q
p q

= –
x
b b ac
a
=
± – –
2
4
2
viii
STATISTICAL MEASURES
● Mean for ungrouped data: and
where N and x

are the population and sample means respectively.
● Mean for grouped data: μ = Σmf/N and x

= Σmf/n
where m is the midpoint and f is the frequency of a class.
● Median for ungrouped data:
= Value of the th observation in a ranked data set, where the number of
= observations is odd and where n is the number of observations.
● Range = Largest value – Smallest value.
● Standard deviation for ungrouped data:
and
where σ and s are the population and sample standard deviations respectively.
● Standard deviation for grouped data:
and
● Pearson’s measure of skewness:
Mean – Mode 3(Mean – Median)
Psk = ––––––––––––––– or ––––––––––––––––
Standard deviation Standard deviation
μ = ∑
x
N
= ∑ x
x
n
n +

1
2
σ =

( )
∑x
x
N
N
2
2

=

( )
∑x
x
n
n
2
2
1

s
σ =

( )
∑m f
mf
N
N
2
2

=

( )
∑m f
mf
n
n
2
2
1

s
ix
STANDARD NORMAL DISTRIBUTION
The table of values of the standard normal distribution set out below provides a means of
determining the probability of an observation (x ) lying within specified standard deviations (σ) of
the mean of the distribution (μ).
0.0 .5000 .4960 .4920 .4880 .4840 .4801 .4761 .4721 .4681 .4641
0.1 .4602 .4562 .4522 .4483 .4443 .4404 .4364 .4325 .4286 .4247
0.2 .4207 .4168 .4129 .4090 .4052 .4013 .3874 .3936 .3897 .3859
0.3 .3821 .3783 .3745 .3707 .3669 .3632 .3594 .3557 .3520 .3483
0.4 .3446 .3409 .3372 .3336 .3300 .3264 .3228 .3192 .3156 .3121
0.5 .3085 .3050 .3015 .2981 .2946 .2912 .2877 .2843 .2810 .2776
0.6 .2743 .2709 .2676 .2643 .2611 .2578 .2546 .2514 .2483 .2451
0.7 .2420 .2389 .2358 .2327 .2296 .2266 .2236 .2206 .2177 .2148
0.8 .2119 .2090 .2061 .2033 .2005 .1977 .1949 .1922 .1894 .1867
0.9 .1841 .1814 .1788 .1762 .1736 .1711 .1685 .1660 .1635 .1611
1.0 .1587 .1562 .1539 .1515 .1492 .1469 .1446 .1423 .1401 .1379
1.1 .1357 .1335 .1314 .1292 .1271 .1251 .1230 .1210 .1190 .1170
1.2 .1151 .1131 .1112 .1093 .1075 .1056 .1038 .1020 .1003 .0985
1.3 .0968 .0951 .0934 .0918 .0901 .0885 .0869 .0853 .0838 .0823
1.4 .0808 .0793 .0778 .0764 .0749 .0735 .0721 .0708 .0694 .0681
1.5 .0668 .0655 .0643 .0630 .0618 .0606 .0594 .0582 .0571 .0559
1.6 .0548 .0537 .0526 .0516 .0505 .0495 .0485 .0475 .0465 .0455
1.7 .0446 .0436 .0427 .0418 .0409 .0401 .0392 .0384 .0375 .0367
1.8 .0359 .0351 .0344 .0336 .0329 .0322 .0314 .0307 .0301 .0294
1.9 .0287 .0281 .0274 .0268 .0262 .0256 .0250 .0244 .0239 .0233
2.0 .02275 .02222 .02169 .02118 .02068 .02018 .01970 .01923 .01876 .01831
2.1 .01786 .01743 .01700 .01659 .01618 .01578 .01539 .01500 .01463 .01426
2.2 .01390 .01355 .01321 .01287 .01255 .01222 .01191 .01160 .01130 .01101
2.3 .01072 .01044 .01017 .00990 .00964 .00939 .00914 .00889 .00866 .00842
2.4 .00820 .00798 .00776 .00755 .00734 .00714 .00695 .00676 .00657 .00639
2.5 .00621 .00604 .00587 .00570 .00554 .00539 .00523 .00508 .00494 .00480
2.6 .00466 .00453 .00440 .00427 .00415 .00402 .00391 .00379 .00368 .00357
2.7 .00347 .00336 .00326 .00317 .00307 .00298 .00289 .00280 .00272 .00264
2.8 .00256 .00248 .00240 .00233 .00226 .00219 .00212 .00205 .00199 .00193
2.9 .00187 .00181 .00175 .00169 .00164 .00159 .00154 .00149 .00144 .00139
3.0 .00135
.00 .01 .02 .03 .04 .05 .06 .07 .08 .09
AREAS IN TAIL OF THE STANDARD NORMAL DISTRIBUTION
x μ
(x – μ)
–––––-
σ
x
1
Study Unit 1
Basic Numerical Concepts
Contents Page
Introduction 3
A. Number Systems 4
Denary Number System 4
Roman Number System 4
Binary System 4
B. Numbers – Approximation and Integers 5
Approximation 5
Integers 6
C. Arithmetic 7
Subtraction (symbol ÷) 8
Multiplication (symbol ×) 8
Division (symbol ÷) 9
The Rule of Priority 9
Brackets 10
D. Dealing with Negative Numbers 11
Multiplication and Division 13
E. Fractions 13
Basic Rules for Fractions 14
Adding and Subtracting with Fractions 17
Multiplying and Dividing Fractions 20
F. Decimals 22
Arithmetic with Decimals 23
Limiting the Number of Decimal Places 25
(Continued over)
2 Basic Numerical Concepts
G. Percentages 26
Arithmetic with Percentages 27
Percentages and Fractions 27
Percentages and Decimals 27
Calculating Percentages 28
H. Ratios 29
I. Further Key Concepts 31
Indices 31
Expressing Numbers in Standard Form 32
Factorials 33
Answers to Questions for Practice 35
Basic Numerical Concepts 3
INTRODUCTION
The skill of being able to handle numbers well is very important in many jobs. You may often
be called upon to handle numbers and express yourself clearly in numerical form – and this
is part of effective communication, as well as being essential to understanding and solving
Sometimes we can express numerical information more clearly in pictures – for example, in
charts and diagrams. We shall consider this later in the course. Firstly though, we need to
examine the basic processes of manipulating numbers themselves. In this study unit, we
shall start by reviewing a number of basic numerical concepts and operations. It is likely that
you will be familiar with at least some of these, but it is essential to ensure that you fully
understand the basics before moving on to some of the more advanced applications in later
units.
Throughout the course, as this is a practical subject, there will be plenty of practice questions
for you to work through. To get the most out of the material, we recommend strongly that you
do attempt all these questions.
One initial word about calculators. It is perfectly acceptable to use calculators to perform
virtually all arithmetic operations – indeed, it is accepted practice in examinations. You should
therefore get to know how to use one quickly and accurately. However, you should also
ensure that you know exactly how to perform all the same operations manually.
Understanding how the principles work is essential if you are to use the calculator correctly.
Objectives
When you have completed this study unit you will be able to:
 identify some different number systems
 round-up numbers and correct them to significant figures
 carry out calculations involving the processes of addition, subtraction, multiplication
and division
 manipulate negative numbers
 cancel fractions down to their simplest terms
 change improper fractions to mixed numbers or integers, and vice versa
 add, subtract, multiply and divide fractions and mixed numbers
 carry out calculations using decimals
 calculate ratios and percentages, carry out calculations using percentages and divide a
given quantity according to a ratio
 explain the terms index, power, root, reciprocal and factorial.
4 Basic Numerical Concepts
A. NUMBER SYSTEMS
Denary Number System
We are going to begin by looking at the number system we use every day.
Because ten figures (i.e. symbols 0 to 9) are used, we call this number system the base ten
system or denary system. This is illustrated in the following table.
Figure 1.1: The denary system
Seventh
column
Sixth
column
Fifth
column
Fourth
column
Third
column
Second
column
First
column
Millions
Hundreds
of
thousands
Tens of
thousands
Thousands Hundreds Tens Units
The number of columns indicates the actual amounts involved. Consider the number 289.
This means that, in this number, there are:
two hundreds (289)
eight tens (289)
nine units (289).
Before we look at other number systems, we can note that it is possible to have negative
numbers – numbers which are less than zero. Numbers which are more than zero are
positive. We indicate a negative number by a minus sign (÷). A common example of the use
of negative numbers is in the measurement of temperature – for example, ÷20°C (i.e. 20°C
below zero).
We could indicate a positive number by a plus sign (+), but in practice this is not necessary
and we adopt the convention that, say, 73 means +73.
Roman Number System
We have seen that position is very important in the number system that we normally use.
This is equally true of the Roman numerical system, but this system uses letters instead of
figures. Although we invariably use the normal denary system for calculations, Roman
numerals are still in occasional use in, for example, the dates on film productions, tabulation,
house numbers, etc.
Here are some Roman numbers and their denary equivalent:
XIV 14 XXXVI 36
IX 9 LVlll 58
Binary System
As we have said, our numbering system is a base ten system. By contrast, computers use
base two, or the binary system. This uses just two symbols (figures), 0 and 1.
In this system, counting from 0 to 1 uses up all the symbols, so a new column has to be
started when counting to 2, and similarly the third column at 4.
Basic Numerical Concepts 5
Figure 1.2: The binary system
Denary Binary
0 0
1 1
2 10
3 11
4 100
5 101
6 110
7 111
8 1000
9 1001
This system is necessary because computers can only easily recognise two figures – zero or
one, corresponding to "components" of the computer being either switched off or on. When a
number is transferred to the computer, it is translated into the binary system.
B. NUMBERS – APPROXIMATION AND INTEGERS
Approximation
Every figure used in a particular number has a meaning. However, on some occasions,
precision is not needed and may even hinder communication. For example, if you are
calculating the cost of the journey, you need only express the distance to the nearest mile.
It is important to understand the methods of approximation used in number language.
(a) Rounding off
Suppose you are given the figure 27,836 and asked to quote it to the nearest thousand.
Clearly, the figure lies between 27,000 and 28,000. The halfway point between the two
is 27,500. The number 27,836 is more than this and so, to the nearest thousand, the
There are strict mathematical rules for deciding whether to round up or down to the
nearest figure:
Look at the digit after the one in the last required column. Then:
 for figures of four or less, round down
 for figures of five or more, round up.
Thus, when rounding off 236 to the nearest hundred, the digit after the one in the last
required column is 3 (the last required column being the hundreds column) and the
answer would be to round down to 200. When rounding off 3,455 to the nearest ten,
the last required column is the tens column and the digit after that is 5, so we would
round up to 3,460.
6 Basic Numerical Concepts
(However, you should note that different rules may apply in certain situations – for
example, in VAT calculations, the Customs and Excise rule is always to round down to
the nearest one pence.)
(b) Significant figures
By significant figures we mean all figures other than zeros at the beginning or end of a
number. Thus:
 632,000 has three significant figures as the zeros are not significant
 000,632 also has only three significant figures
 630,002 has six significant figures – the zeros here are significant as they occur
in the middle of the number.
We can round off a number to a specified number of significant figures. This is likely to
be done to aid the process of communication – i.e. to make the number language
easier to understand – and is particularly important when we consider decimals later in
the unit.
Consider the figure 2,017. This has four significant figures. To write it correct to three
significant figures, we need to drop the fourth figure – i.e. the 7. We do this by using the
rounding off rule – round down from 4 or round up from 5 – so 2,017 correct to three
significant figures becomes 2,020. To write 2,017 correct to one significant figure, we
consider the second significant figure, the zero, and apply the rule for rounding off,
giving 2,000. Do not forget to include the insignificant zeros, so that the digits keep
their correct place values.
Integers
Integers are simply whole numbers. Thus, 0, 1, 2, 3, 4, etc, are integers, as are all the
negative whole numbers.
The following are not integers:
¾, 1½, 7¼, 0.45, 1.8, 20.25
By adjusting any number to the nearest whole number, we obtain an integer. Thus, a number
which is not an integer can be approximated to become one by rounding off.
Questions for Practice 1
1. Round off the following the figures to the nearest thousand:
(a) 3,746
(b) 8,231
(c) 6,501
2. Round off the following figures to the nearest million:
(a) 6,570,820
(b) 8,480,290
(c) 5,236,000
Basic Numerical Concepts 7
3. Correct the following to three significant figures:
(a) 2,836
(b) 3,894
(c) 7,886
(d) 9,726
4. How many significant figures are in the following?
(a) 1,736
(b) 22,300
(c) 1,901
(d) 26,000
Now check your answers with those given at the end of the unit.
C. ARITHMETIC
We will now revise some basic arithmetic processes. These may well be familiar to you, and
although in practice it is likely you will perform many arithmetic operations using a calculator,
it is important that you know how to do the calculations manually and understand all the rules
that apply and the terminology used.
When carrying out arithmetic operations, remember that the position of each digit in a
number is important, so layout and neatness matter.
When adding numbers, it is important to add each column correctly. If the numbers are
written underneath each other in the correct position, there is less chance of error.
So, to add 246, 5,322, 16, 3,100 and 43, the first task is to set the figures out in columns:
246
5,322
16
3,100
43
Total 8,727
The figures in the right hand column (the units column) are added first, then those in the next
column to the left and so on. Where the sum of the numbers in a column amount to a figure
greater than nine (for example, in the above calculation, the sum of the units column comes
to 17), only the last number is inserted into the total and the first number is carried forward to
be added into the next column. Here, it is 1 or, more precisely, one ten which is carried
forward into the "tens" column.
Where you have several columns of numbers to add (as may be the case with certain
statistical or financial information), the columns can be cross-checked to produce check
totals which verify the accuracy of the addition:
8 Basic Numerical Concepts
Columns
A B C D Totals
246 210 270 23 749
211 203 109 400 923
29 107 13 16 165
300 99 7 3 409
47 102 199 91 439
Rows
Totals 833 721 598 533 2,685
Note that each row is added up as well as each column, thus providing a full cross-check.
Subtraction (symbol ÷)
Again, it is important to set out the figures underneath each other correctly in columns and to
work through the calculation, column by column, starting with the units column.
So, to subtract 21 from 36, the first task is to set the sum out:
36
÷21
Total 15
As with addition, you must work column by column, starting from the right – the units column.
Now consider this calculation:
34
÷18
Total 16
Where it is not possible to subtract one number from another (as in the units column where
you cannot take 8 from 4), the procedure is to "borrow" one from the next column to the left
and add it to the first number (the 4). The one that you borrow is, in fact, one ten – so adding
that to the first number makes 14, and subtracting 8 from that allows you to put 6 into the
total. Moving on to the next column, it is essential to put back the one borrowed and this is
added to the number to be subtracted. Thus, the second column now becomes 3 ÷ (1 + 1), or
3 ÷ 2, which leaves 1.
Multiplication (symbol ×)
The multiplication of single numbers is relatively easy – for example, 2 × 4 = 8. When larger
numbers are involved, it is again very helpful to set out the calculation in columns:
246
× 23
In practice, it is likely that you would use a calculator for this type of "long multiplication", but
we shall briefly review the principles of doing it manually.
Before doing so though, we shall note the names given to the various elements involved in a
multiplication operation:
Basic Numerical Concepts 9
 the multiplicand is the number to be multiplied – in this example, it is 246
 the multiplier is the number by which the multiplicand is to be multiplied – here it is 23
and
 the answer to a multiplication operation is called the product – so to find the product of
two or more numbers, you multiply them together.
Returning to our example of long multiplication, the first operation is to multiply the
multiplicand by the last number of the multiplier (the number of units, which is 3) and enter
the product as a subtotal – this shown here as 738. Then we multiply the multiplicand by the
next number of the multiplier (the number of tens, which is 2). However, here we have to
remember that we are multiplying by tens and so we need to add a zero into the units column
of this subtotal before entering the product for this operation (492). Thus, the actual product
of 246 × 20 is 4,920. Finally, the subtotals are added to give the product for the whole
operation:
246
× 23
Subtotal (units) 738
Subtotal (tens) 4,920
Total product 5,658
Division (symbol ÷)
In division, the calculation may be set out in a number of ways:
360 ÷ 24, or 360/24, or most commonly
24
360
Just as there were three names for the various elements involved in a multiplication
operation, so there are three names or terms used in division:
 the dividend is the number to be divided – in this example, it is 360
 the divisor is the number by which the dividend is to be divided – here it is 24 and
 the answer to a division operation is called the quotient – here the quotient for our
example is 15.
Division involving a single number divisor is relatively easy – for example, 16 ÷ 2 = 8. For
larger numbers, the task becomes more complicated and you should use a calculator for all
long division.
The Rule of Priority
Calculations are often more complicated than a simple list of numbers to be added and/or
subtracted. They may involve a number of different operations – for example:
8 × 6 ÷ 3 +
2
8
When faced with such a calculation, the rule of priority in arithmetical operations requires
that:
division and multiplication are carried out before addition and subtraction.
10 Basic Numerical Concepts
Thus, in the above example, the procedure would be:
(a) first the multiplication and division: 8 × 6 = 48;
2
8
= 4
(b) then, putting the results of the first step into the calculation, we can do the addition and
subtraction: 48 ÷ 3 + 4 = 49.
Brackets
Brackets are used extensively in arithmetic, and we shall meet them again when we explore
algebra.
Look at the previous calculation again. It is not really clear whether we should:
(a) multiply the eight by six; or
(b) multiply the eight by "six minus three" (i.e. by three); or even
(c) multiply the eight by "six minus three plus four" (i.e. by seven).
We use brackets to separate particular elements in the calculation so that we know exactly
which operations to perform on which numbers. Thus, we could have written the above
calculation as follows:
(8 × 6) ÷ 3 +
2
8
This makes it perfectly clear what numbers are to be multiplied.
Consider another example:
12 ÷ 3 + 3 × 6 + 1
Again there are a number of ways in which this calculation may be interpreted, for example:
(12 ÷ 3) + (3 × 6) + 1
12 ÷ (3 + 3) × (6 + 1)
(12 ÷ 3) + 3 × (6 + 1)
(12 ÷ [3 + 3]) × 6 + 1
Note that we can put brackets within brackets to further separate elements within the
calculation and clarify how it is to be performed.
The rule of priority explained above now needs to be slightly modified. It must be strictly
followed except when brackets are included. In that case, the contents of the brackets are
evaluated first. If there are brackets within brackets, then the innermost brackets are
evaluated first.
Example:
2 + 3 × (6 ÷ 8 ÷ 2)
= 2 + 3 × (6 ÷ 4) (within the brackets, the division first)
= 2 + 3 × 2
= 2 + 6 (multiplication before addition)
= 8
Thus we can see that brackets are important in signifying priority as well as showing how to
treat particular operations.
Basic Numerical Concepts 11
Questions for Practice 2
1. Work out the answers to the following, using a calculator if necessary.
(a) 937 × 61
(b) 1,460 × 159
(c) 59,976 ÷ 63
(d) 19,944 ÷ 36
A B C D Totals
322 418 42 23
219 16 600 46
169 368 191 328
210 191 100 169
Totals
3. Work out the answers to the following questions.
(a) (6 + 3) × 6 + 4
(b) (6 + 3) × (6 + 4)
(c) 10 × (5 + 2) ÷ (10 ÷ 5)
(d) 24 + 24 ÷ (2 × 2)
(e) 6 × (24 ÷ [4 × 2])
(f) (15 ÷ [3 + 2] + 6 + [4 × 3] × 2) ÷ 11
Now check your answers with those given at the end of the unit.
D. DEALING WITH NEGATIVE NUMBERS
We can use a number scale, like the markings on a ruler, as a simple picture of addition and
subtraction. Thus, you can visualise the addition of, say, three and four by finding the position
of three on the scale and counting off a further four divisions to the right, as follows:
0 1 2 3 4 5 6 7 8
This brings us to the expected answer of seven (3 + 4).
Similarly we can view subtraction as moving to the left on the scale. Thus, 6 minus 4 is
represented as follows:
12 Basic Numerical Concepts
0 1 2 3 4 5 6 7 8
This illustrates the sum 6 ÷ 4 = 2.
Negative numbers can also be represented on the number scale. All we do is extend the
scale to the left of zero:
÷6 ÷5 ÷4 ÷3 ÷2 ÷1 0 1 2 3 4 5 6
We can now visualise the addition and subtraction of negative numbers in the same way as
above. Consider the sum (÷3) + 4. Start at ÷3 on the scale and count four divisions to the
right:
÷4 ÷3 ÷2 ÷1 0 1 2 3 4
Therefore, (÷3) + 4 = 1.
(Note that, here, we have shown the figure of "minus 3" in brackets to avoid confusion with
The order of the figures of these calculations does not matter, so we can also write:
(÷3) + 4 = 4 + (÷3) = 1
We can illustrate the second sum as follows:
÷3 ÷2 ÷1 0 1 2 3 4 5
This is the same as 4 ÷ 3. In effect, therefore, a plus and a minus together make a minus.
With subtraction, the process is slightly different. To subtract a minus number, we change the
subtraction sign to a plus and treat the number as positive. For example:
4 ÷ (÷3) = 4 + 3 = 7
To sum up:
 a plus and minus make a minus
 two minuses make a plus.
Consider the following examples. (Draw your own number scales to help visualise the sums
7 ÷ (÷3) = 7 + 3 = 10
2 + (÷6) = 2 ÷ 6 = ÷4
Basic Numerical Concepts 13
Multiplication and Division
The rules for multiplication and division are as follows:
Type of calculation Product/Quotient Example
Minus times plus minus ÷3 × 7 = ÷21
Minus times minus plus ÷7 × (÷3) = 21
Minus divide by plus minus ÷21 ÷ 3 = ÷7
Minus divide by minus plus ÷21 ÷ (÷3) = 7
Plus divide by minus minus 21 ÷ (÷3) = ÷7
Questions for Practice 3
Solve the following sums, using a number scale if necessary:
1. 7 + (÷4) 2. (÷4) + 7 3. 8 ÷ (÷6)
4. 6 + (÷8) 5. (÷9) + 3 6. (÷8) ÷ (÷2)
7. (÷5) + (÷4) 8. (÷3) × (÷5) 9. (÷7) ÷ (÷7)
10. 10 × (÷10)
Now check your answers with those given at the end of the unit.
E. FRACTIONS
A fraction is a part of a whole number.
When two or more whole numbers are multiplied together, the product is always another
whole number. In contrast, the division of one whole number by another does not always
result in a whole number. Consider the following two examples:
12 ÷ 5 = 2, with a remainder of 2
10 ÷ 3 = 3, with a remainder of 1
The remainder is not a whole number, but a part of a whole number – so, in each case, the
result of the division is a whole number and a fraction of a whole number. The fraction is
expressed as the remainder divided by the original divisor and the way in which it is shown,
then, is in the same form as a division term:
12 ÷ 5 = 2, with a remainder of 2 parts of 5, or two divided by five, or
5
2
14 Basic Numerical Concepts
10 ÷ 3 = 3, with a remainder of 1 part of 3, or one divided by three, or
3
1
Each part of the fraction has a specific terminology:
 the top number is the numerator and
 the bottom number is the denominator.
Thus, for the above fractions, the numerators are 2 and 1, and the denominators are 5 and 3.
When two numbers are divided, there are three possible outcomes.
(a) The numerator is larger than the denominator
In this case, the fraction is known as an improper fraction. The result of dividing out an
improper fraction is a whole number plus a part of one whole, as in the two examples
above. Taking two more examples:
9
42
= 4 whole parts and a remainder of 6 parts of 9, or 4
9
6
24
30
= 1 whole part and a remainder of 6 parts of 24, or 1
24
6
(b) The denominator is larger than the numerator
In this case, the fraction is known as a proper fraction. The result of dividing out a
proper fraction is only a part of one whole and therefore a proper fraction has a value of
less than one.
Examples of proper fractions are:
5
1
,
3
2
,
7
4
,
5
4
, etc.
(c) The numerator is the same as the denominator
In this case, the expression is not a true fraction. The result of dividing out is one whole
part and no remainder. The answer must be 1. For example:
5
5
= 5 ÷ 5 = 1
Fractions are often thought of as being quite difficult. However, they are not really hard as
long as you learn the basic rules about how they can be manipulated. We shall start by
looking at these rules before going on to examine their application in performing arithmetic
operations on fractions.
Basic Rules for Fractions
 Cancelling down
Both the numerator and the denominator in a fraction can be any number. The
following are all valid fractions:
5
2
,
50
20
,
93
47
,
9
6
The rule in expressing fractions, though, is always to reduce the fraction to the smallest
possible whole numbers for both its numerator and denominator. The resulting fraction
is then in its lowest possible terms.
Basic Numerical Concepts 15
The process of reducing a fraction to its lowest possible terms is called cancelling
down. To do this, we follow another simple rule. Divide both the numerator and
denominator by the largest whole number which will divide into them exactly.
Consider the following fraction from those listed previously:
50
20
Which whole number will divide into both 20 and 50? There are several: 2, 5, and 10.
However the rule states that we use the largest whole number, so we should use 10.
Dividing both parts of the fraction by 10 reduces it to:
5
2
Note that we have not changed the value of the fraction, just the way in which the parts
of the whole are expressed. Thus:
50
20
=
5
2
A fraction which has not been reduced to its lowest possible terms is called a vulgar
fraction.
Cancelling down can be a tricky process since it is not always clear what the largest
whole number to use might be. For example, consider the following fraction:
231
165
In fact, both terms can be divided by 33, but you are very unlikely to see that straight
away. However, you would probably quickly see that both are divisible by 3, so you
could start the process of cancelling down by dividing by three:
3 231
3 165
÷
÷
=
77
55
This gives us a new form of the same fraction, and we can now see that both terms can
be further divided – this time by 11:
11 77
11 55
÷
÷
=
7
5
So, large vulgar fractions can often be reduced in stages to their lowest possible terms.
 Changing the denominator to required number
The previous process works in reverse. If we wanted to express a given fraction in a
different way, we can do that by multiplying both the numerator and denominator by the
same number. So, for example,
5
2
could be expressed as
20
8
by multiplying both terms by 4.
If we wanted to change the denominator to a specific number – for example, to express
½ in eighths – the rule is to divide the required denominator by the existing
denominator (8 ÷ 2 = 4) and then multiply both terms of the original fraction by this
result:
4 2
4 1
×
×
=
8
4
16 Basic Numerical Concepts
Note again that we have not changed the value of the fraction, just the way in which the
parts of the whole are expressed.
 Changing an improper fraction into a whole or mixed number
Suppose we are given the following improper fraction:
8
35
The rule for converting this is to divide the numerator by the denominator and place
any remainder over the denominator. Thus we find out how many whole parts there are
and how many parts of the whole (in this case, eighths) are left over:
8
35
= 35 ÷ 8 = 4 and a remainder of three = 4
8
3
This result is known as a mixed number – one comprising a whole number and a
fraction.
 Changing a mixed number into an improper fraction
This is the exact reverse of the above operation. The rule is to multiply the whole
number by the denominator in the fraction, then add the numerator of the fraction to
this product and place the sum over the denominator.
For example, to write 3
3
2
as an improper fraction:
(a) first multiply the whole number (3) by the denominator (3) = 3 × 3 = 9, changing
the three whole ones into nine thirds
(b) then add the numerator (2) to this = 2 + 9 = 11, thus adding the two thirds to the
nine thirds from the first operation
(c) finally place the sum over the denominator =
3
11
Questions for Practice 4
1. Reduce the following fractions to their lowest possible terms:
(a)
36
9
(b)
64
8
(c)
70
21
(d)
54
18
(e)
49
14
(f)
150
25
(g)
81
36
(h)
135
45
(i)
192
24
(j)
88
33
Basic Numerical Concepts 17
2. Change the following fractions as indicated:
(a)
5
1
to 10ths (b)
5
3
to 20ths
(c)
9
4
to 81sts (d)
11
9
to 88ths
3. Change the following improper fractions to mixed numbers:
(a)
5
12
(b)
8
17
(c)
6
25
(d)
3
31
4. Change the following mixed numbers to improper fractions:
(a) 2
3
2
(b) 3
8
3
(c) 4
8
1
(d) 9
4
3
Now check your answers with those given at the end of the unit.
(a) Proper fractions
Adding or subtracting fractions depends upon them having the same denominator. It is
not possible to do these operations if the denominators are different.
Where the denominators are the same, adding or subtracting is just a simple case of
applying the operation to numerators. For example:
7
3
+
7
2
=
7
5
8
5
÷
8
1
=
8
4
=
2
1
Where the denominators are not the same, we need to change the form of expression
so that they become the same. This process is called finding the common denominator.
Consider the following calculation:
5
2
+
10
3
We cannot add these two fractions together as they stand – we have to find the
common denominator. In this case, we can change the expression of the first fraction
by multiplying both the numerator and denominator by 2, thus making its denominator
10 – the same as the second fraction:
10
4
+
10
3
=
10
7
18 Basic Numerical Concepts
Not all sums are as simple as this. It is often the case that we have to change the form
of expression of all the fractions in a sum in order to give them all the same
denominator.
Consider the following calculation:
5
3
+
4
1
÷
10
1
The lowest common denominator for these three fractions is 20. We must then convert
all three to the new denominator by applying the rule explained above for changing a
fraction to a required denominator:
5
3
+
4
1
÷
10
1
=
20
12
+
20
5
÷
20
2
=
20
15
=
4
3
The rule is simple:
 find the lowest common denominator – and where this is not readily apparent, a
common denominator may be found by multiplying the denominators together
 change all the fractions in the calculation to this common denominator
 then add (or subtract) the numerators and place the result over the common
denominator.
(b) Mixed numbers
When adding mixed numbers, we deal with the integers and the fractions separately.
The procedure is as follows:
 first add all the integers together
 then add the fractions together as explained above, and then
 add together the sum of the integers and the fractions.
Consider the following calculation:
3 + 4
4
1
+ 5
2
1
+ 2
4
3
= (3 + 4 + 5 + 2) + |
.
|

\
|
+ +
4
3
2
1
4
1
= 14 + |
.
|

\
|
+ +
4
3
4
2
4
1
= 14 +
4
6
= 14 + 1
2
1
= 15
2
1
In general, the same principles are applied to subtracting mixed numbers. For example:
4
6
5
÷ 1
2
1
= (4 ÷ 1) + |
.
|

\
|
÷
2
1
6
5
= 3 + |
.
|

\
|
÷
6
3
6
5
= 3 +
6
2
= 3
3
1
Basic Numerical Concepts 19
However, the procedure is slightly different where the second fraction is larger than the
first. Consider this calculation:
4
2
1
÷ 1
6
5
Here, it is not possible to subtract the fractions even after changing them to the
common denominator:
6
3
÷
6
5
To complete the calculation we need to take one whole number from the integer in the
first mixed number, convert it to a fraction with the same common denominator and add
it to the fraction in the mixed number. Thus:
4
2
1
÷ 3 + |
.
|

\
|
+
2
1
1 = 3 + |
.
|

\
|
+
6
3
6
6
= 3
6
9
We can now carry out the subtraction as above:
4
2
1
÷ 1
6
5
= 3
6
9
÷ 1
6
5
= (3 ÷ 1) + |
.
|

\
|
÷
6
5
6
9
= 2 +
6
4
= 2
3
2
Questions for Practice 5
Carry out the following calculations:
1.
3
1
+
4
1
2.
6
1
+
10
1
3.
8
1
+
3
2
4.
5
1
+
6
1
5.
5
2
+
4
3
6.
3
2
÷
4
1
7.
6
5
÷
3
1
8.
8
7
÷
5
1
9.
7
5
÷
3
1
10.
4
3
÷
8
3
11. 2
2
1
+ 1
4
3
12. 4
8
3
+ 2
4
3
+ 1
2
1
13.
3
2
+ 3 + 1
2
1
+ 2
5
1
14. 1
4
3
÷
3
2
15. 4
9
5
÷ 2
3
1
16. 6
8
1
÷ 2
3
1
17. 2
8
1
+ 2
3
1
÷ 2
6
1
18. 4
4
3
+ 1
3
2
÷ 2
2
1
Now check your answers with those given at the end of the unit.
20 Basic Numerical Concepts
Multiplying and Dividing Fractions
(a) Multiplying a fraction by a whole number
In this case, we simply multiply the numerator by the whole number and place the
product (which is now the new numerator) over the old denominator. For example:
3
2
× 5 =
3
5 2×
=
3
10
= 3
3
1
4
3
× 7 =
4
7 3 ×
=
4
21
= 5
4
1
We often see the word "of" in a calculation involving fractions – for example, ¼ of
£12,000. This is just another way of expressing multiplication.
(b) Dividing a fraction by a whole number
In this case, we simply multiply the denominator by the whole number and place the old
numerator over the product (which is now the new denominator). For example:
4
3
÷ 5 =
5 4
3
×
=
20
3
3
2
÷ 5 =
5 3
2
×
=
15
2
(c) Multiplying one fraction by another
In this case, we multiply the numerators by each other and the denominators by each
other, and then reduce the result to the lowest possible terms. For example:
4
3
×
3
2
=
3 4
2 3
×
×
=
12
6
=
2
1
8
3
×
5
3
=
5 8
3 3
×
×
=
40
9
(d) Dividing by a fraction
In this case, we turn the divisor upside down and multiply. For example:
4 ÷
4
3
= 4 ×
3
4
=
3
16
= 5
3
1
6
1
÷
3
1
=
6
1
×
1
3
=
6
3
=
2
1
(e) Dealing with mixed numbers
All that we have said so far in this section applies to the multiplication and division of
proper and improper fractions. Before we can apply the same methods to mixed
numbers, we must complete one further step – convert the mixed number to an
improper fraction. For example:
5
2
1
×
11
4
=
2
11
×
11
4
=
22
44
= 2
4
4
3
÷
2
1
=
4
19
×
1
2
=
4
38
= 9
2
1
Basic Numerical Concepts 21
(f) Cancelling during multiplication
We have seen how it is possible to reduce fractions to their lowest possible terms by
cancelling down. This involves dividing both the numerator and the denominator by the
same number, for example:
8
6
cancels down to
4
3
by dividing both terms by 2.
We often show the process of cancelling down during a calculation by crossing out the
cancelled terms and inserting the reduced terms as follows:
3 12
2 8
=
3
2
(It is always helpful to set out all the workings throughout a calculation. Some
calculations can be very long, involving many steps. If you make a mistake somewhere,
it is much easier to identify the error if all the workings are shown. It is also the case
that, in an examination, examiners will give "marks for workings", even if you make a
mistake in the arithmetic. However, if you do not show all the workings, they will not be
able to follow the process and see that the error is purely mathematical, rather than in
the steps you have gone through.)
Cancellation can also be used in the multiplication of fractions. In this case it can be
applied across the multiplication sign – dividing the numerator of one fraction and the
denominator of the other by the same number. For example:
4
1 3
×
4 12
1
=
16
1
Remember that every cancelling step must involve a numerator and a denominator.
Thus, the following operation would be wrong:
1 4
3
×
3 12
1
=
3
3
Note, too, that you cannot cancel across addition and subtraction signs.
However, you can use cancellation during a division calculation at the step when,
having turned the divisor upside down, you are multiplying the two fractions. So,
repeating the calculation involving mixed numbers from above, we could show it as
follows:
4
4
3
÷
2
1
=
2 4
19
×
1
1 2
=
2
19
= 9
2
1
22 Basic Numerical Concepts
Questions for Practice 6
Carry out the following calculations:
1. 7 ×
4
3
2. 8 ×
3
2
3. 5 ×
6
3
4. 6 ×
7
3
5.
3
2
÷ 6 6.
4
3
÷ 6
7.
8
7
÷ 3 8.
7
5
÷ 4
Now check your answers with those given at the end of the unit.
F. DECIMALS
Decimals are an alternative way of expressing a particular part of a whole. The term
"decimal" means "in relation to ten", and decimals are effectively fractions expressed in
tenths, hundredths, thousandths, etc.
Unlike fractions though, decimals are written completely differently and this makes carrying
out arithmetic operations on them much easier.
Decimals do not have a numerator or a denominator. Rather, the part of the whole is shown
by a number following a decimal point:
 the first number following the decimal point represents the number of tenths, so 0.1 is
one-tenth part of a whole and 0.3 is three-tenths of a whole
 the next number to the right, if there is one, represents the number of hundredths, so
0.02 is two-hundredths of a whole and 0.67 is six-tenths and seven-hundredths (i.e. 67
hundredths) of a whole
 the next number to the right, again if there is one, represents the number of
thousandths, so 0.005 is five-thousandths of a whole and 0.134 is 134 thousandths of a
whole
and so on up to as many figures as are necessary.
We can see the relationship between decimals and fractions as follows:
0.1 =
10
1
, 0.3 =
10
3
, 1.6 = 1
10
6
0.01 =
100
1
, 0.67 =
100
67
, 3.81 = 3
100
81
0.001 =
1000
1
, 0.054 =
1000
54
, 5.382 = 5
1000
382
We can see that any fraction with a denominator of ten, one hundred, one thousand, ten
thousand, etc. can be easily converted into a decimal. However, fractions with a different
denominator are also relatively easy to convert by simple division.
Basic Numerical Concepts 23
Arithmetic with Decimals
Performing addition, subtraction, multiplication and division with decimals is exactly the same
as carrying out those operations on whole numbers. The one key point to bear in mind is to
get the position of the decimal point in the right place. Layout is all important in achieving
this.
To add or subtract decimals, arrange the numbers in columns with the decimal points
all being in the same vertical line. Then you can carry out the arithmetic in the same
way as for whole numbers.
For example, add the following: 3.6 + 7.84 + 9.172
The layout for this is as follows:
3.6
7.84
9.172
Total 20.612
Similarly, subtraction is easy provided that you follow the same rule of ensuring that the
decimal points are aligned vertically. For example, subtract 18.857 from 63.7:
63.7
– 18.857
Total 44.843
(b) Multiplying decimals
It is easier to use a calculator to multiply decimals, although you should ensure that you
take care to get the decimal point in the correct place when both inputting the figures
The principles of multiplying decimals manually are the same as those considered
earlier in the study unit for whole numbers. Again, it is important to get the decimal
point in the right place.
There is a simple rule for this – the answer should have the same number of decimal
places as there were in the question.
Consider the calculation of 40 × 0.5. The simplest way of carrying this out is to ignore
the decimal points in the first instance, multiply the numbers out as if the figures were
whole numbers, and then put back the decimal point in the answer. The trick is to get
the position of the decimal point right!
So, ignoring the decimal points, we can multiply the numbers as follows:
40 × 5 = 200
Now we need to put the decimal point into the answer. To get it in the right place, we
count the number of figures after a decimal point in the original sum – there is just one,
in the multiplier (0.5) – and place the decimal point that number of places from the right
20.0 or just 20
If the sum had been 0.4 × 0.5, what would the answer be?
24 Basic Numerical Concepts
The multiplication would be the same, but this time there are 2 figures after decimal
points in the original sum. So, we insert the decimal point two places to the right in the
0.2
One final point to note about multiplying decimals is that, to multiply a decimal by ten,
you simply move the decimal point one place to the right. For example:
3.62 × 10 = 36.2
Likewise, to multiply by 100, you move the decimal point two places to the right:
3.62 × 100 = 362
This can be a very useful technique, enabling quick manual calculations to be done
with ease.
(c) Division with decimals
Again, it is far easier to use a calculator when dividing with decimals.
However, when dividing by ten or by one hundred (or even by a thousand or a million),
we can use the technique noted above in respect of multiplication to find the answer
quickly without using a calculator:
 to divide by ten, simply move the decimal point one place to the left – for
example, 68.32 ÷ 10 = 6.832
 to divide by one hundred, simply move the decimal point two places to the left –
for example, 68.32 ÷ 100 = 0.6832.
(d) Converting fractions to decimals, and vice versa
Fractions other than tenths, hundredths, thousandths, etc. can also be converted into
decimals. It is simply a process of dividing the numerator by the denominator and
expressing the answer in decimal form. You can do this manually, but in most
instances it is easier to use a calculator.
For example, to change
5
3
into decimal form, you divide 3 by 5 (i.e. 3 ÷ 5).
Using a calculator the answer is easily found: 0.6
However, not all fractions will convert so easily into decimals in this way.
Consider the fraction
3
1
.
Dividing this out (or using a calculator) gives the answer 0.3333333 or however many
3s for which there is room on the paper or the calculator screen. This type of decimal,
where one or more numbers repeat infinitely, is known as a recurring decimal.
In most business applications, we do not need such a degree of precision, so we
abbreviate the decimal to a certain size – as we shall discuss in the next section.
To change a decimal into a fraction, we simply place the decimal over ten or one
hundred etc, depending upon the size of the decimal. Thus, the decimal becomes the
numerator and the denominator is 10 or 100 etc. For example:
0.36 =
100
36
=
25
9
3.736 = 3
1000
736
= 3
125
92
Basic Numerical Concepts 25
Limiting the Number of Decimal Places
We noted that certain fractions do not convert into exact decimals. The most common
examples are one-third and two-thirds, which convert into 0.3333333... and 0.6666666...
respectively. (When referring to these types of recurring decimals, we would normally just
quote the decimal as, say, "0.3 recurring".) However, there are also plenty of other fractions
which do convert into exact decimals, but only at the level of thousandths or greater. For
example:
7
2
= 0.2857142, or
17
11
= 0.6470568
This gives us a problem: how much detail do we want to know? It is very rare that, in
business applications, we would want to know an answer precise to the millionth part of a
whole, although in engineering, that level of precision may be critical. Such large decimals
also present problems in carrying out arithmetic procedures – for example, it is not easy to
add the above two decimals together and multiplying them does not bear thinking about.
In most business circumstances then, we limit the number of decimal places to a
manageable number. This is usually set at two or three decimal places, so we are only
interested in the level of detail down to hundredths or thousandths. We then speak of a
number or an answer as being "correct to two (or three) decimal places".
For decimals with a larger number of decimal places than the required number, we use the
principles of rounding discussed earlier in the study unit to reduce them to the correct size.
Thus, we look at the digit in the decimal after the one in the last required place and then:
 for figures of five or more, round up the figure in the last required place
 for figures of four or less, round down by leaving the figure in the last required place
unchanged.
So, if we want to express the decimal form of two-sevenths (0.2857142) correct to three
decimal places, we look at the digit in the fourth decimal place: 7. Following the rule for
rounding, we would then round up the number in the third decimal place to give 0.286.
Now consider the decimal forms of one-third and two-thirds. If we were to express these
correct to two decimal places, these would be 0.33 and 0.67 respectively.
Working to only two or three decimal places greatly simplifies decimals, but you should note
that the level of accuracy suffers. For example, multiplying 0.33 by two gives us 0.66, but we
have shown two-thirds as being 0.67. These types of small error can increase when adding,
subtracting, multiplying or dividing several decimals which have been rounded – something
which we shall see later in the course in respect of a set of figures which should add up to
100, but do not. In such cases, we need to draw attention to the problem by stating that
"errors are due to rounding".
Questions for Practice 7
1. Carry out the following calculations:
(a) 3.67 + 4.81 + 3.2 (b) 8.36 + 4.79 + 6.32
(c) 7.07 + 8.08 + 11.263 (d) 11.14 + 8.16 + 3.792
(e) 8.84 ÷ 3.42 (f) 7.36 ÷ 2.182
(g) 10.8 ÷ 4.209 (h) 12.6 ÷ 3.43
26 Basic Numerical Concepts
(i) 3.62 × 2.8 (j) 4.8 × 3.3
(k) 5.8 × 4.1 (l) 0.63 × 5.2
(m) 154 ÷ 2.2 (n) 2.72 ÷ 1.7
(o) 11.592 ÷ 2.07 (p) 2.6 ÷ 0.052
2. Change the following fractions into decimal form, correct to the number of decimal
places (dp) specified:
(a)
5
2
to 2 dp (b)
8
3
to 3 dp
(c)
4
3
to 2 dp (d)
8
7
to 3 dp
(e)
8
5
to 3 dp (f)
6
5
to 3 dp
(g)
7
4
to 3 dp (h)
16
5
to 3 dp
3. Change the following decimals into fractions and reduce them to their lowest possible
terms:
(a) 0.125 (b) 0.60
(c) 1.75 (d) 0.750
Now check your answers with those given at the end of the unit.
G. PERCENTAGES
A percentage is a means of expressing a fraction in parts of a hundred – the words "per cent"
simply mean "per hundred", so when we say "percentage", we mean "out of a hundred". The
expression "20 per cent" means 20 out of 100. The per cent sign is %, so 20 per cent is
written as 20%. Some examples:
 4% means four-hundredths of a whole, or four parts of a hundred =
100
4
 28% means twenty-eight-hundredths of a whole, or twenty-eight parts of a hundred =
100
28
 100% means one hundred-hundredths of a whole, or a hundred parts of a hundred =
100
100
= 1
Percentages are used extensively in many aspects of business since they are generally
more convenient and straightforward to use than fractions and decimals. It is therefore
important that you are fully conversant with working with them.
Basic Numerical Concepts 27
Arithmetic with Percentages
Multiplication and division have no meaning when applied to percentages, so we need only
be concerned with their addition and subtraction. These processes are quite straightforward,
but you need to remember that we can only add or subtract percentages if they are parts of
the same whole.
Take an example: if Peter spends 40% of his income on rent and 25% on household
expenses, what percentage of his income remains? Here both percentages are parts of the
same whole – Peter's income – so we can calculate the percentage remaining as follows:
100% ÷ (40% + 25%) = 35%
However, consider this example: if Alan eats 20% of his cake and 10% of his pie, what
percentage remains? Here the two percentages are not parts of the same whole – one refers
to cake and the other to pie. Therefore, it is not possible to add the two together.
Percentages and Fractions
Since a percentage is a part of one hundred, to change a percentage to a fraction you simply
divide it by 100. In effect all this means is that you place the percentage over 100 and then
reduce it to its lowest possible terms.
For example:
60% =
100
60
=
5
2
24% =
100
24
=
25
6
Changing a fraction into a percentage is the reverse of this process – simply multiply by 100.
For example:
8
3
× 100 =
8
300
% = 37½% or 37.5%
3
2
× 100 =
3
200
% = 66.67%
As you become familiar with percentages you will get to know certain equivalent fractions
well. Here are some of them:
12½% =
8
1
20% =
5
1
25% =
4
1
33.3% =
3
1
50% =
2
1
75% =
4
3
Percentages and Decimals
To change a percentage to a decimal, again you divide it by 100. As we saw in the last
section, this is easy – you simply move the decimal point two places to the left. Thus:
11.5% = 0.115
Working the conversion the other way is equally easy. To change a decimal to a percentage,
move the decimal point two places to the right:
0.36% = 36%
28 Basic Numerical Concepts
Calculating Percentages
To find a particular percentage of a given number, change the percentage rate into a fraction
or a decimal and then multiply the given number by that fraction/decimal.
For example, to find 15% of £260:
£260 ×
100
15
=
100
3900 £
= £39
or £260 × 0.15 = £39
To express one number as a percentage of another, we convert the numbers into a fraction
with the first number as the numerator and the second the denominator, and then change the
fraction into a percentage by multiplying by 100.
For example, what percentage of cars have been sold if 27 have gone from a total stock of
300?
300
27
× 100 =
300
2700
= 9%
Do not forget that percentages refer to parts of the same whole and therefore can only be
calculated where the units of measurement of the quantities concerned are the same.
For example, to show 23 pence as a percentage of £2, we need to express both quantities in
the same units before working out the percentage. This could be both as pence or both as
pounds:
200
23
× 100 = 11.5%
or
2
23 . 0
× 100 = 11.5%
There are occasions when we deal with percentages greater than 100%.
For example, if a woman sells a car for £1,200 and says that she made 20% profit, how
much did she buy it for?
The cost plus the profit = 120% of cost = £1,200.
The cost, therefore, is 100%
To work out the cost, we need to multiply the selling price by
120
100
:
Cost = 1,200 ×
120
100
= £1,000
To prove that this is correct, we work the calculation through the other way. The cost was
£1,000 and she sold it at 20% profit, so the selling price is:
£1,000 + 20% of £1,000 = £(1,000 + 0.2 × 1,000) = £1,200, as given above.
Basic Numerical Concepts 29
Questions for Practice 8
1. Change the following fractions into percentages:
(a)
100
3
(b)
8
3
2. Change the following percentages into fractions and reduce them to their lowest
possible terms
(a) 2% (b) 8%
(c) 7½% (d) 18%
3. Calculate the following:
(a) 15% of £200 (b) 25% of 360°
(c) 12½% of 280 bottles (d) 27% of £1,790
(e) 17.5% of £138
4. Calculate the following:
(a) 60 as a percentage of 300 (b) 25 as a percentage of 75
(c) £25 as a percentage of £1,000 (d) 30 as a percentage of 20
Now check your answers with those given at the end of the unit.
H. RATIOS
A ratio is a way of expressing the relationship between two quantities. It is essential that the
two quantities are expressed in the same units of measurement – pence, number of people,
etc. – or the comparison will not be valid.
For example, if we wanted to give the ratio of the width of a table to its depth where the width
is 2 m and the depth is 87 cm, we would have to convert the two quantities to the same units
before giving the ratio. Expressing both in centimetres would give the ratio of 200 to 87.
Note that the ratio itself is not in any particular unit – it just shows the relationship between
quantities of the same unit.
We use a special symbol (the colon symbol : ) in expressing ratios, so the correct form of
showing the above ratio would be:
200: 87 (pronounced "200 to 87")
There are two general rules in respect of ratios:
 They should always be reduced to their lowest possible terms (as with vulgar fractions).
To do this, express the ratio as a fraction by putting the first figure over the second and
cancel the resulting fraction. Then re-express the ratio in the form of numerator :
denominator.
30 Basic Numerical Concepts
For example, consider the ratio of girls to boys in a group of 85 girls and 17 boys. The
ratio would be 85 : 17. This can be reduced as follows:
17
85
=
5
1
The ratio of 85 to 17 is, therefore, 5 : 1.
 Ratios should always be expressed in whole numbers. There should not be any
fractions or decimals in them.
Consider the ratio of average miles per gallon for cars with petrol engines to that for
cars with diesel engines, where the respective figures are 33½ mpg and 47 mpg. We
would not express this as 33½ : 47, but convert the figures to whole numbers by, here,
multiplying by 2 to give: 67 : 94
Ratios are used extensively in business to provide information about the way in which one
element relates to another. For example, a common way of analysing business performance
is to compare profits with sales. If we look at the ratio of profits to sales across two years, we
may be able to see if this aspect of business performance is improving, staying the same or
deteriorating. (This will be examined elsewhere in your studies.)
Another application is to work out quantities according to a particular ratio. For example,
ratios are often used to express the relationship between partners in a partnership – sharing
profits on a basis of, say, 50 : 50 or 80 : 20.
Consider the case of three partners – Ansell, Boddington and Devenish – who share the
profits of their partnership in the ratio of 3 : 1 : 5. If the profit for a year is £18,000, how much
To divide a quantity according to a given ratio:
 first add the terms of the ratio to find the total number of parts
 then find what fraction each term of the ratio is to the whole, and
 finally divide the total quantity into parts according to the fractions.
For the Ansell, Boddington and Devenish partnership, this would be calculated as follows:
The profit is divided into 3 + 1 + 5 = 9 parts.
Therefore:
9
3
of the profits
9
1
of the profits and
9
5
of the profits.
Therefore:
Ansell gets
9
3
× £18,000 = £6,000
Boddington gets
9
1
× £18,000 = £2,000
Devenish gets
9
5
× £18,000 = £10,000
Basic Numerical Concepts 31
Questions for Practice 9
1. Express the following sales and profits figures as profit to sales ratios for each firm:
Firm Sales Profits
£ £
A 100,000 25,000
B 63,000 7,000
C 30,000 10,000
D 75,000 15,000
2. Calculate the distribution of profits for the following partnerships:
(a) X, Y and Z share profits in the ratio of 3 : 4: 5. Total profit is £60,000.
(b) F, G and H share profits in the ratio of 2 : 3 : 3. Total profit is £7,200.
Now check your answers with those given at the end of the unit.
I. FURTHER KEY CONCEPTS
In this last section we shall briefly introduce a number of concepts which will be used
extensively later in the course. Here we concentrate mostly on what these concepts mean,
and you will get the opportunity to practice their use in later study units. However, we do
include practice in relation to the expression of numbers in standard form.
Indices
Indices are found when we multiply a number by itself one or more times. The number of
times that the multiplication is repeated is indicated by a superscript number to the right of
the number being multiplied. Thus:
2 × 2 is written as 2
2
(the little "2" raised up is the superscript number)
2 × 2 × 2 is written as 2
3
2 × 2 × 2 × 2 × 2 × 2 × 2 is written as 2
7
etc.
The number being multiplied (here, 2) is termed the base and the superscript number is
called the exponent or index of its power. Thus in the case of 2
6
we refer to the term as "two
raised to the sixth power" or "two to the power of six". (Strictly, 2
2
should be referred to as
"two to the power of two", but we normally call it "2 squared"; similarly, 2
3
is referred to as "2
cubed".)
The power of a number is the result of multiplying that number a specified number of times.
Thus, 8 is the third power of 2 (2
3
) and 81 is the fourth power of 3 (3
4
).
32 Basic Numerical Concepts
It is quite possible to have a negative exponent, as in the case of 2
÷3
. A negative exponent
indicates a reciprocal. The reciprocal of any number is one divided by that number. For
example:
the reciprocal 5 is
5
1
and
2
÷3
represents
3
2
1
÷
=
8
1
= 0.125
Indices also do not have to be whole numbers – they can be fractional, as in the case of 16
¼
.
This indicates that the base is raised to the power of a quarter. It is more usual though to
refer to this as the fourth root of the base and to write it as follows:
4
16
The symbol \ indicates a root. A root is the number that produces a given number when
raised to a specified power. Thus, the fourth root of 16 is 2 (2 × 2 × 2 × 2 = 16).
A square root of any number is the number which, when multiplied by itself, is equal to the
first number. Thus
2
9 = 3 (3 × 3 = 9)
The root index (the little 4 or little 2 in the above examples) can be any number. It is, though,
customary to omit the 2 from the root sign when referring to the square root. Thus
36 refers to the square root of 36, which is 6.
Expressing Numbers in Standard Form
The exploration of indices leads us on to this very useful way of expressing numbers.
Working with very large (or very small) numbers can be difficult, and it is helpful if we can find
a different way of expressing them which will make them easier to deal with. Standard form
provides such a way.
Consider the number 93,825,000,000,000. It is a very big number and you would probably
have great difficulty stating it, let alone multiplying it by 843,605,000,000,000.
In fact, can you even tell which is the bigger? Probably not. It is made a little easier by the
two numbers being very close to each other on the page, but if one was on a different page,
it would be very difficult indeed.
However, help is at hand.
The method of expressing numbers in standard form reduces the number to a value between
1 and 10, followed by the number of 10s you need to multiply it by in order to make it up to
the correct size. The above two numbers, under this system are:
9.3825 × 10
13
and 8.43605 × 10
14
.
The 13 and 14 after the 10 are index numbers, as discussed above. Written out in full, the
first number would be:
9.3825 × (10 × 10 × 10 × 10 × 10 × 10 × 10 × 10 × 10 × 10 × 10 × 10 × 10).
Basic Numerical Concepts 33
This is not particularly useful, so an easier way of dealing with a number expressed in
standard form is to remember that the index number after the 10 tells you how many places
the decimal point has to be moved to the right in order to create the full number
(remembering also to put in the 0s). So, the second number from above, written out in full
would be:
8 . 4 3 6 0 5 0 0 0 0 0 0 0 0 0 = 843,605,000,000,000
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Converting numbers in this way makes arithmetic operations much easier – something which
we shall return to later in your studies – and also allows us to compare numbers at a glance.
Thus, we can see immediately that 8.43605 × 10
14
is larger than 9.3825 × 10
13
.
Just as with indices we can also have numbers expressed in standard form using a negative
index – for example, 2.43 × 10
÷2
.
Before we consider this in more detail, it would be helpful to recap exactly what the index
number means.
We know that 10 × 10 = 100, so we can say that 10
2
= 100.
We can go on multiplying 10 by itself to get:
10
3
= 1,000
10
4
= 10,000, etc.
We also know from the above that, when there is a minus number in the index, it indicates a
reciprocal – i.e. one divided by the number as many times as the index specifies. Thus:
10
÷1
=
10
1
= 0.1.
We could go on to say that 10
÷2
= 0.01 and 10
÷3
= 0.001, etc. Again, you will note that a
minus index number after the 10 tells you how many places the decimal point has to be
moved to the left in order to create the full number (remembering also to put in the 0s as
well).
Now you should be able to work out that 2.43 × 10
÷2
= 0.0243.
Factorials
A factorial is the product of all the whole numbers from a given number down to one.
For example, 4 factorial is written as 4! (and is sometimes read as "four bang") and is equal
to 4 × 3 × 2 × 1.
You may encounter factorials in certain types of statistical operations.
34 Basic Numerical Concepts
Questions for Practice 10
1. Express the following numbers in standard form:
(a) 234
(b) 0.045
(c) 6,420
(d) 0.0005032
(e) 845,300,000
(f) 0.00000000000000003792
2. Write out in full the following numbers which are expressed in standard form.
(a) 4.68 × 10
2
(b) 9.35 × 10
÷2
(c) 1.853 × 10
6
(d) 6.02 × 10
÷5
(e) 7.653 × 10
12
(f) 8.16 × 10
÷10
Now check your answers with those given at the end of the unit.
Basic Numerical Concepts 35
Questions for Practice 1
1. (a) 4,000 (b) 8,000
(c) 7,000
2. (a) 7,000,000 (b) 8,000,000
(c) 5,000,000
3. (a) 2,840 (b) 3,880
(c) 7,890 (d) 9,730
4. (a) Four (b) Three
(c) Four (d) Two
Questions for Practice 2
1. (a) 57,157 (b) 232,140
(c) 952 (d) 554
2. The check on your answers is achieved by adding a column into which you put the sum
of each row. The total of the check column should equal the sum of the four column
totals.
A B C D Check
322 418 42 23 805
219 16 600 46 881
169 368 191 328 1,056
210 191 100 169 670
Totals 920 993 933 566 3,412
3. (a) (6 + 3) × 6 + 4 = 9 × 6 + 4 = 54 + 4 = 58
(b) (6 + 3) × (6 + 4) = 9 × 10 = 90
(c) 10 × (5 + 2) ÷ (10 ÷ 5) = 10 × 7 ÷ 2 = 70 ÷ 2 = 68
(d) 24 + 24 ÷ (2 × 2) = 24 + 24 ÷ 4 = 24 + 6 = 30
(e) 6 × (24 ÷ [4 × 2]) = 6 × (24 ÷ 8) = 6 × 3 = 18
(f) (15 ÷ [3 + 2] + 6 + [4 × 3] × 2) ÷ 11
= (15 ÷ 5 + 6 + 12 × 2) ÷ 11 = (3 + 6 + 24) ÷ 11 = 33 ÷ 11 = 3
Questions for Practice 3
1. 3 2. 3
3. 14 4. ÷2
5. ÷6 6. ÷6
7. ÷9 8. 15
9. 1 10. ÷100
36 Basic Numerical Concepts
Questions for Practice 4
1. (a)
4
1
(b)
8
1
(c)
10
3
(d)
3
1
(e)
7
2
(f)
6
1
(g)
9
4
(h)
3
1
(i)
8
1
(j)
8
3
2. (a)
10
2
(b)
20
12
(c)
81
36
(d)
88
72
3. (a) 2
5
2
(b) 2
8
1
(c) 4
6
1
(d) 10
3
1
4. (a)
3
8
(b)
8
27
(c)
8
33
(d)
4
39
Questions for Practice 5
1.
12
7
2.
15
4
3.
24
19
4.
30
11
5. 1
20
3
6.
12
5
7.
2
1
8.
40
27
9.
21
8
10.
8
3
11. 4
4
1
12. 8
8
5
13. 7
30
11
14. 1
12
1
Basic Numerical Concepts 37
15. 2
9
2
16. 3
24
19
17. 2
24
7
18. 3
12
11
Questions for Practice 6
1.
4
21
= 5
4
1
2.
3
16
= 5
3
1
3.
6
15
= 2
2
1
4.
7
18
= 2
7
4
5.
18
2
=
9
1
6.
24
3
=
8
1
7.
24
7
8.
28
5
Questions for Practice 7
1. (a) 11.68 (b) 19.47
(c) 26.413 (d) 23.092
(e) 5.42 (f) 5.178
(g) 6.591 (h) 9.17
(i) 10.136 (j) 15.84
(k) 23.78 (l) 3.276
(m) 70 (n) 1.6
(o) 5.6 (p) 50
2. (a) 0.40 (b) 0.375
(c) 0.75 (d) 0.875
(e) 0.625 (f) 0.833
(g) 0.571 (h) 0.313
3. (a)
1000
125
=
8
1
(b)
100
60
=
5
3
(c)
100
175
= 1
4
3
(d)
1000
750
=
4
3
38 Basic Numerical Concepts
Questions for Practice 8
1. (a) 3% (b) 37.5%
2. (a)
50
1
(b)
25
2
(c)
40
3
(d)
50
9
3. (a) £30 (b) 90°
(c) 35 bottles (d) £483.30
(e) £24.15
4. (a) 20% (b) 33.33%
(c) 2.5% (d) 150%
Questions for Practice 9
1. Firm A: 1 : 4
Firm B: 1 : 9
Firm C: 1 : 3
Firm D: 1 : 5
2. (a) X: £15,000
Y: £20,000
Z: £25,000
(b) F: £1,800
G: £2,700
H: £2,700
Questions for Practice 10
1. (a) 2.34 × 10
2
(b) 4.5 × 10
÷2
(c) 6.42 × 10
3
(d) 5.032 × 10
÷4
(e) 8.453 × 10
8
(f) 3.792 × 10
÷15
2. (a) 468 (b) 0.0935
(c) 1,853,000 (d) 0.0000602
(e) 7,653,000,000,000 (f) 0.000000000816
39
Study Unit 2
Algebra, Equations and Formulae
Contents Page
Introduction 40
A. Some Introductory Definitions 40
Algebra 40
Equations 42
Formulae 42
B. Algebraic Notation 43
Multiplication 44
Division 45
Indices, Powers and Roots 45
Brackets in Algebra 48
C. Solving Equations 50
Equality of Treatment to Both Sides of the Equation 50
Transposition 52
Equations with the Unknown Quantity on Both Sides 54
D. Formulae 55
Formulating a Problem 55
Constants and Variables 56
Substitution 56
Rearranging Formulae 57
40 Algebra, Equations and Formulae
INTRODUCTION
In this study unit we continue our review of basic concepts by looking at algebra. Algebra
essentially involves the substitution of letters for numbers in calculations, so that we can
establish rules and procedures for carrying out mathematical operations which can be
applied whatever the actual numbers involved. In this sense then, algebra enables general
expressions to be developed and general results to be obtained.
We shall then go on to introduce the subject of equations. An equation is simply a statement
of equality between two mathematical expressions, and is generally used to enable one or
more unknown quantities to be worked out. There are various rules and procedures which
can be applied to equations to help this process, and these will be examined in full. Here we
shall be concerned with "simple" equations, but later in the course we shall consider more
complex forms. However, the same basic rules are applied, so it is important to understand
these at the outset.
Finally, we look briefly at formulae. A formula is one particular form of generalised
mathematical statement, usually expressed in the form of an equation, that sets out a rule
which may be applied in particular situations. It enables us to work out an unknown quantity
or value provided we know certain specific quantities or values. There are formulae which
describe a great many physical situations or mathematical problems and you will encounter
these throughout your studies, both in this course and many others.
Objectives
When you have completed this study unit you will be able to:
 outline the basic principles of algebra
 apply the basic arithmetic operations of addition, subtraction, multiplication and division
to algebraic notation, and simplify algebraic expressions by the process of collecting
like terms
 define equations and formulae and outline their uses
 find unknown quantities and values from simple equations by using transposition
 outline the principles of formulae and how they are constructed
 rearrange the terms in a formula to isolate different unknowns.
A. SOME INTRODUCTORY DEFINITIONS
Algebra
Algebra is a branch of mathematics in which, instead of using numbers, we use letters to
represent numbers.
We all know that 2 + 3 = 5.
Suppose, though, that we substitute letters for the first two numbers, so that:
2 = a
3 = b
We can then write:
a + b = 5
All that has happened is that we have replaced the numbers with letters. However, a number
is a specific quantity – for example, 5 is more than 4, but less than 6 – whereas a letter can
be used to represent any number. Thus, in the previous expression, "a" could be 4 and "b"
Algebra, Equations and Formulae 41
could be 1. We only know that they are 2 and 3 respectively because we defined them as
such before.
The main consequence of this is that algebra uses general expressions and gives general
results, whereas arithmetic (using numbers) uses definite numbers and gives definite results.
Arithmetic is specific, whereas algebra is general.
Let us consider some examples to illustrate this further.
Example 1:
Suppose you have a piece of wood which is 7 metres long and from it you wish to cut a piece
4 metres long. The length of the remaining piece is 3 m, calculated as follows:
7 ÷ 4 = 3
This is a specific arithmetic statement relating to cutting a specific amount from this particular
piece of wood.
We could translate this into an algebraic expression by substituting letters for the specific
lengths:
let the original length of the piece of wood = x metres
and the length of the piece cut off = y metres.
The calculation can now be shown as:
x ÷ y = 3
This is now a general statement for cutting one length of wood from another to leave a piece
3 metres in length.
Example 2:
To find the area of a floor measuring 10 m long and 9 m wide, we multiply one dimension by
the other:
area = 10 × 9 square metres
= 90 sq metres.
Substituting the letters "l" and "w" to represent the actual length and width, we can
reformulate the expression as:
area = l × w
Again, this is a general expression which can be made specific by putting in particular values
for "l" and "w".
Example 3:
The distance travelled by a train in 3 hours at a speed of 60 miles per hour is easily
calculated as:
3 × 30 = 180 miles.
Here again, letters may be substituted to give us a general expression:
s × t = d
where: speed = s mph
time = t hours
distance travelled = d miles.
42 Algebra, Equations and Formulae
Thus, using algebra – working with letters instead of numbers – allows us to construct
general mathematical expressions. This is not particularly helpful in itself, but is very
important when we come to consider equations.
Equations
An equation is simply a mathematical statement that one expression is equal to another. So,
for example, the statement that "2 + 2 = 4" is an equation.
Note that an equation has two sides. Here, they are "2 + 2" and "4". The two sides must
always be in equality for the statement to be an equation.
If we now introduce the concept of algebra into this, we have the makings of an extremely
useful mathematical tool.
For example, given that a certain number multiplied by 3 is 6, we can write this as:
3 × the required number = 6.
To save writing "the required number" (or the " unknown" number), it is more convenient to
call it "x", so that we can then write:
3 × x = 6
Here are a few simple statements which you can easily write as equations for practice:
(a) A certain number is added to 4 and the result is 20.
(b) A certain number is multiplied by 4 and the result is 20.
(c) If 4 is taken from a certain number the result is 5.
(d) If a certain number is divided by 3 the result is 1.
The correct equations are:
(a) x + 4 = 20
(b) x × 4 = 20
(c) x ÷ 4 = 5
(d) x ÷ 3 = 1
The point of constructing equations in this way is that it provides a means by which you can
work out the unknown value (here, "x"). These examples are very simple and you can
probably solve them (i.e. work out the unknown) very easily. However, equations can get
very complicated. Don't worry, though – there are a number of simple rules which can be
used to help solve them.
Formulae
A mathematical formula (plural "formulae") is a special type of equation which can be used
for solving a particular problem.
In fact, we have already introduced two formulae in the preceding section on algebra:
 area = length × width
 distance travelled = speed × time.
The essence of these two statements is that they are always true, whatever the actual values
of length, width, speed etc. are. Thus, a formula is an equation which always applies to a
particular mathematical problem, whatever the actual values. It provides a set of rules which
can be used in a particular situation in order to solve the problem.
Algebra, Equations and Formulae 43
If we know the actual values, we could put them into the equation, but if not we would have
to treat them as unknown. Thus, we can formulate a general statement using algebra as
follows:
 A = l × w
where: A = area
l = length
w = width.
 d = s × t
where: speed = s mph
time = t hours
distance travelled = d miles.
Having a formula for a particular problem set out as a general equation using algebra
enables us to work out the answer by substituting actual values back into it for the unknowns.
For example, if we know we have been driving for 2 hours at an average speed of 60 mph,
we can easily work out the distance travelled by substituting 60 for "s" and 2 for "t" in the
above formula:
d = s × t
= 60 × 2
= 120 miles.
There are many formulae used in mathematics and other physical sciences such as
chemistry or physics. You will meet a good many in your studies – principally in this subject
and in financial subjects. Again, do not worry too much about them. They are usually quite
straightforward and, by using the general rules governing equations, can easily be used to
solve the problem faced.
B. ALGEBRAIC NOTATION
As algebraic letters simply represent numbers, the operations of addition, subtraction,
multiplication and division are still applicable in the same way. However, in algebra it is not
always necessary to write the multiplication sign. So, instead of "a × b", we would write
simply "ab" (or sometimes "a.b", using the full stop to represent multiplication).
Addition and subtraction follow the same rules as in arithmetic. Thus, just as:
3 + 3 + 3 + 3 + 3 + 3 = 3 × 6
so
y + y + y + y + y + y = y × 6, or 6y
You should note that this holds good only for like terms. Like terms are those of the same
symbol or value – for example, all the terms to be added above were "y".
44 Algebra, Equations and Formulae
When unlike terms are to be added, we can simplify an arithmetical statement, but not an
algebraic statement. For example:
3 + 6 + 7 = 16
y + x + z = y + x + z
The three numbers, when added, can be reduced to a simpler form, i.e. 16. However, there
is not a simpler means of expressing "x + y + 3".
We can, though, simplify expressions when they involve like terms.
Just as the result of adding two numbers does not depend upon their order, so the result of
addition in algebra is not affected by order. Thus:
2 + 7 = 7 + 2 = 9
and
x + y + 3 = 3 + y + x = y + 3 + x, etc.
In simplifying an algebraic expression, we can collect like terms together whenever possible.
For example:
5x + 2y ÷ x + 3y + 2 = (5x ÷ x) + (2y + 3y) + 2 = 4x + 5y + 2
We have collected together all the "x"s (5x ÷ x) and all the "y"s (2y + 3y), but cannot do
anything with the "2" which is unlike the other terms.
This distinction between like and unlike terms, and the way in which they can be treated, is
an important point.
Multiplication
As we have seen, the product of a number and a letter, for example 3 × a, may be written as
3a. Always place the number before the letter, thus 3a, not a3. Since multiplication by 1 does
not alter the multiplicand, the expression 1a is not used; you just write a.
To multiply unlike terms, we simply write them together with no sign. Thus:
a × b × c × d = abcd
If numerical factors are involved then they are multiplied together and placed in front of the
simplified term for the letters:
3a × 2b × 4c × d = 24abcd
That this is correct can be shown by writing it in full and changing the order of multiplication:
3 × a × 2 × b × 4 × c × d
= 3 × 2 × 4 × a × b × c × d
= 24abcd
Multiplying like terms (for example, x × x) gives a result of the term being raised to the power
of the multiplier (here, x
2
). We shall consider this in more detail later.
Algebra, Equations and Formulae 45
Division
Division should present little difficulty if we follow the rules used in arithmetic. Consequently,
just as we write:
2
4
for 4 ÷ 2
so we can write:
y
x
for x ÷ y
Remember that any number goes into itself once. This will prevent you from saying that 6 ÷ 6
= 0 instead of 6 ÷ 6 = 1.
In the same way x ÷ x = 1. It is then easy to cancel algebraic fractions. For example:
4x
2xy
can be cancelled by dividing both the numerator and denominator by x to give
4
2y
4
2y
can be further cancelled by dividing both the numerator and denominator by 2 to
give
2
y
or we could simply say we can cancel
4x
2xy
by dividing both parts by 2x.
We can always cancel like terms in the numerator and denominator.
Indices, Powers and Roots
You will remember that indices are found when we multiply the same number by itself several
times. The same rules apply in algebra, but we can investigate certain aspects further by
using algebraic notation.
To recap:
x × x is written as x
2
(referred to as "x squared")
x × x × x is written as x
3
(referred to as "x cubed")
x × x × x × x is written as x
4
(referred to as "x to the power four" or "x raised to the
fourth power")
and so on.
In, for example, x
5
, x is termed the base and 5 is called the exponent or index of the power.
Note two particular points before we go on:
x = x
1
x
0
= 1
(a) Multiplication of indices
You will note that, for example:
x
2
× x
4
= (x × x) × (x × x × x × x)
= x × x × x × x × x × x
= x
6
= x
2 + 4
46 Algebra, Equations and Formulae
and
x
3
× x
5
= (x × x × x) × (x × x × x × x × x)
= x × x × x × x × x × x × x × x
= x
8
= x
3 + 5
There is in fact a general rule that, when we have two expressions with the same base,
multiplication is achieved by adding the indices. Thus:
x
m
× x
n
= x
m + n
(Rule 1)
(b) Division of indices
When we divide expressions with the same base, we find we can achieve this by
subtracting the indices. For example:
3
6
x
x
=
x x x
x x x x x x
× ×
× × × × ×
= x × x × x
= x
3
= x
6 ÷ 3
and
x
x
2
=
x
x x ×
= x
= x
1
= x
2 ÷ 1
The general rule is:
n
m
x
x
= x
m ÷ n
(Rule 2)
(c) Showing reciprocals
In Rule 2 above, let us put m = 0, so that the expression becomes :
n
0
x
x
= x
0 ÷ n
We know that x
0
= 1, so
n
x
1
= x
0 ÷ n
= x
÷ n
Thus, negatives denote reciprocals.
(d) Raising one power by another
To work out an expression such as (x
2
)
3
– i.e. x squared all cubed – we have x
2
× x
2
×
x
2
as it is the cube of x
2
.
Algebra, Equations and Formulae 47
Thus (x
2
)
3
= x
2
× x
2
× x
2
= x × x × x × x × x × x
= x
6
= x
2 × 3
In general, we have (x
m
)
n
= x
mn
(Rule 3)
(e) Roots
Indices do not have to be whole numbers. They can be fractional.
Let us consider x
½
(i.e. x raised to the power half). We know from Rule (3) that:
(x
½
)
2
= x
½ × 2
= x
Thus, x
½
when multiplied by itself gives x. However, this is exactly the property that
defines the square root of x. Therefore, x
½
is the square root of x which you will
sometimes see denoted as:
2
x or, more commonly, as . x
Therefore, x
½
÷ . x (The symbol "÷" means "identical with".)
Similarly, (
3
1
x )
3
= x. Thus,
3
1
x is the cube root of x which is denoted by the symbol
3
x .
Similarly, x
¼
is the fourth root of x denoted by
4
x .
Thus, in general, we can say that:
n
1
x is the n
th
root of x or
n
x .
Note that in all these expressions the number to the left of the root sign – for example
the 3 in
3
x – is written so that it is level with the top of the root sign. A number written
level with the foot of the root sign – for example, x 3 – would imply a multiple of the
square root of x – in this example, 3 times the square root of x.
We shall often need to use square roots in later work, particularly in statistics and in
stock control.
(f) Roots of powers
There is one more type of index you may meet – an expression such as:
or
3
5
x , or in general
n
m
x .
To interpret these, remember that the numerator of a fractional index denotes a power
and that the denominator denotes a root.
Thus,
2
3
x is (x
3
)
½
, i.e. the square root of (x cubed) or
2 3
x . Alternatively, it is (x
½
)
3
, i.e.
the cube of (the square root of x) or ( x )
3
.
Let us consider its value when x = 4:
2
3
4 =
2 3
4 = 4 4 4 × × = 64 = 8 (since 8 × 8 = 64)
or
2
3
4 = ( ) 4
3
= 2
3
= 2 × 2 × 2 = 8
Therefore it does not matter in which order you perform the two operations, and we
have:
n
m
x =
n m
x = (
n
x )
m
48 Algebra, Equations and Formulae
(g) Collecting like terms
We saw earlier how we can collect like items. Let us look at a couple of examples
where indices are involved. It is no more complicated, even though it may look that
way.
a
2
+ 2a
2
= a
2
+ a
2
+ a
2
= 3a
2
These are like terms and can be added.
b
2
+ 2b
2
+ 2b
3
= b
2
+ b
2
+ b
2
+ b
3
+ b
3
= 3b
2
+ 2b
3
Note that b
2
and b
3
are not like terms, so they cannot be combined.
Brackets in Algebra
The rules for brackets are exactly the same in algebra as in arithmetic. However, if we have
unlike terms inside the brackets, it is not possible to collect them together before removing
the brackets. In such cases, we need to be very careful when removing them and always
obey the following three rules
 If the bracket has a term (a number or a letter) in front of it, everything within the
bracket must be multiplied by that term.
 If a bracket has a + sign before it, or before the term by which the terms inside the
brackets are to be multiplied, the signs + or ÷ within the brackets remain unchanged.
 If a bracket has a ÷ sign before it, or before the term by which the terms inside the
brackets are to be multiplied, the signs + or ÷ within the brackets are changed so that +
becomes ÷ and vice versa.
These rules need a little explanation, so let us consider some examples.
Example 1:
4x + 2(x ÷ y)
The terms inside the brackets are unlike and cannot be further collected.
All the terms within the brackets must be multiplied by the term outside before removing the
brackets, so this gives:
4x + 2x ÷ 2y = 6x ÷ 2y
Example 2:
x ÷ (y + 3z)
Again, the terms inside the brackets are unlike and cannot be further collected, so we can
proceed to remove the brackets as the next step.
Since there is a ÷ sign before the brackets, we must reverse the signs of the terms inside
when removing them. Thus:
x ÷ y ÷ 3z
We can clarify this by imagining that there is an unwritten "1" before the bracket, making the
expression effectively:
x ÷ 1(y + 3z)
So "÷1" is multiplying the whole bracket:
x ÷ 1 × (y × 3z)
Algebra, Equations and Formulae 49
This gives us three separate terms:
x, ÷1 × y, and ÷1 × 3z
Thus, we have:
x ÷ y ÷ 3z
Example 3:
3a(a ÷ 3) ÷ 4b(b ÷ 4)
The first bracket is multiplied by 3a and the second by ÷4b. This gives us:
3a
2
÷ 9a ÷ 4b
2
+ 16b
Note that the ÷ sign before the second bracket changes the ÷ sign inside to +.
Often we find it necessary to insert brackets into an algebraic expression. This is known as
factorisation because, when we include the brackets, we place any common factor outside
the brackets.
Thus:
x + 5y + 5z = x + 5(y + z)
x ÷ 5y + 5z = x ÷ 5(y ÷ z)
Notice how we must keep an eye on the signs as well.
Questions for Practice 1
Simplify the following expressions, writing "no simpler form" if you think any cannot be further
simplified:
1. a + a + a + a 2. ab + ab
3. 3x + 2x + x 4. 3x + 2y + y
5. 5x ÷ 3x + x 6. 7xy ÷ 2x + y
7. 3a ÷ 2a 8. a + a + b + a ÷ 3
9. x + y + xy 10. x + x + x + y + y + y
11. a + 2b + 3a + 4b +5 12 3x ÷ (x ÷ 3)
13 x + 2(y + z) 14. 4 ÷ 3(÷2x + y) + (÷x ÷y)
15. x ÷ (y ÷ 3z) 16. 4x + 2(x ÷ y)
17. a(b ÷ c) ÷ b(c ÷ a) 18. 3a(a ÷ 3) ÷ 4b(b ÷ 4)
Now check your answers with those given at the end of the unit.
50 Algebra, Equations and Formulae
C. SOLVING EQUATIONS
When we talk about solving an equation, we mean finding the value of the unknown or
unknowns using the other numbers in the equation. For example, in:
4x = 20
the value of the unknown (x) can be discovered because of the stated relationship between 4
and 20. You can easily see that if 4 times the unknown = 20 then the unknown must be 5.
There are a number of rules which can be used to help solve equations and we shall
consider these next in relation to equations which have only one unknown. In a later study
unit we shall examine more complex equations – those with two or more unknowns – and
consider additional rules. However, these are based on the simple rules which we set out
here.
Equality of Treatment to Both Sides of the Equation
This is the golden rule that you must always follow when handling equations and there are no
exceptions.
What you do to one side of the equation, you must also do to the other.
Remember that an equation has two sides and those two sides must always be in equality
for the whole statement to remain an equation. So, anything we do to the equation must
maintain this equality.
Another way of thinking about this is that an equation is like a balance. If the weights of a
balance are equal on both sides it is "in balance". You can add an equal weight to each side,
or take an equal weight from each side, and it will remain "in balance".
Looking at the rule in more detail, we can see a number of possible operations which may be
carried out without changing the equality of the equation.
(a) The same number may be added to both sides of the equation
For example:
x ÷ 4 = 6
If we add 4 to each side, then we get:
x ÷ 4 + 4 = 6 + 4
x = 10
(b) The same number may be subtracted from both sides of the equation
For example:
x + 6 = 18
If we deduct 6 from each side, we get:
x + 6 ÷ 6 = 18 ÷ 6
x = 12
(c) Both sides of the equation may be multiplied by the same number.
For example:
3
1
of a number = 10, or
3
1
x = 10, or
3
x
= 10
Algebra, Equations and Formulae 51
If we multiply both sides by 3, we get:
3
x
× 3 = 10 × 3
3
x 3
= 30
x = 30
(d) Both sides of the equation may be divided by the same number.
For example:
2 times a number is 30, or 2x = 30
If we divide both sides by 2, we get:
2
x 2
=
2
30
x = 15
You should have noticed that the object of such operations is to isolate a single unknown, x,
on one side of the equals sign.
Questions for Practice 2
Simplify the following expressions, writing "no simpler form" if you think any cannot be further
simplified:
1. x ÷ 7 = 21
2. x ÷ 3 = 26
3. x + 9 = 21
4.
5
x
= 11
5.
8
x
= 7
6. 3x = 30
Now check your answers with those given at the end of the unit.
52 Algebra, Equations and Formulae
Transposition
When you understand the rules in the previous subsection you may shorten your working by
Transposition is a process of transferring a quantity from one side of an equation to another
by changing its sign of operation. This is done so as to isolate an unknown quantity on one
side.
You must observe the following rules:
 An added or subtracted term may be transposed from one side of an equation to the
other if its sign is changed from + to ÷, or from ÷ to +.
Consider the following examples:
(a) x ÷ 4 = 6
Transfer the 4 to the right-hand side of the equation and reverse its sign:
x = 6 + 4
x = 10
(b) x + 6 = 18
Transfer the 6 to the right-hand side and reverse its sign:
x = 18 ÷ 6
x = 12
(c) 12 + x = 25
Transpose the 12 and reverse its sign:
x = 25 ÷ 12
x = 13
(d) 18 ÷ x = 14
First transpose the x and reverse its sign:
18 = 14 + x
Then transpose the 14 and reverse its sign:
18 ÷ 14 = x
x = 4
Note that we can keep transposing terms from one side to the other until we have
isolated the unknown quantity.
 A multiplier may be transposed from one side of an equation by changing it to the
divisor on the other.
Similarly, a divisor may be transposed from one side of an equation by changing it to
the multiplier on the other.
Consider the following examples:
Algebra, Equations and Formulae 53
(a)
4
x
= 12
To get x by itself, move 4 across the = sign and change it from divisor to
multiplier:
x = 12 × 4
x = 48
(b) 2x = 20
To get x by itself, transpose the 2 and change it from multiplier to divisor:
x =
2
20
x = 10
(c)
x
24
= 6
In this case, we need to transpose x and change it from divisor to multiplier:
24 = 6x
Then, to get x by itself, we transpose the multiplier (6) and change it to a divisor:
6
24
= x
x = 4
This rule is called cross multiplication.
Questions for Practice 3
Find the value of x by transposition:
1.
6
x
= 36
2.
x
72
= 24
3. 3x + 3 = 21
4. 5x = 25
5. 2x ÷ 3 = 15
6. 21 ÷ x = 15
Now check your answers with those given at the end of the unit.
54 Algebra, Equations and Formulae
Equations with the Unknown Quantity on Both Sides
These equations are treated in the same way as the other equations we have met so far. We
simply keep transposing terms as necessary until we have collected all the unknown terms
on one side of the equation.
For example, consider 3x ÷ 6 = x + 8
First move the ÷6 to the right-hand side, making it +6:
3x = x + 8 + 6
Then transpose x from right to left, changing its sign:
3x ÷ x = 8 + 6
2x = 14
Finally, move the 2 to the right-hand side changing it from multiplier to divisor, and thus
isolating x on the left:
x =
2
14
x = 7
Questions for Practice 4
Find the value of x:
1. x + 3 = 8 ÷ 2x
2. 2x = 24 ÷ x
3.
2
1
x = 15 ÷ x
4. 5x ÷ 3 = 27 + 2x
5. 18 + x = 5x ÷ 2
6.
3
x 2
= 24 + 6x
Now check your answers with those given at the end of the unit.
Algebra, Equations and Formulae 55
D. FORMULAE
A formula is a mathematical model of a real situation.
The easiest way to explain this is to give an example.
Formulating a Problem
When a person borrows money, he or she has to pay interest on the loan (called the
"principal"). The amount of interest payable is determined by the annual rate of interest
charged on the money borrowed, taking into account the time over which the loan is repaid. If
we want to work out the amount of interest repayable on a particular loan, the calculation is
as follows:
Interest = the principal multiplied by the percentage rate of interest, multiplied by the
number of time periods the interest is to apply.
We could model this situation by using algebraic notation and then show the whole
calculation as an equation:
let:: I = simple interest
P = principal
r = rate of interest (%)
n = number of periods
then:
I = P ×
100
r
× n
This equation is the formula for calculating simple interest. It can be used for any simple
interest calculation and enables us to find the value of any one of the elements, provided of
course that we know all the others.
Thus, what we have done is express the real situation as a mathematical model using an
algebraic equation.
This particular formula is very important in financial studies and can be applied in many
situations. However, it is possible to develop formulae to model any situation involving
mathematical relationships. Some are just useful for helping to understand how to work out a
particular problem or situation you are faced with, whilst others describe key mathematical
relationships which you will meet time and again throughout the course.
For example, consider the following situation:
A man runs at a certain number of kilometres per hour for a certain number of hours, and
then cycles at a certain number of kph for a certain number of hours. What is his average
speed?
let: x = the speed at which he runs (in kph)
y = the number of hours he runs
p = the speed at which he cycles (in kph)
q = the number of hours he cycles.
Therefore, we can say he covers (xy + pq) kilometres in (y + q) hours.
then:
average speed =
q + y
pq + xy
kph
56 Algebra, Equations and Formulae
Note that when we use a letter such as x to represent an unknown quantity, it only represents
that quantity and the units must be exactly specified. For example, we said "let x = the speed
at which he cycles (in kph)"; to have said "let x = the speed at which he cycles" would have
been inaccurate.
The steps to take in formulating the problem, then, are as follows:
 select a letter to represent each of the quantities required
 convert each block of information from the problem into an algebraic term or statement,
using the letters selected
 form an equation by combining the symbolic terms, ensuring that all terms are
expressed in the same units.
The last step is to test the formula by going back to the problem and check that it works by
using real number values in the formula through the process of substitution. We shall look at
substitution shortly.
Activity
As practice, see if you can work the following problem out by developing a formula. Note that
the unknown quantities have already been given letters.
A grocer mixes m kg of tea at x pence per kg with n kg of tea at y pence per kg. If she sells
the mixture at v pence per kg, what profit, in pence, does she make? Bring your answer to £.
Now check your answer with that given at the end of the unit.
Constants and Variables
As we have seen, formulae consist of letters (algebraic notation) and, sometimes, numbers.
These are combined together in particular ways to express the mathematical relationships
between the terms such that a particular result is obtained.
The numbers in a formula always stay the same and are known as constants. So, for
example, in the formula for simple interest, the constant will always be 100, no matter what
the values given to the P, r and n in a particular situation.
The letters on the other hand are variables. The value that they have in working out the
formula in any particular situation may vary – it depends on the particular circumstances of
the problem.
Substitution
We defined a formula as a mathematical model of a real situation. In order to use that model
in a real situation, we have to give the variables the values which apply in that particular
situation. Doing this is the process of substitution.
Turning back to the formula for simple interest, consider the following example:
What is the simple interest on £1,000 invested at 10% per annum (i.e. per year) for two
years?
First of all we should restate the formula:
I = P ×
100
r
× n
Algebra, Equations and Formulae 57
Now we substitute the values from the problem into the formula:
P = £1,000
r = 10
n = 2
So:
I = £1,000 ×
100
10
× 2
=
100
20 10 000 1 × × , £
= £200
This explains the principle of substitution. In the next study unit you will get much more
practice in this.
Rearranging Formulae
Finally, we should note that a formula can be rearranged to find a different unknown quantity.
So for example, when considering simple interest, we may wish to know how long it would
take to earn a specified amount of interest from a loan made to someone else, at a particular
rate of interest.
The way in which the formula is expressed at present does not allow us to do that, but by
using the rules we have discussed in relation to simple equations, we can rearrange the
formula so that it does.
Again, we start by restating the original formula:
I = P ×
100
r
× n
Note that we can also express this as:
I =
100
n r P
To find n, we need to rearrange the formula so that n is isolated on one side of the equation.
We can do this by dividing both sides by P ×
100
r
or, as it may also be expressed,
100
r P
:
|
.
|

\
|
100
Pr
I
=
|
.
|

\
|
|
.
|

\
|
100
Pr
n
100
Pr
n =
|
.
|

\
|
100
Pr
I
n = I ÷
100
Pr
= I ×
Pr
100
n =
Pr
I 100
58 Algebra, Equations and Formulae
Don't worry too much about the mechanics of this for now – again you will have plenty of
practice at these types of operation as you work through the course. However, see if you can
rearrange the simple interest formula to isolate the other two unknowns.
Questions for Practice 5
Rearrange the simple interest formula to allow the following calculations to be made:
1. To find P – the principal which would provide a certain amount of interest at a particular
rate of interest over a specified period.
2. To find r – the rate of interest which would have to be charged to earn a particular
amount of interest on a specified principal over a certain period of time.
Now check your answers with those given at the end of the unit.
Algebra, Equations and Formulae 59
Questions for Practice 1
1. 4a
2. 2ab
3. 6x
4. 3x + 3y
5. 3x
6. No simpler form.
7. a
8. 3a + b ÷ 3
9. No simpler form.
10. 3x + 3y
11. 4a + 4b + 5
12 3x ÷ x + 3 = 2x + 3
13 x + 2y + 2z
14. 4 + 6x ÷ 3y ÷ x ÷ y = 4 + 5x ÷ 4y
15. x ÷ y ÷ 3z
16. 4x + 2x ÷ 2y = 6x ÷ 2y
17. ab ÷ ac ÷ bc + ab = 2ab ÷ ac ÷ bc
18. 3a
2
÷ 9a ÷ 4b
2
+ 16b
Questions for Practice 2
1. x = 28
2. x = 29
3. x = 12
4. x = 55
5. x = 56
6. x = 10
Questions for Practice 3
1. x = 36 × 6 = 218
2. 72 = 24x,
24
72
= x, x = 3
3. 3x = 21 ÷ 3, x =
3
18
, x = 6
60 Algebra, Equations and Formulae
4. x =
5
25
= 5
5. 2x = 15 + 3, x =
2
18
= 9
6. 21 = 15 + x, 21 ÷ 15 = x, x = 6
Questions for Practice 4
1. x + 2x = 8 ÷ 3, x =
3
5
= 1
3
2
2. 2x + x = 24, x = 8
3.
2
1
x + x = 15, x = 10
4. 5x ÷ 2x = 27 + 3, x = 10
5. 18 + 2 = 5x ÷ x, x = 5
6. ÷24 = 6x ÷
3
x 2
= 5
3
1
x, x = ÷24 ÷ 5
3
1
= ÷24 ×
16
3
=
16
72
=
2
9
= ÷4½
Questions for Practice 5
1. We always start by restating the original formula:
I = P ×
100
r
× n or I =
100
n r P
To isolate P, we divide both sides by
100
r
× n, or
100
n r
|
.
|

\
|
100
n r
I
=
|
.
|

\
|
|
.
|

\
|
100
n r
100
n r
P
|
.
|

\
|
100
n r
I
= P
P = I ÷
100
n r
= I ×
n r
100
So P =
n r
I 100
I = P ×
100
r
× n, or I =
100
n r P
To find
100
r
, we divide both sides by P × n:
Algebra, Equations and Formulae 61
Pn
I
=
Pn
Pn
100
r
|
.
|

\
|
Pn
I
=
100
r
100
r
= r%. which is what we want to find
so r% =
Pn
I
Activity
The solution would be as follows:
profit = selling price ÷ cost price
selling price = quantity sold × price = v(m + n) pence
cost price = mx + ny pence
profit = v(m + n) ÷ (mx + ny) pence
To convert this to £, we simply divide the result by 100:
profit = £
100
ny) + (mx n) + (m v ÷
62 Algebra, Equations and Formulae
63
Study Unit 3
Contents Page
Introduction 64
A. Currency and Rates of Exchange 64
Rates of Exchange 64
Currency Conversion 66
B. Discounts and Commission 68
Discounts 68
Commission 68
C. Simple and Compound Interest 70
Simple Interest 71
Compound Interest 72
D. Depreciation 76
Straight-Line Depreciation 76
Reducing Balance Depreciation 77
E. Pay and Taxation 79
Make Up of Pay 79
Taxation 81
Answers to Questions for Practice 84
INTRODUCTION
In this study unit we shall apply some of the numerical concepts introduced in the previous
two units to some of the basic problems encountered regularly in the business world. The
intention here is not to go into detail about the nature of these problems, but to concentrate
on the principles of calculation involved in solving them. Thus, in most cases, we shall use
very simplified situations and limited amounts of information.
At first sight, some of these business applications may seem quite hard to work out, but if
you follow the logical steps to breaking down the problem shown here, applying the correct
formulae and setting out the calculations clearly, you should not find them difficult in practice.
Objectives
When you have completed this study unit you will be able to:
 explain the rate of exchange between different currencies and convert values in one
currency into another
 explain discounts and commissions and calculate them from given information
 state the formulae for simple interest and compound interest, and use them to calculate
any of the amount of interest payable (or receivable), the principal, the rate of interest
or the time period covered, given information about the other variables
 explain depreciation and calculate it according to the straight-line method and the
reducing-balance method
 outline the way in which gross pay is made up and the main deductions made from
gross pay in order to leave net or take-home pay
 calculate the income tax payable on pay, taking into account non-taxable items and
personal tax allowances.
A. CURRENCY AND RATES OF EXCHANGE
Each country has its own system of coinage and this is known as its currency. The process
of finding the value of the currency of one country in terms of the currency of another country
is called exchange.
Rates of Exchange
You are no doubt familiar with various currency systems. They are invariably based on the
decimal number system – for example, America uses the dollar (\$) and the cent, which is
equivalent to one-hundredth of a dollar. Europe has the euro (€) and the cent, which is one-
hundredth of a euro, and present British currency comprises the pound sterling (£) and the
penny, which is one-hundredth of a pound.
The relative values of the different currencies used to depend upon the amount of gold which
a country possessed, but now that we are no longer on the gold standard, their values
depend on the political and economic circumstances of the countries concerned. Such value
is measured by how much one unit of currency costs in another currency. For example, one
unit of British currency (£1) might cost 2.11 units of US currency (\$2.11). This relationship of
one currency to another is known as the rate of exchange.
These values vary from day-to-day and are quoted in most daily papers. The changes in
value are brought about by such things as changes in government, changes in government
policies, the national budget, the relationship between the value of exports and imports, etc.
To give you some idea of the scale of changes over time, before the Second World War the
value of the dollar was about 21p (i.e. \$1 = £0.21), whereas at the time of writing (2008) it is
about 50p (i.e. \$1 = £0.50). However, such large changes usually only occur over a long
period or because of some particular economic problem. In general, the changes are very
small, although they can take place quite rapidly. Even then, although the changes may be
very slight, their effects can be very large when you consider that large amounts of money
could be involved. Companies may have many millions of pounds sterling invested in
overseas currencies, and a small change in the value of those currencies can produce
substantial losses or gains, depending on whether the value of the currency rises or falls
against the pound.
The rate of exchange is expressed as a ratio. It shows the value of one unit of currency in
terms of another currency. Thus, one side of the ratio is always "1". Here are some examples
of typical rates of exchange of the British pound:
£1 = 2.00 Swiss francs
£1 = \$1.98 (US dollars)
£1 = €1.43 (euros)
£1 = 237.7 Japanese yen
Note that, although it is a ratio, the rate of exchange is usually shown as "x = y", rather than
"x : y".
It is this ratio that determines how much of one currency can be bought with another
currency, or how much a particular amount of one currency is worth in another.
Before we look at the relevant calculations involved in converting one currency to another,
there are two further points to note about the rate of exchange.
 For any relationship between two different currencies, the rate of exchange can be
expressed in two ways – how much one unit of one currency is worth in terms of the
other, and vice versa. Thus, the rate of exchange between the British pound and the
Swiss franc could be shown as:
£1 = 2.00 Swiss francs or
1 Swiss franc = £0.50
 The rate of exchange is usually quoted in two forms – a selling price and a buying
price. The selling price determines the amount of foreign currency you will be sold for
return for your foreign currency. (The difference between the two enables the bank or
bureau de change etc. undertaking the exchange to make a profit.) As an example of
the two different prices, the rates quoted might be:
£1 = 2.0076 Swiss francs 2.0069 Swiss francs
£1 = \$1.9832 (US dollars) \$1.9830 (US dollars)
£1 = €1.286 (euros) €1.2845 (euros)
Currency Conversion
To find the value of a sum of money in one currency in terms of another currency, we could
develop a simple formula using the rate of exchange, as follows.
Let: e = rate of exchange expressed as the number of units of a second currency (e.g.
dollars) equivalent to one unit of a first currency (e.g. pounds, as in \$1.98 = £1)
x = amount in first currency
y = amount in second currency.
To convert a certain amount of the first currency (say, £x) into the second currency (say \$y),
we multiply the amount of the first currency by the rate of exchange. Thus the formula would
be:
y = x × e
So, to convert £100 into \$ at the rate of £1 = \$1.98, we would substitute the following values
into the formula:
x = 100
e = 1.98
The formula gives us the following calculation:
\$y = 100 × 1.98 = \$198
If we wanted to know how much a particular sum of the second (say \$y) currency would be
worth in the first (say £x) currency, we would have to rearrange the above formula to isolate
x. To do this, we simply divide each side of the equation by e – the rate of exchange:
e
y
=
e
e x ×
x =
e
y
We can now apply these formulae to any given problem.
Example 1:
What is the cost in £ of an invoice for €1,275 euros if the rate of exchange is £1 = 1.29?
The rate of exchange is expressed as 1.29 units of one currency is equivalent to 1 unit of the
other. So, euros are the second currency and pounds are the first, and x is the amount in
pounds and y the amount in euros.
Substituting in the second formula (to find x) we get:
£x = £
1.29
1,275
= £988.37
(Note that the actual result of the calculation is £988.37209, but we round down to express it
correct to two decimal places as required for currency.)
Example 2:
Find the value of £63.40 in Swiss francs if the rate of exchange is £1 = SF2.02.
The rate of exchange is expressed as 2.02 units of the second currency (SF) is equivalent to
1 unit of the first (£). So, again x is the amount in pounds and y the amount in Swiss francs.
Substituting in the first formula (to find y) we get:
y SF = 63.40 × 2.02 = 128.068 = 128.07 Swiss francs (rounded up).
Example 3:
Convert \$3,000 to pounds given that £1 = \$1.99.
The rate of exchange is expressed as 1.99 units of the second currency (\$) is equivalent to
1 unit of the first (£). So, again x is the amount in pounds and y the amount in dollars.
Substituting in the second formula (to find x) we get:
£x = £
99 . 1
000 , 3
= £1,507.54 (rounded up).
Example 4:
Convert 100,000 Japanese yen to pounds given that £1 = 195.3 yen.
The procedure should be becoming clear to you now.
£x = £
3 . 195
100,000
= £512.03 (rounded down).
Example 5:
Find the value of £250 in dollars if the exchange rate is £1 = \$2.01
\$y = 250 × 2.01 = £502.50
Questions for Practice 1
1. John and Paul ate a meal in a restaurant whilst on holiday in Cyprus. The meal for two
cost £ (Cyprus) 35.70. Calculate the cost of the meal in pounds. The exchange rate
was 0.7888 £ Cyprus to the British pound.
2. The Smiths spent a holiday touring in France. While travelling they used 200 litres of
petrol which cost €1.35 per litre. The exchange rate was 1.29 euros to the pound.
(a) how much did the petrol cost them in pounds?
(b) what was the price per litre of the petrol in pence?
3. Before going on holiday to Switzerland and Italy, the Jones family changed £800 into
Swiss francs. While in Switzerland they spent 720 Swiss francs and then changed the
remaining francs into euros as they crossed the border into Italy. The exchange rate
was 2.05 Swiss francs to the pound and 0.642 euros to the Swiss franc. Calculate:
(a) the number of Swiss francs they received
(b) the number of euros they bought.
4. A bank buys Swedish kroner (SEK) at 10.05 kroner to the pound and sells at 9.45
kroner to the pound.
(a) Mr Watson changed £100 into Swedish Kroner for a trip to Sweden. How many
(b) Sadly the excursion was cancelled and so he changed all his Kroner back into
pounds. How much money did he lose because of the cancellation?
Now check your answers with those given at the end of the unit.
B. DISCOUNTS AND COMMISSION
You've probably heard the terms "discount" and "commission". What do they mean to you?
Discounts and commissions are a practical use of percentage calculations. Let us look in
more detail at their exact meanings and the way in which they are calculated.
Discounts
A discount is a reduction in the selling price of an item, usually offered as an inducement for
quantity buying or for prompt payment. The discount is expressed as a percentage of the list
price (the list price is the original price) and what the buyer actually pays is the net price – the
list price minus the discount.
To calculate a discount (D), you use the following rule or formula:
multiply the list price (or base, B) by the rate of the discount (R) and divide by 100.
Thus:
D = B ×
100
R
Consider the following examples.
Example 1:
A three-piece suite was listed as £720, but was sold at a discount of 15%. What is the net
price?
discount = £720 ×
100
15
= £108
net price = £720 ÷ £108 = £612
Example 2:
A car was listed at £3,600 but was sold for cash less a discount of 20%. What was the final
selling price?
discount = £3,600 ×
100
20
= £720
net price = £3,600 ÷ £720 = £2,800
Commission
A commission is the amount of money paid to an agent, broker or salesman for buying or
selling goods or services. This is expressed as a percentage of the value of the
goods/services traded and is called the rate of commission.
To find the commission, we simply multiply the value of the goods/services traded (known as
the principal) by the rate of commission.
Consider the following example.
Pauline obtained £6,400 of new business for the PR agency for which she works. She is paid
commission of 12½%. Calculate the amount due to her.
commission = 12½% of £6,400
= £6,400 ×
100
5 . 12
= £6,400 ×
200
25
= £800
Commissions are also usually charged by banks, travel agents and bureaux de change to
cover their costs when carrying out a currency exchange. The commission on changing cash
is commonly £3 or 2% whichever is larger.
When going on holiday, most people take a limited amount of cash in the currency of each
country to be visited and take the remainder that they will need in the more secure form of
traveller's cheques. The commission charged on traveller's cheques is commonly £3 or 1%
whichever is larger.
In these cases, the person exchanging the currency or buying traveller's cheques is charged
either:
 a fixed cost – £3 per transaction or
 a variable cost – the value of 1% or 2% of the amount of the transaction.
We shall consider fixed and variable elements in more detail later, but consider the following
example in relation to currency exchanges.
Using the rates of commission quoted above, and given that £1 = \$1.47, how many dollars
would you get when converting:
(a) £100 to American dollars in cash?
(b) £200 to American dollars in cash and in traveller's cheques?
In each case, we would need to work out the commission at both the fixed and the variable
rate, and then apply whichever is the larger.
(a) The commission payable on converting £100 into dollars cash is either £3 or £2 (2% of
£100). The fixed rate is the larger so that would be applied:
£100 ÷ £3 = £97
£97 × 1.47 = \$142.50
(b) The commission payable on converting £200 into dollars cash is either £3 or £4 (2% of
£200). The variable rate is the larger so that would be applied:
£200 ÷ £4 = £196
£196 × 1.47 = \$288.12
However, the commission payable on converting £200 into \$ traveller's cheques is
either £3 or £2 (1% of £200). The fixed rate is the larger so that would be applied:
£200 ÷ £3 = £197
£197 × 1.47 = \$289.59
Questions for Practice 2
1. What is the price paid for a £310 washing machine with a discount of 10%?
2. What is the price paid for a £7,000 car with a discount of 15%?
3. An estate agent charges a commission of 1.5% of the value of each house she sells.
How much commission is earned by selling a house for £104,000?
4. Calculate the commission earned by a shop assistant who sold goods to the value of
£824 if the rate of commission is 3%.
5. A car salesman is paid 2.5% commission on his weekly sales over £6,000. In one
particular week, he sold two cars for £7,520 and £10,640. What was his commission for
that week?
6. On a holiday touring New Zealand, Australia and Hong Kong, a family of four decided
to take the equivalent of £900 abroad. The travel agent charges commission on each
currency exchange at the rate of £3 or 2% (whichever is the larger) for cash and £3 or
1% (whichever is the larger) for travellers cheques.
(a) They changed £110 into New Zealand dollars, £100 into Australian dollars and
£120 into Hong Kong dollars. How much of each currency did they receive after
commission was deducted?
(b) After buying the foreign currencies, they changed as much as possible of the
remaining money into traveller's cheques. The smallest value of traveller's
cheque which can be bought is £10.
(i) How much did they exchange for travellers cheques?
(ii) How much commission did they pay for the cheques?
(c) What was the total cost per person of the foreign currencies and travellers
cheques?
Use the following exchange rates:
£1 = .798 New Zealand dollars (NZD)
£1 = 2.458 Australian dollars (AUD)
£1 = 19.65 Hong Kong dollars (HKD)
Now check your answers with those given at the end of the unit.
C. SIMPLE AND COMPOUND INTEREST
We introduced the subject of interest in the last study unit and will now consider it in more
detail. Interest calculations involve the application of simple formulae to specific problems.
Interest is what we call the amount of money someone will give you for letting them borrow
your money or what you pay for borrowing money from someone else. So you can receive
interest on a loan you make to someone and you can also be asked to pay interest on a loan
which you take out. Note that when you receive interest on savings from a bank or a building
society, this is the bank or building society paying interest to you on money which you have
loaned to them.
Let us consider the position from the point of view of the borrower. It is often the case that
individuals or companies wish to spend money which they do not currently have, but which
they feel sure they will have in the future. Therefore they borrow that money from a lender –
a bank, building society, credit card company, etc. – and pay the amount borrowed back to
the lender over a particular period. The cost charged by the lender can vary a great deal
depending on the method of obtaining credit, the amount borrowed and the time over which
the loan is repaid. This cost is then quoted as an annual percentage rate of interest to be
charged on the loan (known as the principal).
Usually finance companies quote a particular rate of interest applicable to either loans taken
out or savings placed with them, again the rate varying according to the amount of the
loan/savings and also the period of the loan (or the amount of time you guarantee to leave
the savings with them).
There are two ways of looking at interest – simple interest and compound interest.
Simple Interest
Simple interest (SI) is calculated on the basis of having a principal amount, say P, in the bank
for a number of years, T, with a rate of interest, R. As we have seen, there is a simple formula
to work out the amount of interest your money will earn:
100
PRT
SI =
This means that the interest is found by multiplying the principal by the rate, then by the time,
and then dividing by 100.
(Note that the way in which this formula is expressed is slightly different from that used in the
previous unit. It works in exactly the same way, but is stated in a rather easier fashion. Note,
too, that the notation is also slightly different. It is often the case that formulae can be
expressed in slightly different ways.)
As we have seen, we can also rearrange this formula to allow us to work out any of the
variables. Thus:
P
RT
100SI
=
R%
PT
SI
=
T
PR
100SI
=
Consider the following examples.
Example 1:
John had £16.40 in an account that paid simple interest at a rate of 9%. Calculate how much
interest would be paid to John if he kept the money in the account for 5 years.
The principal (P) is £16.40, the rate (R) is 9% and the time (T) is 5 years. So, substituting in
the formula, we get:
SI =
100
5 9 16.40 × ×
= £5.90
Example 2:
Calculate the length of time taken for £2,000 to earn £270 if invested at 9%.
Here we need to work out T where P = £2,000, SI = £270 and R = 9. Substituting in the
formula, we get:
T =
9 2,000
270 100
×
×
=
000 , 18
000 , 27
= 1.5 years
Questions for Practice 3
1. What is the simple interest on £250 over 3 years at 8.75% interest rate?
2. If you invest £6,750 at 8.5% per annum, how much interest will you earn and how
much will you have in your account after 4 years?
3. If you invest £5,000 at 9.25% per annum for 6 months, how much interest will you
4. £10,800 was invested and at the end of each year the interest was withdrawn. After 4
years, a total of £3,240 in interest had been withdrawn. At what annual rate of interest
was the money invested?
Now check your answers with those given at the end of the unit.
Compound Interest
Let us now look at how compound interest differs from simple interest.
If an amount of money is invested and the interest is added to the investment, then the
principal increases and so the interest earned gets bigger each year.
Suppose you begin with £1,000 and the interest is 8% per year. The interest in the first year
is 8% of £1,000 which is £80. Adding this to the original principal, the amount invested at the
end of the first year is £1,080.
Now, in the second year, the interest is 8% of £1,080 which is £86.40. The amount accrued
at the end of the second year is £1,166.50.
You can continue working out the amounts in this way. However, to find the amount at the
end of any year, you are finding 100% + 8% = 108% of the amount at the beginning of the
year. So you can multiply by 1.08 each time.
There is a standard formula which can be used to calculate compound interest, as follows:
A =
n
100
r
1 P |
.
|

\
|
+
where: A = accrued amount
P = original principal
r = rate of interest (for a particular time period, usually annual)
n = number of time periods.
Consider the following examples.
Example 1:
£3,000 is invested in an account paying 12% compound interest per year. Find the value of
the investment after 3 years.
Working this out year by year, we get:
investment at the beginning of year 1 = £3,000
interest at 12% (of £3,000) = £360
value of the investment after 1 year = £3,000 + £360 = £3,360
interest at 12% (of £3,360) = £403.20
value of investment after 2 years = £3,360 + £403.20 = £3,763.20
interest at 12% (of £3,763.20) = £451.58
value of the investment after 3 years = £3,763.20 + £451.58 = £4,214.78
Alternatively, we could use the formula:
A =
n
100
r
1 P |
.
|

\
|
+
where: P = £3,000
r = 12%
n = 3
Substituting these values in the formula gives:
A =
n
100
r
1 P |
.
|

\
|
+
= £3,000
3
100
12
1 |
.
|

\
|
+ = £3,000 × 1.12
3
= £3,000 × 1.404928
= £4,214.78 (to the nearest p).
Example 2:
£2,000 is invested in a high interest account for 1½ years. The interest rate is 10.5% per
annum, and it is paid into the account every six months. Calculate the value of the
investment after this time and the amount of interest earned.
The rate of interest for 1 year = 10.5%.
Therefore the rate of interest for 6 months = 5.25%.
Working this out by each six month period, we get:
investment at the beginning of year 1 = £2,000
interest at 5.25% (of £2,000) = £105
value of the investment after 6 months = £2,000 + £105 = £2,105
interest at 5.25% (of £2,105) = £110.51
value of investment after one year = £2,105 + £110.51 = £2,215.51
interest at 5.25% (of £2,215.51) = £116.31
value of the investment after 1½ years = £2,215.51 + £116.31 = £2,331.82
Total compound interest earned = £2,331.82 ÷ £2,000 = £331.82
Note that at each stage the amount of interest has been rounded to the nearest penny.
Alternatively, we could use the formula:
A =
n
100
r
1 P |
.
|

\
|
+
where: P = £2,000
r = 5.25%
n = 3
Substituting these values in the formula gives:
A =
n
100
r
1 P |
.
|

\
|
+
= £2,000
3
100
25 . 5
1 |
.
|

\
|
+ = £2,000 × 1.0525
3
= £2,000 × 1.1659134
= £2,331.83 (to the nearest p).
Total compound interest earned = £2,331.83 ÷ £2,000 = £331.83
So far we have used the compound interest formula
A =
n
100
r
1 P |
.
|

\
|
+
to work out how much a sum of money today (i.e. the original principal P) would be worth at
some point in the future (i.e. the accrued amount (A)). In other words, looking at our formula,
the original principal (P), the rate of interest (r) and the number of time periods (n) are known
and the uncertain variable is the accrued amount (A).
However, there are some situations when we might want to know how much we would need
to invest today (i.e. the original principal (P)) to achieve a certain level of known return at
some point in the future (i.e. the accrued amount (A)). For example, a financial manager
might want to know how much to invest now in a bank account in order to accrue a certain
sum of money for a known future expenditure (e.g. the replacement of an asset – such as
machinery and equipment – where the future cost of the asset is known). In other words,
looking at our formula, the accrued amount (A), the rate of interest (r) and the number of time
periods (n) are known with certainty and the uncertain variable this time is the original
principal (P).
Consider the following example:
A manufacturing company is considering replacing its production line in three years' time.
The cost of this is expected to total £40,000. Calculate how much must be invested now in a
bank account to purchase the new production line in three years' time, if the interest rate is
12% per annum. (Interest earned each year will be reinvested and no money will be
withdrawn).
Using our compound interest formula:
A =
n
100
r
1 P |
.
|

\
|
+
We know that:
P = £40,000
r = 12.0%
n = 3
Substituting these values in the formula gives:
40,000 =
3
100
12
1 P |
.
|

\
|
+
To solve for P, we need to rearrange our compound interest formula. You will recall from
Study Unit 2 Section D how we go about rearranging formulae.
Thus:
A =
n
100
r
1 P |
.
|

\
|
+ becomes:
P =
n
|
.
|

\
|
+
100
r
1
A
So
40,000 =
3
100
12
1 P |
.
|

\
|
+ becomes:
P =
3
100
12
1
40,000
|
.
|

\
|
+
=
3
1.12
40,000
=
.404928 1
40,000
= 8,471.21 2 £ (to the nearest p)
Thus, the manufacturing company needs to invest £28,471.21 today at an interest rate of
12% per annum if it is to accrue the £40,000 required to replace its production line in three
year's time.
Questions for Practice 4
1. £6,000 is invested at 10% per annum compound interest which is paid annually. How
much is in the account after 3 years?
2. £2,000 is invested at 10% per annum payable every 6 months. How much is in the
account at the end of 1½ years?
3. One building society offers an interest rate of 8.5% per annum payable half-yearly. A
second offers 8.75% payable annually. If you had £2,000 to invest, which savings
account should you choose?
4. An office has to replace some of its machinery in three years' time which it anticipates
will cost £9,000. Find out how much must be invested now, to buy this machinery in
three years' time, if the interest rate which can be earned on investments is 10%.
(Interest earned each year will be reinvested and no money will be withdrawn). Give
Now check your answers with those given at the end of the unit.
D. DEPRECIATION
Depreciation is a measure of the wearing out, consumption or other loss of value of an asset
arising from use, passage of time, or obsolescence through technology or market changes.
For example, a car loses value over time and is worth less each year.
This is an important concept in business where certain types of asset need to be accurately
valued in order to help determine the overall value of the business. The amount by which an
asset depreciates each year may also have important consequences.
There are several methods for calculating depreciation, but the most commonly used are the
straight-line method (using fixed instalments) and the reducing-balance method (applying a
fixed percentage). It is acceptable to use different methods for different categories of asset,
as long as the method chosen is applied consistently from year to year.
Straight-Line Depreciation
In the straight-line method, depreciation is calculated by dividing the cost of the asset by the
number of years the asset is expected to be used in the business. We can express this as a
formula as follows:
annual depreciation =
life useful
asset of cost
Thus, over the period of years selected, the value of the asset will be reduced to zero.
In the case of certain types of asset, such as cars, it may be anticipated that the asset will be
used for say three years and then sold. In this case, the amount of depreciation is the
difference between the cost of the asset and its anticipated value at the end of its useful life,
divided by the number of years over which it will be used. The formula would then be:
annual depreciation =
( ) ( )
life useful
life useful of end at value - asset of cost
Consider the following examples.
Example 1:
A computer costing £1,500 will have a useful life of 3 years. Show the process of
depreciation.
annual depreciation =
3
500 , 1 £
= £500
end of year 1: value = £1,500 ÷ £500 = £1,000
end of year 2: value = £1,000 ÷ £500 = £500
end of year 3: value = £500 ÷ £500 = £0
Example 2:
A machine cost £4,800. How much will it be worth after 3 years if it depreciates at £600 per
year, and how many years will it be before the machine is valueless?
Depreciation after 3 years will be 3 × £600 = £1,800
The value of the machine will be £4,800 ÷ £1,800 = £3,000
If the annual depreciation is £600, the period over which the machine is depreciated to zero
will be:
£4,800 ÷ £600 = 8 years.
Reducing Balance Depreciation
In the reducing-balance method, depreciation is calculated each year by applying a fixed
percentage to the diminishing balance. Unlike the straight-line method then, the total
provision for depreciation will never equal the cost price at the end of the asset's useful life
(i.e. the final balance never reaches zero).
The percentage to be applied will depend on the type of asset. For example, assets having
long lives, such as buildings, will have a small percentage (say 2½%), but assets which
quickly depreciate, such as motor vehicles, will bear a large percentage (say 15%).
Consider the following example.
A company bought a machine for £6000. If it depreciates at 12% per annum what will it be
worth after 3 years?
Amount of depreciation in year 1 = 12% of £6,000
= £6,000 ×
100
12
= £6,000 × 0.12 = £720
Therefore, value of machine at the end of year 1
= £6,000 ÷ £720 = £5,280
Amount of depreciation in year 2 = 12% of £5,280
= £5,280 ×
100
12
= £5,280 × 0.12 = £633.60
Therefore, value of machine at the end of year 2
= £5,280 ÷ £633.60 = £4,646.40
Amount of depreciation in year 3 = 12% of £4,646.40
= £4,646.40 ×
100
12
= £4,646.40 × 0.12 = £557.57
Therefore, value of machine at the end of year 3
= £4,646.40 ÷ £557.57 = £4,088.83
However this is obviously a time-consuming method. A quicker and simpler way is to realise
that 12% is 12/100, which is 0.12. If you are reducing the amount by a factor of 0.12 (12%),
you are left with 0.88 (88%) of the amount. So, multiplying the amount by 0.88 gives the
same result as multiplying the amount by 0.12 and then subtracting this from the original
amount.
We can demonstrate this in respect of the above example as follows:
Amount of depreciation in year 1 = 12% of £6,000
= £6,000 ×
100
12
= £6,000 × 0.12 = £720
Therefore, value of machine at the end of year 1
= £6,000 ÷ £720 = £5,280
This is the same as:
value of machine at the end of year 1 = £6,000 × 88%
= £6,000 × 0.88 = £5,280
This is both quicker and easier and you are less likely to make a mistake. Repeatedly
multiplying by 0.88 will give the amount left each year. Thus:
At the end of the year 2,
the value of the machine = £5,280 x 0.88 = £4,646.40
At the end of the year 3,
the value of the machine = £4,646.40 × 0.88 = £4,088.83
We can express this method by the following formula:
D = B(1 ÷ i)
n
where: D = depreciated value at the end of the n
th
time period
B = original value at beginning of time period
i = depreciation rate (as a proportion)
n = number of time periods (normally years).
Questions for Practice 5
1. A launderette buys a washing machine which cost £2,000.
(a) How much will it be worth after 2 years if it depreciates at £400 per year?
(b) How many years before the machine is valueless?
2. A factory buys a machine for £7,000. If the machine will be valueless after 4 years,
what should be the amount of depreciation per year?
3. A small business buys a computer costing £4,500. The rate of depreciation is 20% per
annum. What is its value after 3 years?
4. A second-hand motorbike is bought for £795. Its value depreciates by 11% per year.
For how much could it be sold two years later? (Give your answer to the nearest
pound.)
5. Company A bought a fleet of 20 cars for its sales force, each costing £15,250. It
decided to write-off equal instalments of the capital cost over 5 years. Company B
bought an exactly similar fleet, but it decided to write-off 20% of the value each year.
Calculate the value of the cars to Company A and to Company B at the end of the 3
rd
year.
Now check your answers with those given at the end of the unit.
E. PAY AND TAXATION
Pay and taxation are complex issues and it is not our intention here to explore them in any
detail. Rather, we shall concentrate on the principles of their calculation, using only a basic
construction of how pay is made up and the way in which the taxation system works.
Again, we shall be applying a number of the basic numerical concepts we have considered
previously – including the distinction between fixed and variable values, percentages and
simple formulae. This is also an area where it is very important to adopt clear layout when
working through calculations.
Make Up of Pay
All employees receive payment for their labour. At its most basic, this is paid in the form of
either a salary or wages.
Salaries are generally paid monthly and are based on a fixed annual amount. Thus:
monthly pay =
12
salary annual
For example, an employee with an annual salary of £18,000 would receive £1,500 every
month (£18,000 ÷ 12).
Wages are generally paid weekly and are usually based on a fixed rate for working a certain
number of hours. For example, many wage earners are required to work a fixed number of
hours in a week, and they are paid for these hours at a basic hourly rate.
Thus, an office clerk is paid £5.30 per hour for a 40 hour week. What is her weekly wage?
weekly wage = rate of pay × hours worked
= £5.30 × 40 = £212
An employee can often increase a basic wage by working longer than the basic week – i.e.
by doing overtime. A higher hourly rate is usually paid for these additional hours, the most
common rates being time and a half (i.e. 1½ times the basic hourly rate) and double time
(twice the basic hourly rate).
Consider a factory worker who works a basic 36 hour week for which she is paid a basic rate
of £5.84 per hour. In addition, she works 5 hours overtime at time and a half and 3 hours
overtime at double time. Calculate her weekly wage.
Basic pay: = 36 hours × £5.84 = £210.24
Overtime at time and a half: = 5 hours × (£5.84 × 1.5) = £43.80
Overtime at double time: = 3 hours × (£5.84 × 2) = £35.04
Total pay: = £289.08
People who are employed as salespersons or representatives, and some shop assistants,
are often paid a basic salary or wage plus a percentage of the value of the goods they have
sold. In some cases, their basic salary or wage can be quite small, or even non-existent, and
the commission on their sales forms the largest part of their pay.
For example, a salesman earns a basic salary of £8,280 per year plus a monthly commission
of 5% on all sales over £5,000. What would be his income for a month in which he sold
goods to the value of £9,400?
basic monthly pay =
12
salary annual
=
12
280 , 8 £
= £690
He earns commission on (£9,400 ÷ £5,000) worth of sales:
commission = 5% of (£9,400 ÷ £5,000) = 0.05 × £4,400 = £220
Total monthly pay = basic pay + commission
= £690 + £220 = £910
Questions for Practice 6
1. Calculate the weekly wage of a factory worker if she works 42 hours per week at a rate
of pay of £5.25 per hour
2. If an employee's annual salary is £18,750 how much is she paid per month?
3. A basic working week is 36 hours and the weekly wage is £306. What is the basic
hourly rate?
4. A trainee mechanic works a basic 37.5 hour week at an hourly rate of £4.90. He is paid
overtime at time and half. How much does he earn in a week in which he does 9 hours
overtime?
5. An employee's basic wage is £6.20 per hour, and she works a basic 5-day, 40-hour
week. If she works overtime during the week, she is paid at time and a half. Overtime
worked at the weekend is paid at double time. Calculate her wage for a week when she
worked five hours overtime during the week and four hours overtime on Saturday.
6. An insurance representative is paid purely on a commission basis. Commission is paid
at 8% on all insurance sold up to a value of £4,000 per week. If the value of insurance
sold exceeds £4,000, he is paid a commission of 18% on the excess. Calculate his
total pay for the 4 weeks in which his sales were £3,900, £4,500, £5,100 and £2,700.
Now check your answers with those given at the end of the unit.
Taxation
The total earnings made up in the way just discussed is known as gross pay. Unfortunately,
very few employees actually receive all the money they have earned. Certain amounts of
money are deducted from gross pay and the pay the employee actually ends up with is the
net pay or take-home pay.
The two main deductions are:
 income tax and
 pension contributions.
There are often other deductions – some being further national or local taxes on pay (such as
the UK system of National Insurance) and others being deductions made by the employer for
work-specific reasons (such as repayments of a loan for a car or for a public transport
season ticket).
Income tax is a major source of finance for the government and is used to fund all types of
public expenditure. The tax is a percentage of gross pay, based on the amount a person
earns in a tax year that begins on 6
th
April and ends on 5
th
April the following year. The
percentage rate at which the tax is applied varies according to the level of income. The
amount of gross income on which tax is payable (known as taxable pay) may be affected by
the provision of certain allowances within the tax system and also by certain elements of pay
or deductions being exempted from tax (non-taxable items). The rates of tax applied to
particular bands of income, and the levels of allowances, are decided each year by the
government as part of the Budget.
Most people have income tax deducted from their pay by their employer before they receive
it; the employer then pays the tax to the government. This method of paying income tax is
called PAYE (Pay As You Earn).
FromApril 2008, the UK system divides taxable pay into two bands of income, with the
following rates of tax applied to each:
Table 3.1: Taxation bands
Annual taxable income Tax rate
0 – £36,000 20%
£36,001 and above 40%
This means that, for a person with a total taxable pay of £20,000 in a year, the total tax
liability would be:
£20,000 × 20% = £4,000
For a person with a total taxable pay of £45,000 in a year, the total tax liability would be:
£36,000 × 20% £7,200
(£45,000 ÷ £36,000) × 40% = £3,600
£10,800
These annual tax bands can be converted into monthly or weekly amounts for calculating tax
on a PAYE basis and deducting the correct amount from an employees gross pay:
Table 3.2: Annual, monthly and weekly taxable income
Tax rate Annual taxable
income
Monthly taxable
income
Weekly taxable
income
20% 0 – £36,000 0 – £3,000 0 – £692
40% £36,001 and above £3,001 and above £693 and above
There are many possible allowances and non-taxable items which affect the level of taxable
pay. Here we shall consider just two.
 Personal allowances – the amount of money an individual is allowed to earn each year
before he or she starts to pay tax. This amount (sometimes called free-pay) depends
on the individual's circumstances – for example, whether they are single or married,
number of children, etc. If your personal allowances are greater than your actual pay,
then you would pay no income tax at all.
The level of personal allowances are generally set as an annual amount, but this can
be converted into monthly or weekly amounts of "free-pay".
 Pension contributions – the amounts paid by an individual into a pension scheme, such
as an employer's occupational pension scheme, are not taxable. Therefore, pension
contributions, where they are deducted from pay by the employer, may also be
deducted from gross pay before calculating tax. (Note that this does not necessarily
apply to any other types of deductions from pay. Where pension contributions are not
paid through the employer, an adjustment may be made to a person's personal
allowances to allow for them.)
Now we can develop a framework for calculating the amount of tax payable annually, using
the tax bands set out here.
gross taxable pay = gross pay ÷ non-taxable items
net taxable pay = gross taxable pay ÷ personal allowances
tax liability = first £36,000 of net taxable pay × 20%
plus remainder of net taxable pay × 40%
Consider the following examples, using the bands and tax rates set out here.
Example 1:
How much annual tax will be paid by someone who earns £18,600 per annum and has
personal allowances of £5,800? What is the monthly tax deduction?
There are no non-taxable items specified, so:
gross taxable pay = gross pay ÷ non-taxable items
= £18,600
net taxable pay = gross taxable pay ÷ personal allowances
= £18,600 ÷ £5,800 = £12,800
tax liability = £12,800 × 20% = £2,560
The monthly income tax deducted will be £2,560 divided by 12, i.e. £213 rounded down.
Example 2:
How much annual tax will be paid by someone who earns £43,000 per annum, has personal
allowances of £4,400 and pays pension contributions through his employer at a rate of 5% of
his annual salary.
Here, the pension contributions are non-taxable, so:
gross taxable pay = £43,000 ÷ (£43,000 × 5%) = £40,850
net taxable pay = £40,850 ÷ £4,400 = £36,450
tax liability = £36,000 × 20% + (£36,450 ÷ £36,000) × 40%
= £7,200 + £180 = £7,380
Note that taxation questions are often set without reference to income bands with different
taxation rates applying, or without reference to deductions from gross pay or personal
allowances. Always answer the question using the information supplied. For example:
How much tax will be paid each month by someone who earns £24,000 per annum if
the tax rate is 25%?
Net taxable pay per month, based on the information supplied =
12
000 , 24 £
= £2,000
Tax liability = £2,000 × 25% = £500 per month.
Questions for Practice 7
1. An employee is paid a salary of £24,492 per annum which he receives in 12 monthly
payments. He has personal allowances amounting to £4,800. If tax is payable at the
rate of 24%, calculate his monthly pay.
2. If, in the next Budget, the Chancellor changes the rate of income tax from 24% to 23%,
but wages remain the same, how much less tax will be paid by the employee in
Question 1 above.
3. Someone whose personal tax allowance is £4,800 is paid a weekly wage of £80. Would
this person have to pay tax?
4. An employee has a gross income of £23,000 and personal allowances of £5,000. She
pays £3,600 tax. What is the rate of income tax?
5. A weekly paid worker receives a basic wage of £5.50 per hour for a 37.5 hour week.
Overtime is paid at a rate of time and a half for the first five hours, and at double time
She also belongs to an occupational pension scheme, making contributions at a rate of
6% of basic earnings only. As a single mother, she has a personal taxation allowance of
£5,850.
Calculate her net pay in a week when she works 48 hours if the tax bands and rates
are as set out above (i.e. as per 2008 in the UK).
Now check your answers with those given at the end of the unit.
Questions for Practice 1
1. £45.26
2. (a) £209.30
(b) 105p (rounded).
3. (a) 1640 Swiss francs
(b) 590.64 euros.
4. (a) 945 Swedish kroner
(b) £5.97
Questions for Practice 2
1. £31
2. £5,950
3. £1,560
4. £24.72
5. £304
6. There are three parts to this question.
(a) They would be charged £3 for each cash transaction. Therefore, they would be
changing:
£107 into NZD = 299.39 NZD
£97 into AUD = 238.43 AUD
£117 into HKD = 2299.05 HKD.
(b) (i) £570
(ii) £5.70
(c) Total cost is the £900 worth of foreign currency plus the commission of £9 for the
cash transactions and £5.70 for the travellers cheques = £914.70. Spread
between the four members of the family, cost per person is £228.68.
Questions for Practice 3
1. £65.63
2. £2,295 interest and £9,045 in the account.
3. £231.25
4. 7.5%
Questions for Practice 4
1. £7,986
2. £2,315.25
3. To compare the effects of the two rates, work out the accrued amount after one year in
each case:
For the first building society, the rate of 8.5% payable six monthly equates to 4.25% as
a six-monthly rate. Thus, using the formula for calculating the accrued amount over two
six month periods is:
A =
n
100
r
1 P |
.
|

\
|
+
= £2,000
2
100
25 . 4
1 |
.
|

\
|
+ = £2,000 × 1.0425
2
= £2,000 × 1.1659134
= £2,173.61 (to the nearest p)
The calculation for the second building society rate of 8.75% paid annually is:
A = £2,000
1
100
75 . 8
1 |
.
|

\
|
+ = £2,000 × 1.0875
= £2,175.00
Therefore, the second building society, with the rate of 8.75% paid annually, is the
better investment.
4. Using the formula:
P =
n
|
.
|

\
|
+
100
r
1
A
=
3
100
10
1
9,000
|
.
|

\
|
+
=
( )
3
.10 1
9,000
=
1.331
9,000
= £6,762
Therefore, £6,762 would need to be invested today at an interest rate of 10% per
annum, if the office is to accrue the £9,000 required to replace some of its machinery in
three year's time.
Questions for Practice 5
1. (a) £1,200
(b) 5 years
2. £1,750
3. The value at the end of each year is 80% of the value at the beginning of the year.
Thus:
value at the end of year 1 = £4,500 × 80% = £3,600
value at the end of year 2 = £3,600 × 80% = £2,880
value at the end of year 3 = £2,880 × 80% = £2,304.
4. The value at the end of each year is 89% of the value at the beginning of the year.
Thus:
value at the end of year 1 = £795 × 89% = £707.55
value at the end of year 2 = £707.55 × 89% = £629.72 = £630 to the nearest
pound.
5. Total cost of the fleets to both companies = £15,250 × 20 = £305,000.
Company A:
annual amount of depreciation =
5
000 , 305 £
= £61,000
after 3 years, total depreciation = £61,000 × 3 = £183,000
value of the fleet after 3 years = £305,000 ÷ £183,000 = £122,000
Company B:
The value at the end of each year is 80% of the value at the beginning of the
year. Thus:
value at the end of year 1 = £305,000 × 80% = £244,000
value at the end of year 2 = £244,000 × 80% = £195,200
value at the end of year 3 = £195,200 × 80% = £156,160.
Questions for Practice 6
1. £220.50
2. £1,562.50
3. £8.50
4. £249.90
5. £344.10
6. £1,456
Questions for Practice 7
1. £1,247.16 per month
2. £16.41 per month
3. No
4. 20%
5. Gross pay is made up as follows:
37.5 hours at the basic rate of £5.50 per hour £206.25
5 hours at (£5.50 × 1.5) per hour £41.25
5.5 hours at (£5.50 × 2) per hour £60.50
gross pay £308.00
Non-taxable items are pension contributions amounting to:
£206.25 × 6% = £12.38
Gross taxable pay = gross pay ÷ non-taxable items
= £308.00 ÷ £12.38
= £295.62
Net taxable pay = gross taxable pay ÷ weekly personal allowances
= £295.62 ÷
52
850 , 5 £
= £295.62 ÷ £54.81
= £240.81
Tax liability is based on the weekly income tax bands, and this level of weekly net
taxable income falls within the 20% liability band (0 – £692).
tax liability = £240.81 × 20%
= £48.16
Net pay = gross pay ÷ pension contribution ÷ tax
= £308.00 ÷ £12.38 ÷ £48.16
= £247.46
89
Study Unit 4
Equations
Contents Page
Introduction 90
A. Simultaneous Equations 91
Solution by Elimination 91
B. Formulating Problems as Simultaneous Equations 96
Solving a Quadratic Equation by Factorisation 99
Solving Quadratic Equations with no Terms in x 102
Solving Quadratic Equations by Formula 103
D. Logarithmic and Exponential Functions 106
The Exponential Number e 106
Logarithms 106
Relationship between Logarithms and the Exponential Number 108
Laws of Logs and Exponents 108
Solving Equations using Logarithms and the Exponential Number 110
Answers to Questions for Practice 111
90 Simultaneous, Quadratic, Exponential and Logarithmic Equations
INTRODUCTION
In Study Unit 2 we introduced the concept of equations and looked at the basic principles of
solving them – i.e. finding out the value of the unknown quantity. In that study unit, we
considered simple equations where there was only one unknown quantity. Now we shall take
our study of equations further by examining equations where there is more than one
unknown quantity – simultaneous equations.
We shall also consider the distinction between linear and quadratic equations, a difference
which will be developed in the next study unit when we look at drawing graphs to represent
equations.
Finally, we will examine the principles and rules of logarithms and exponential values, and
learn how to solve equations using them.
Before we start though, we should recap the key principles already examined in relation to
equations. It is essential that you fully understand these before studying this unit, so if you
are unsure of any points, go back to Study Unit 2 and revise the section on equations there.
 An equation is a statement of equality between two expressions. It has two sides and
they must always be equal.
 Solving an equation means finding the value of the unknown quantity (or quantities) in
the equation – the variable – by using the other numbers, the constants. To do this, the
expressions which form the equation may be manipulated with the objective of isolating
the unknown on one side of the equation.
 The golden rule in manipulating equations is equality of treatment to both sides of the
equation. What you do to one side of the equation, you must also do to the other. Thus,
the same number may be added or subtracted from both sides of the equation without
changing its equality, and both sides may be multiplied or divided by the same number
without changing its equality.
 Quantities may be transposed – transferred from one side of the equation to the other –
by changing their signs. Thus, an added or subtracted term may be transposed if its
sign is changed from + to ÷, or from ÷ to +. Similarly, a multiplier may be transferred
from one side of an equation by changing it to the divisor on the other, and a divisor
may be transferred from one side by changing it to the multiplier on the other.
 Unknowns may also be transposed, using the same rules, in order to collect all the
unknown terms on the same side of the equation.
Objectives
When you have completed this study unit you will be able to:
 use the elimination method to solve simultaneous equations including two unknowns
 formulate problems as simultaneous equations
 define a quadratic equation and distinguish it from a linear equation
 solve quadratic equations by the factorisation method
 simplify and solve equations using both base 10 and natural logarithms.
Simultaneous, Quadratic, Exponential and Logarithmic Equations 91
A. SIMULTANEOUS EQUATIONS
As we said in the introduction, up until now, when solving equations, we had only one
unknown quantity to find. However, in practice, many situations will arise when there is more
than one unknown.
If an equation involves two unknown quantities, we may find any number of pairs of values to
satisfy it. Consider the following example:
x + y = 7
This may be satisfied by:
x = 3 and y = 4
x = 2½ and y = 4½
x = ÷1.3 and y = +8.3, etc.
To determine a particular pair of unknown values then, we need more information. One way
of getting more information is to have another equation – for example:
x + y = 7 and 2x + y = 12
Now there is only one solution pair: x = 5 and y = 2.
Similarly, if x + y = 7 and x ÷ y = 2, there is only one solution: x = 4½ and y = 2½.
Therefore to find two unknowns we need two equations; to find three unknowns we need
three equations, etc. These groups of equations are known as simultaneous equations.
Note though, that for equations to be simultaneous they must satisfy two conditions – they
must be consistent and independent. For example:
 the equations x + y = 7 and 2x + 2y = 5 are inconsistent – they cannot be true
simultaneously, and no pair of values which satisfies one will satisfy the other
 the equations 3x + 3y = 21 and 2x + 2y = 14 are not independent – they are really the
same equations, and all pairs of values which satisfy the one will satisfy the other.
Solution by Elimination
If two equations are true simultaneously, any other equation obtained from them will also be
true. The method of solution by elimination depends on this fact.
By adding or subtracting suitable multiples of the given equations, we can eliminate one of
the unknowns and obtain an equation containing only one unknown. Once we have done
that, the resulting equation can be solved using the methods we already know, and then the
other unknown may be discovered by substitution in the original equation.
The following examples explain the method. Study them carefully.
Example 1:
Consider the following pair of equations:
3x + y = 11 (1)
x ÷ y = ÷3 (2)
We can add the two equations together to derive a third equation which will eliminate y:
92 Simultaneous, Quadratic, Exponential and Logarithmic Equations
3x + y = 11
x ÷ y = ÷3
4x = 8
Thus, by adding the two equations, we are left with 4x = 8. Therefore, x = 2.
If we now substitute that value for x in (1), so that 3x becomes 6:
6 + y = 11
y = 5 (the symbol means "therefore").
We can check that the values we have determined for x and y are correct by substituting both
in equation (2):
x ÷ y = 2 ÷ 5 = ÷3
Note that eliminating one of the unknowns (in this case, y) was possible because both the
"y"s in the original two equations had a coefficient of 1 – i.e. the coefficients were equal. (The
coefficient is simply the factor or number applied to an algebraic term, so the coefficient of
the term 3x is 3, for example.)
If we want to eliminate one of the unknowns by adding or subtracting the two equations, the
coefficients of the unknowns must be equal.
Example 2:
Consider the following pair of equations:
6x + 5y = ÷6 (1)
18x + 7y = 6 (2)
We want one of the unknowns to have equal coefficients, but unlike in Example 1, this is not
the case here. However, we can see that the coefficient of x in (2) is a multiple of the
coefficient of x in (1). To make the coefficients equal, we derive a changed version of the
same equation by multiplying (1) by 3 and then subtract (2) from this equation:
Multiplying (6x + 5y = ÷6) by 3 gives:
18x + 15y = ÷18
Note that, providing we apply the same proportionate change to each term in the equation
(by multiplying or dividing each by the same number), the equation has not been changed. It
is not a new, independent equation, but simply a different way of expressing the same
equation.
We can now subtract equation (2) from this:
18x + 15y = ÷18
18x ÷ 7y = 6
8y = ÷24
y = ÷3
To find x, we substitute the value of y = ÷3 in (1) so that 5y becomes ÷15:
6x + (÷15) = ÷6
By transposition, we get:
6x = ÷6 + 15 = 9
x = 1.5
Simultaneous, Quadratic, Exponential and Logarithmic Equations 93
It is always good practice to check the values by substitution in the other equation, but we will
skip that step from now on.
Example 3:
Consider the following pair of equations:
13x ÷ 7y = 33 (1)
9x + 2y = 16 (2)
This time, there are no equal or multiple coefficients, so there is no obvious unknown term
which can be easily eliminated. We can then choose which unknown to eliminate and, to
avoid large numerical terms, we shall decide to eliminate y.
Where there are no equal or multiple coefficients, we need to change the way in which both
equations are expressed so that we can make the coefficients of one of the unknowns equal.
To do this here, we multiply (1) by 2 and (2) by 7, so that the coefficient of y in both equations
is 14. We can then eliminate y by adding the two equations:
26x ÷ 14y = 66
63x + 14y = 112
89x = 178
x = 2
To find y, we substitute the value of x = 2 in (2) so that 9x becomes 18:
18 + 2y = 16
By transposition, we get:
2y = 16 ÷ 18 = ÷2
y = ÷1
Important note: When working through the processes of transposition and substitution, it is
not necessary to write ÷ (÷7) and + (÷15), etc., but be sure to pay special
attention to all calculations involving negative numbers.
Questions for Practice 1
Solve the following simultaneous equations by finding values for x and y.
1. 3x + 4y = 1
9x + 5y = 10
2. 5x ÷ 3y = 14
7x + 5y = 38
3. 3x ÷ y = 8
x + 5y = 4
4. 2x ÷ 11y = 23
7x ÷ 8y = 50
94 Simultaneous, Quadratic, Exponential and Logarithmic Equations
5. 9x + 10y = 12
2x + 15y = ÷5
6. x +
3
2
y = 16
2
1
x + 2y = 14
7.
2
1
1 x ÷
5
3
y = 12
x +
15
7
y = 21
Now check your answers with those given at the end of the unit.
Now we shall look at some harder examples!
(a) Where the unknowns are denominators
Consider the following equations:
x
4
÷
y
3
= 4 (1)
x
3

y
7
= 2 (2)
Here we can treat
x
1
and
y
1
as the unknowns and then solve the problem in the usual
way.
To eliminate
y
1
, multiply (1) by 7 and (2) by 3 and then subtract:
x
28
÷
y
21
= 28 (1)
x
9
÷
y
21
= 6 (2)
x
19
= 22

x
1
=
19
22
x =
22
19
Substituting for
x
1
in (1) gives:
19
22 4×
÷
y
3
= 4
19
88
÷ 4 =
y
3
Simultaneous, Quadratic, Exponential and Logarithmic Equations 95
y
3
=
19
76 88 ÷
=
19
12
y
1
=
19
4
y =
4
19
=
4
3
4
Alternatively, you may find it easier to put
x
1
= m and
y
1
= n, and rewrite the equations as:
4m ÷ 3n = 4 (1)
3m ÷ 7n = 2 (2)
Solve these in the usual way, but remember to write your answer as x = ? and y = ?,
not as m and n.
(b) Forming simultaneous equations from single equations
Equations are sometimes given in the form:
x + 5y = 2x ÷ 4y + 3 = 3x + 2y ÷ 4
In order to solve this, we need to reformulate the equation as two simultaneous
equations.
Taking the first pair of expressions, we can rearrange the terms to form one equation in
the usual form:
x + 5y = 2x ÷ 4y + 3
÷x + 9y = 3
Now take the second pair of expressions and rearrange them:
2x ÷ 4y + 3 = 3x + 2y ÷ 4
2x ÷ 3y = 4
We now have two simultaneous equations as follows:
÷x + 9y = 3 (1)
2x ÷ 3y = 4 (2)
These can be solved in the usual way. To eliminate y, multiply (2) by 3 and add to (1):
5x = 15
x = 3
y =
3
2
96 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Questions for Practice 2
Solve the following equations for x and y:
x
3
+
y
20
= 13
x
7
+
y
10
= 2
Now check your answers with those given at the end of the unit.
B. FORMULATING PROBLEMS AS SIMULTANEOUS
EQUATIONS
As we have seen previously, it is very useful to develop mathematical models of real
situations to help analyse and solve problems. The use of algebraic notation to represent
variables in a situation and their combination with numerical constants to form equations are
the building blocks of such mathematical models.
We have seen that simple equations can be developed to model a situation and enable us to
find the value of one unknown variable, where we know the values of all the other variables.
In the same way, we can develop models which will enable us to find two or more unknown
variables by using simultaneous equations.
The key points about formulating problems as simultaneous equations are the same as doing
the same with simple equations:
 Always state clearly what your unknowns represent and give the units you are using.
 Having found the number for which the unknowns stand, give the answer to the
question in words.
Consider the following examples and try to form the equations for yourself before looking at
the solutions given. Note too, that unless you are told to find both or all of the unknowns, it is
not essential to do so, but it is most advisable, as otherwise you have no means of checking
Example 1:
The expenditure of 10 men and 8 boys amounts to £160. If 4 men together spend £18 more
than 6 boys, how much does each man and boy spend? (Assume each man spends the
same amount and each boy spends the same amount.)
Let: x = expenditure of 1 man in £
y = expenditure of 1 boy in £
Then we can express the information given as equations:
10x + 8y = 160 (all terms must be expressed in the same unit, i.e. here pounds)
and
4x ÷ 6y = 18
Solving these two equations will answer the problem.
Simultaneous, Quadratic, Exponential and Logarithmic Equations 97
Example 2:
In a bag containing black and white balls, half the number of white is equal to a third the
number of black, and twice the total number of balls exceeds three times the number of black
balls by four. How many balls did the bag contain?
Let: x = number of white balls
y = number of black balls.
Expressing the information given as equations, we get:
2
1
x =
3
1
y
and
2(x + y) ÷ 3y = 4
Example 3:
Pay special attention to the following problem. Note how the equations are built up.
A certain number, of two digits, is such that it is diminished by 9 if the digits are reversed. The
number itself is six times the sum of its digits. Find the number.
Let: x = the tens figure
y = the units figure
Then the number is 10x + y, and the reversed number is 10y + x.
Expressing the information given as equations, we get:
10x + y ÷ 9 = 10y + x (1)
and
10x + y = 6(x + y) (2)
Rearranging these equations to isolate the constant in (1) and simplify both, we get:
for (1):
9x ÷ 9y = 9
or
x ÷ y = 1 (A)
for (2):
4x = 5y (B)
If we now multiply (A) by 4, we get:
4x ÷ 4y = 4
We can then substitute the value of 4x from (B) to get:
5y ÷ 4y = 4
y = 4
Substituting for y in (A), we get:
x = 5
The required number is 54.
98 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Questions for Practice 3
1. Half the sum of two numbers is 20 and three times their difference is 18. Find the
numbers.
2. 6 kg of tea and 11 kg of sugar cost £13.30.
11 kg of tea and 6 kg of sugar cost £17.30.
Find the cost of tea and sugar per kilo.
3. The income from advertisements and sales for a college magazine amounted, in a
year, to £670. In the following year the income from advertisements was increased by
12½% and the income from sales decreased by 16
3
2
%. The total income decreased by
£12.50.
Find the original income from advertisements and sales using a calculation method.
4. A train from London to Birmingham takes 2 hours for the journey. If the average speed
of the train were decreased by 5 miles per hour, the train would take 12 minutes longer.
Find the distance from London to Birmingham and the average speed of the train,
using simultaneous equations.
Now check your answers with those given at the end of the unit.
All the equations we have studied so far have been of a form where all the variables are to
the power of 1 and there is no term where x and y are multiplied together – i.e. xy does not
feature in the equation. (The types of equation studied so far are known as linear equations
since, as we shall see in the next study unit, if we draw a graph of the outcome of the
equation for all the possible values of the unknowns, the result is a straight line.)
By contrast, a quadratic equation is one that contains the square of the unknown number, but
no higher power. For example, the following are quadratic expressions in x:
4x
2
+ 7x
2x
2
÷ x + 1
2
1
x
2
÷ 2
We can also have quadratic expressions in x
2
. For example, 3x
4
+ 2x
2
+ 5 is a quadratic in x
2
,
since it may be also be written as:
3(x
2
)
2
+ 2(x
2
) + 5, or
3y
2
+ 2y + 5, where y stands for x
2
Note that these are examples of quadratic expressions. A quadratic equation is obtained by
making such an expression equal to a similar expression or to a number.
One of the features of quadratic equations is that, usually, they cannot be solved definitively.
There are generally two possible values for the unknown.
Simultaneous, Quadratic, Exponential and Logarithmic Equations 99
A second feature, reflecting this, is that when we draw a graph of the outcome of a quadratic
equation for all the possible values of the unknown, the result is a curve. Again, we shall
examine this later.
Solving a Quadratic Equation by Factorisation
If an equation contains two unknown numbers, we cannot determine their values unless we
are told something more. For example, if x and y are two quantities and all we are told about
them is that their product is 12, we cannot tell either of their values – x may be 3, in which
case y must be 4; or x may be ÷½, in which case y must be ÷24, etc. This is true whatever
the value of the product, with the one exception that if xy = 0, then either x = 0 or y = 0.
If x = 0, y may have any value, and if y = 0, x may have any value. They need not both be 0,
but one of them must be, since the product of two numbers is zero only when one of them is
itself zero.
This important fact provides the easiest method of solving a quadratic equation, as is shown
in the following examples.
Example 1:
Consider the following equation:
x(x ÷ 3) = 0
Here, the product of the two factors x and (x ÷ 3) is zero. Therefore, one of the factors must
be zero:
either:
x = 0
or:
(x ÷ 3) = 0
further, if (x ÷ 3) = 0, then x = 3
so the equation is satisfied if x = 0, or if x = 3.
Having understood the principle of this procedure, we do not need to give it in full every time.
The following example simplifies the steps.
Example 2:
Consider the following equation:
x(x ÷ 5) = 0
Either, x = 0, or (x ÷ 5) = 0
x = 0, or x = 5
The important thing to remember is that this applies only to a product the value of which is
zero. Therefore, the above method of solving an equation applies only when the RHS (right-
hand side) of the equation is zero, while the LHS can be expressed in factors.
Example 3:
Consider the following equation:
(x ÷ 5)(x ÷ 2) = 4
We must rearrange this so that the RHS is zero. First of all, we have to expand the product of
the factors:
x
2
÷ 7x + 10 = 4
100 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Then we rearrange the equation so that the RHS is zero:
x
2
÷ 7x + 6 = 0
Next we have to re-express the LHS in factors:
(x ÷ 6)(x ÷ 1) = 0
Now we can say that either (x ÷ 6) = 0, or (x ÷ 1) = 0.
x = 6, or x = 1
Check this in the original equation just to justify to ourselves that the method works:
(x ÷ 5)(x ÷ 2) = 4
if x = 6:
(6 ÷ 5)(6 ÷ 2) = 1 × 4 = 4
if x = 1:
(1 ÷ 5)(1 ÷ 2) = (÷4)(÷1) = +4
Questions for Practice 4
Solve the following equations:
1. x(x ÷ 1) = 0
2. (x ÷ 1)(x + 2) = 0
3. 3x(x + 1) = 0
4. 4x
2
= 0
5. (2x ÷ 1)(x + 3) = 0
6. (3x ÷ 2)(2x + 3) = 0
Now check your answers with those given at the end of the unit.
You should have found these questions very simple and you may even have been able to
work them out in your head. However, no matter how simple an equation is, always copy it
down first exactly as it is set. Then, if you are sure of the working, you may give the answer
at once, but do not just guess.
On the other hand, you may prefer to omit the "either ... or ..." step. If you do this, remember
to change the sign of the number as you transfer it to the RHS of the equation. For example:
(4x ÷ 3)(2x + 5) = 0
4x = 3, or 2x = ÷5
x = ¾, or x = ÷2½
One of the reasons that the questions just given were so easy is that the LHS of the
equations were already expressed in factor terms. Commonly, though, equations are not so
simple – the factors may not already have been found, or the terms are not already collected
on one side of the equation. You will, therefore, have to work as follows:
Simultaneous, Quadratic, Exponential and Logarithmic Equations 101
 Collect all the terms on the LHS, so that the RHS is zero.
 Factorise the LHS expression – the product of the factors is then zero.
 Equate each factor in turn to zero, where possible.
Work through the following two examples carefully to make sure that you understand the
process thoroughly.
Example 4:
Consider the following equation:
7x
2
+ 23x = 60
First, collect all the terms on the LHS, so that the RHS is zero:
7x
2
+ 23x ÷ 60 = 0
Next factorise the LHS expression:
(7x ÷ 12)(x + 5) = 0
Then equate each factor in turn to zero:
either
(7x ÷ 12) = 0,
or
(x + 5) = 0
which may be expressed as:
7x = 12, or x = ÷5
x =
7
5
1 or ÷5
Example 5:
Solve x(5x ÷ 2) = 3(2x ÷ 1)
Expand the product of the factors:
5x
2
÷ 2x = 6x ÷ 3
Collect the terms:
5x
2
÷ 8x + 3 = 0
Factorise the LHS:
(5x ÷ 3)(x ÷ 1) = 0
5x = 3, or x = 1
x =
5
3
or 1
102 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Questions for Practice 5
Solve the following equations:
1. 7 ÷ x
2
= 5 ÷ x
2. x
2
+ 4x + 4 = 0
3. x
2
÷ 10x = 24
4. x(x ÷ 2) = 2(2 ÷ x)
5. x
2
÷ 11x + 30 = 0
6. 2x
2
+ 5 = 11x
7. 10x
2
+ 11x + 3 = 0
8. (x + 3)(x ÷ 3) = 8x
9. (x ÷ 1)
2
+ (x + 1)
2
= 34
10. (x + 1)
2
÷ (x ÷ 1)
2
= x
2
11. 9x
2
= 25
12. 12x
2
+ 12x = 9
Now check your answers with those given at the end of the unit.
Solving Quadratic Equations with no Terms in x
In particular circumstances, there is a quicker and easier method than the factor method of
Consider the following equation:
25x
2
÷ 36 = 0
Using the factor method, this would be solved as follows:
(5x ÷ 6)(5x + 6) = 0
5x = 6, or 5x = ÷6
x = 1.2 or ÷1.2.
There is, however, a simpler method of obtaining this answer and you should use it
whenever there is no term in x in the given equation:
25x
2
÷ 36 = 0
25x
2
= 36
x
2
=
25
36
Simultaneous, Quadratic, Exponential and Logarithmic Equations 103
To solve this, take the square root of each side, remembering that there are two square roots
numerically equal, but opposite in sign:
2
x =
25
36
x = ±
5
6
or ±1.2
Questions for Practice 6
Solve the following equations:
1. x
2
= 16
2. 4x
2
= 9
3. 3x
2
÷ 27 = 0
4. x
2
÷ a
2
= 0
5. 9x
2
= 1
Now check your answers with those given at the end of the unit.
Since the method just used for solving a quadratic equation applies to all cases, even where
no real roots can be found, it can be used to establish a formula.
Consider the equation ax
2
+ bx + c = 0, where a, b and c may have any numerical value.
Rearranging the equation we get:
ax
2
+ bx + c = 0
ax
2
+ bx = ÷c
x
2
+
a
b
x = ÷
a
c
At this point, we need to "complete the square", which means adding or subtracting the
expression that is needed in order to be able to factorise the LHS. In this case, it is:
2
a 2
b
|
.
|

\
|
Remember that this needs to be added to both sides of the equation to keep it balanced.
x
2
+
a
b
x +
2
a 2
b
|
.
|

\
|
=
a
c ÷
+
2
2
a 4
b
104 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Factorising the LHS we get:
2
2a
b
+ x |
.
|

\
|
=
2
2
4a
4ac - b
x +
2a
b
= ±
2
2
4a
ac 4 b ÷
x = ÷
2a
b
±
2a
ac 4 b
2
÷
x =
a 2
ac 4 b b
2
÷ ± ÷
This is an important formula. When using it, always write it down first, then write the values of
a, b, c in the particular case, paying great attention to the signs. It can be used to solve any
quadratic equation, although you should use the factor method whenever possible.
Example 1:
Consider the following equation:
5x
2
÷ 15x + 11 = 0
x =
a 2
ac 4 b b
2
÷ ± ÷
where: a = 5, b = ÷15, c = 11
Therefore:
x =
10
220 225 15 ÷ ± +
10
236 . 2 15±
=
= 1.72 or 1.28 (to two decimal places)
Example 2:
Consider the following equation:
3x
2
+ 4x ÷ 5 = 0
x =
a 2
ac 4 b b
2
÷ ± ÷
where: a = 3, b = 4, c = ÷5
Therefore:
x =
6
60 16 4 + ± ÷
=
6
718 . 8 4 ± ÷
=
6
718 . 4
or
6
718 . 12 ÷
= 0.79 or ÷2.12 (to two decimal places)
Simultaneous, Quadratic, Exponential and Logarithmic Equations 105
Example 3:
Consider the following equation:
3 ÷ 2x
2
= 4x
First arrange equation with x
2
term positive:
2x
2
+ 4x ÷ 3 = 0
Then apply the formula as usual.
Questions for Practice 7
Solve the following by means of the formula:
1. 2x
2
+ 13x + 4 = 0
2. 5x
2
÷ 3x ÷ 4 = 0
3. x
2
÷ 5x + 2 = 0
4. 4x
2
+ 7x + 1 = 0
5. x
2
÷ x ÷ 1 = 0
Now check your answers with those given at the end of the unit.
Remember that the square of a number is always positive, and that no real square root can
be found for a negative number.
Therefore, if the term b
2
÷ 4ac is negative, no real roots of the equation can be found.
For example:
2x
2
+ 4x + 5 = 0
Applying the formula:
x =
a 2
ac 4 b b
2
÷ ± ÷
where: a = 2, b = 4, c = 5
Therefore:
x
4
40 16 4 ÷ ± ÷
=
=
4
24 4 ÷ ± ÷
There are no real roots to this expression.
If b
2
÷ 4ac is a perfect square, then the square root will be a whole number and you should
have solved the equation by the factor method.
106 Simultaneous, Quadratic, Exponential and Logarithmic Equations
D. LOGARITHMIC AND EXPONENTIAL FUNCTIONS
So far in this study unit we have considered simultaneous and quadratic equations. In this
section we will now move on to consider logarithmic and exponential functions.
The Exponential Number e
There is a particular constant used in many mathematical calculations whose value (to seven
decimal places) is 2.7182818. This value occurs so frequently when mathematics is used to
model physical and economic phenomena that it is convenient to write it simply as e; it is
known as the exponential constant or exponential number.
Have a go on your scientific calculator now to obtain the value of e. To do so, simply locate
and press the e
x
button on your calculator and enter 1. (Calculators differ, so you may have
to experiment with your own calculator to find out how to obtain the value of e. Some
scientific calculators may require you to enter 1 before pressing the e
x
button.) By performing
this calculation, you have just demonstrated that:
e
1
= 2.7182818
Questions for Practice 8
Using your scientific calculator, find the following values:
1. e
0
2. e
3
3. e
-2
4. e
1/2
5. e
-3
Now check your answers with those given at the end of the unit.
Logarithms
A logarithm (also known simply as a log) is used to write an expression involving powers.
Having mastered the application of powers in Study Unit 2, you should have no problem
handling logarithms! In essence, a logarithm is the power to which a fixed number (base)
must be raised in order to produce a given number.
Let us work through some examples to aid our understanding of the application of logarithms.
It is common to work with logarithms using the number 10 as the base, referred to as
common logarithms.
Let us first consider the expression 10
2
. From our understanding of powers, 10
2
= 100 (i.e.
10 × 10 = 100). In other words, 10 is the base and 2 is the power or index. Using common
logarithms we can rewrite this expression as:
log
10
100 = 2
Simultaneous, Quadratic, Exponential and Logarithmic Equations 107
This statement can be read as the "log to the base 10 of 100 is 2". In other words, 2 is the
power to which the base number 10 must be raised in order to produce the number 100.
Thus, a logarithm is basically an alternative way of writing an expression involving powers.
Using this logic, then:
log
10
1000 = 3 (i.e. 3 is the power to which the base number 10 must be raised in order
to produce the number 1000).
log
10
10 = 1 (i.e. 1 is the power to which the base number 10 must be raised in order to
produce the number 10).
Using your scientific calculator, have a go at confirming the above results. To do so, simply
locate and press the log button on your calculator and enter the number you want to find the
common logarithm of. (Note – calculators differ, so you may have to experiment with your
own calculator to find out how to obtain the common log of a number. Some calculators may
require you to enter the number you want to find the logarithm of before pressing the log
button.)
Questions for Practice 9
Using your scientific calculator, find the logs of the following values:
1. log
10
50
2. log
10
45
3. log
10
2
4. log
10
0.4
5. log
10
4.4
Now check your answers with those given at the end of the unit.
Another base that we frequently use when working with logarithms (in addition to base 10) is
the exponential number e (i.e. the value of 2.7182818). Logarithms to the base e are
commonly referred to as natural logarithms and are denoted by ln or loge. For example:
ln(2) = 0.693
This statement can be read as the "log to the base e of 2 is 0.693" or "the natural log of 2 is
0.693". In other words, 0.693 is the power to which the base value of e (i.e. 2.7182818) must
be raised in order to produce the number 2.
Using your scientific calculator, have a go at confirming this result. To do so, simply locate
and press the ln button on your calculator and enter the number you want to find the natural
logarithm of. (Note – calculators differ, so you may have to experiment with your own to find
out how to obtain the natural log of a number. Some calculators may require you to enter the
number you want to find the natural logarithm of before pressing the ln button.)
108 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Questions for Practice 10
Using your scientific calculator, find the natural logarithms of the following values:
1. ln(1.76)
2. ln(0.3)
3. ln(2002)
4. ln(10)
5. ln(38)
Now check your answers with those given at the end of the unit.
Relationship between Logarithms and the Exponential Number
From this discussion of natural logarithms and the exponential number, you may have
identified that a relationship exists between the two. The natural logarithm function f(x) = e
x
is
the inverse of the exponential function f(y) = ln(x). Notice that the variables x and y have
been reversed. This is not a mistake! When we take the natural logarithm of an exponential
function the result is the exponent, so if y = e
x
, then ln(y) = x.
Use your calculator to confirm this natural logarithmic and exponential relationship. First,
think of a number for y and then calculate the natural logarithm of this number to find x (i.e.
using the formula y = e
x
). Then, using this value for x, calculate the value of y using the
exponential function y = e
x
. The resulting value of y should be the same as the initial number
you chose for y, thus demonstrating that an inverse relationship exists between the natural
logarithm function f(x) = e
x
and the exponential function f(y) = ln(x). Thus, we can conclude
that:
ln(e
x
) = x
Laws of Logs and Exponents
Numbers expressed in logarithmic or exponential form obey the same rules applicable to
powers and indices, which have already been described in Study Unit 2.
Here we give the various rules of logarithms (which can be equally applied to common
logarithms (i.e. to the base 10) and natural logarithms (to the base e)). Since logarithms are
basically powers, there should be some logic to the following rules. Thus, for any base:
- log(p × q) = log p + log q
- log |
.
|

\
|
q
p
= log p ÷ log q
- log p
n
= n log p
- if y = ax
n
then n = (log y ÷ log a) ÷ log x
- log 1 = 0
and finally, do not forget that:
- ln(e
x
) = x
Simultaneous, Quadratic, Exponential and Logarithmic Equations 109
Using these laws, we can simplify some expressions involving logarithms to a single log
term.
Example 1:
Simplify the following logarithm expression into a single log term:
log(x) + 2 log(y)
Using the rule [log p
n
= n log p] we can rearrange the second part of the expression:
log(x) + log(y
2
)
Using the rule [log(p × q) = log p + log q] we can rearrange the expression:
log(xy
2
)
Example 2:
Simplify the following logarithm expression into a single log term:
3ln(A) ÷ 4ln(B)
Using the rule [log p
n
= n log p] we can rearrange both parts of the expression:
ln(A
3
) – ln(B
4
)
Using the rule [log |
.
|

\
|
q
p
= log p ÷ log q] we can rearrange the expression:
ln |
.
|

\
|
4
3
B
A
Questions for Practice 11
Simplify the following logarithmic expressions to a single log term:
1. log(x) + log(x ÷ 4)
2. 4 log(2) + 2 log(3) ÷ log(12)
3. ln(x2y) ÷ ln(y) + 3 ln(x)
Now check your answers with those given at the end of the unit.
110 Simultaneous, Quadratic, Exponential and Logarithmic Equations
Solving Equations using Logarithms and the Exponential Number
Before completing this section on logarithms and exponents, let us bring together everything
we have learnt about them, as well as how to solve equations in general, by answering the
following questions.
Questions for Practice 12
Solve the following equations, giving your answer correct to 2 decimal places (you will need a
scientific calculator to evaluate logs and natural logs):
1. 4x = 100
2. e2x = 10
3. log 3x = 3
Now check your answers with those given at the end of the unit.
Simultaneous, Quadratic, Exponential and Logarithmic Equations 111
Questions for Practice 1
1. 3x + 4y = 1 (i)
9x + 5y = 10 (ii)
To eliminate x, multiply (i) by 3 so that the coefficients are equal and then subtract (ii):
9x + 12y = 3
9x + 5y = 10
7y = ÷7
y = ÷1
Substituting y = ÷1 in (i) we get:
3x ÷ 4 = 1
3x = 5
x =
3
2
1
(Check in (ii): 15 ÷ 5 = 10)
2. 5x ÷ 3y = 14 (i)
7x + 5y = 38 (ii)
To eliminate y, multiply (i) by 5 and (ii) by 3 so that the coefficients are equal and then
25x ÷ 15y = 70
21x + 15y = 114
46x = 184
x = 4
Substituting x = 4 in (i) we get:
20 ÷ 3y = 14
3y = 6
y = 2
(Check in (ii): 28 + 10 = 38)
The full workings for the next three answers are not given, but the method is indicated.
3. 3x ÷ y = 8 (i)
x + 5y = 4 (ii)
To eliminate y, multiply (i) by 5 and add the equations: 16x = 44, x =
4
3
2
or
to eliminate x, multiply (ii) by 3 and subtract (i): 6y = 4, y =
4
1
112 Simultaneous, Quadratic, Exponential and Logarithmic Equations
4. 2x ÷ 11y = 23 (i)
7x ÷ 8y = 50 (ii)
To eliminate y, subtract 8 times (i) from 11 times (ii): x = 6
or
to eliminate x, subtract 7 times (i) from 2 times (ii): y = ÷1
5. 9x + 10y = 12 (i)
2x + 15y = ÷5 (ii)
To eliminate y, subtract 2 times (ii) from 3 times (i): x = 2
or
to eliminate x, subtract 2 times (i) from 9 times (ii): y = ÷
5
3
6. x +
3
2
y = 16 (i)
2
1
x + 2y = 14 (ii)
To eliminate x, subtract 2 times (ii) from (i): 3
3
1
y = 12, 10y = 36, y =
10
6
3
substituting for y in (ii):
2
1
x +
5
6
6 = 14
2
1
x = 14 ÷
5
1
7
2
1
x =
5
4
6
x =
5
3
13
7.
2
1
1 x ÷
5
3
y = 12 (i)
x +
15
7
y = 21 (ii)
Notice that (i) has a factor 3, so divide by this to get a simpler equation:
2
1
x ÷
5
1
y = 4 (i)
To eliminate x, subtract 2 times (i) from (ii): (
5
2
15
7
+ )y = 13,
15
13
y = 13, y = 15
Substituting in (i) for y gives:
2
1
x ÷ 3 = 4
2
1
x = 7
x = 14
Simultaneous, Quadratic, Exponential and Logarithmic Equations 113
Questions for Practice 2
Let a
x
1
= , and b
y
1
= . Then express the equations using these terms:
3a + 20b = 13
7a ÷ 10b = 2
Solving these in the normal way gives:
a = 1, and b =
2
1
Therefore:
x = 1, and y = 2
Questions for Practice 3
1. Let: x = first number
y = second number
Then, if half the sum of two numbers is 20:
2
1
(x + y) = 20
or
x + y = 40 (i)
and, if three times their difference is 18:
3(x ÷ y) = 18
or
x ÷ y = 6 (ii)
Eliminating x by subtraction, we get:
2y = 34
y = 17
Substituting for y in (i), we get:
x + 17 = 40
x = 23
Therefore, the numbers are 17 and 23.
2. As the units must all be the same, change each amount of money to pence.
Let: x = cost of tea per kg in pence
y = cost of sugar per kg in pence.
Then
6x + 11y = 1,330 (i)
11x + 6y = 1,730 (ii)
114 Simultaneous, Quadratic, Exponential and Logarithmic Equations
To eliminate x, subtract 6 times (ii) from 11 times (i) to get:
85y = 4,250
y = 50
Substituting for y in (i) we get:
6x + 550 = 1,330
x =
6
550 1,330÷
x = 130
Therefore, the cost of tea = £1.30 per kilo, and the cost of sugar = £0.50 per kilo.
y = the original income from sales (in £).
Thus, in the first year:
x + y = 670 (i)
In the second year:
(x + 12
2
1
%x) + (y ÷ 16
3
2
%y) = 670 ÷ 12.50
(x +
8
1
x) + (y ÷
6
1
y) = 657
2
1
8
9
x +
6
5
y = 657
2
1
Multiplying this by 24 to eliminate the fractions, we get:
27x + 20y = 15,780 (ii)
To eliminate y from the pair of equations, multiply (i) by 20:
20x + 20y = 13,400
and subtract this from (ii) to get:
7x = 2,380
x = 340
Substituting for x in (i):
340 + y = 670
y = 330
Thus, the original income from:
sales = £330.
4. Let: x = the average speed (in miles per hour)
y = the distance (in miles)
We can use a formula for the time taken:
speed
distance
= time taken
Simultaneous, Quadratic, Exponential and Logarithmic Equations 115
So at average speed:
x
y
= 2 hours
and at the slower speed:
) 5 (x
y
÷
= 2 hours 12 minutes =
5
1
2 hours
Rearranging these equations into standard linear form, we get:
y = 2x (i)
y =
5
1
2 x ÷ 11 (ii)
Subtracting (ii) from (i) gives:
0 =
5
1
x ÷ 11
11 =
5
1
x
55 = x
x = 55
Substituting for x in (i), we get:
y = 110
Thus: average speed for the journey = 55 mph
distance from London to Birmingham = 110 miles.
Questions for Practice 4
1. x = 0 or 1
2. x = 1 or ÷2
3. x = 0 or ÷1
4. x = 0
5. x =
2
1
or ÷3
6. x =
3
2
or 1
2
1
Questions for Practice 5
1. x = 2 or ÷1
2. x = ÷2
3. x = 12 or ÷2
4. x = 2 or ÷2
5. x = 5 or 6
6. x =
2
1
or 5
7. x = ÷
5
3
or ÷
2
1
116 Simultaneous, Quadratic, Exponential and Logarithmic Equations
8. x = 9 or ÷1
9. x = 4 or ÷4 (written as x = ±4)
10. x = 0 or 4
11. x = ±
3
2
1
12. x = ÷1
2
1
or
2
1
Questions for Practice 6
1. x = ±4
2. x = ±1
2
1
3. x = ±3
4. x = ±a
5. x = ±
3
1
Questions for Practice 7
1. ÷0.32, ÷6.18
2. 1.24, ÷0.64
3. 4.56, 0.44
4. ÷0.16, ÷1.59
5. 1.62, ÷0.62
Questions for Practice 8
1. e
0
= 1
2. e
3
= 20.086
3. e
-2
= 0.135
4. e
1/2
= 1.649
5. e
-3
= 0.050
Questions for Practice 9
1. log 10 50 = 1.699
2. log 10 45 = 1.653
3. log 10 2 = 0.301
4. log 10 0.4 = -0.398
5. log 10 4.4 = 0.643
Simultaneous, Quadratic, Exponential and Logarithmic Equations 117
Questions for Practice 10
1. ln(1.76) = 0.5653
2. ln(0.3) = ÷1.204
3. ln(2002) = 7.602
4. ln(10) = 2.303
5. ln(38) = 3.638
Questions for Practice 11
1. Using the rule [log(p × q) = log p + log q]:
log(x) + log(x ÷ 4) = log(x (x ÷ 4))
= log(x
2
÷ 4x)
2. Using the rule [log p
n
= n log p]:
4 log(2) = log(2
4
) = log(16), and
2 log(3) = log(3
2
) = log(9)
Using the rules [log(p × q) = log p + log q] and [log |
.
|

\
|
q
p
= log p ÷ log q]:
4 log(2) + 2 log(3) ÷ log(12) = log(16) + log(9) ÷ log(12)
= log
|
.
|

\
| ×
12
9 16
= log(12)
3. Using the rules [log(p × q) = log p + log q] and [log p
n
= n log p]:
ln(x
2
y) ÷ ln(y) + 3ln(x) = ln(x
2
) + ln(y) ÷ ln(y) + ln(x
3
)
= ln(x
5
)
Questions for Practice 12
1. 4
x
= 10, 4 log 100 log x ÷ = , 32 . 3 x =
2. e
2x
= 10, ) 10 ( In x 2 = , 2 10 In x ÷ = 15 1 x · =
3. log 3x = 3, 3x = 10
3
, x = 10
3
÷3 x = 333.33
118 Simultaneous, Quadratic, Exponential and Logarithmic Equations
119
Study Unit 5
Introduction to Graphs
Contents Page
Introduction 120
A. Basic Principles of Graphs 121
Framework of a Graph 121
Drawing a Graph 122
B. Graphing the Functions of a Variable 124
Representation 125
Ordered Pairs 126
C. Graphs of Equations 126
Linear Equations 127
Equations with x as the Denominator 136
Exponential and Logarithmic Equations 139
Gradient of a Straight Line 142
Equation of a Straight Line and Gradient 145
E. Using Graphs in Business Problems 147
Break-even Analysis 151
Economic Ordering Quantities 152
Plotting Time Series Data 155
Answers to Questions for Practice 160
120 Introduction to Graphs
INTRODUCTION
For many problems and issues which fall within the remit of quantitative methods, it is
possible to analyse and solve them both mathematically and graphically. This study unit
introduces the principles and role of graphs in such problems.
There are many different types of graph, but here we will only be concerned with linear
(straight line), quadratic (curved line), exponential and logarithmic graphs. Line graphs
illustrate the relationships between two variables, showing graphically what happens to one
variable when the other changes. For example, many graphs involve time as one of the
variables, and thus provide a measure of something that changes with time. Whenever one
measurable quantity changes as a result of another quantity changing, it is possible to draw
a graph.
You will immediately realise that since equations are a mathematical expression of the
relationship between two variables, it is possible to represent equations by means of graphs.
We shall spend some time on this in the unit.
The importance of graphs is that they give a visual picture of the behaviour of the two
variables. From this, by knowing how to interpret it, information can very quickly be found.
The picture may, for example, indicate a trend in the data from which we might be able to
make further predictions. The graph might also be used to approximate a solution to a
particular problem.
Graphical approaches to problem solving are often easier to carry out than mathematical
ones. However, they are less accurate, since the answers must be read from the graph and
hence cannot be more accurate than the thickness of the pencil. For this reason, it is
essential that you are very careful when drawing graphs, and we cover some of the key
design aspects here. One element of this is the need to use graph paper. You will find it very
useful to have a supply to hand as you work through this unit.
Objectives
When you have completed this study unit you will be able to:
 explain the construction of graphs
 calculate functions of x (that is, f(x))
 calculate and tabulate the values of y from an equation in x
 draw graphs from linear, quadratic, exponential and logarithmic equations
 identify the characteristics of a straight line in terms of its equation, and calculate its
gradient from its graph or its equation
 decide, from an equation, which way its graph will slope
 identify, from looking at their equations, whether two lines are parallel
 describe a shaded region on a graph in terms of the lines bounding it
 use graphs to solve a variety of problems.
Introduction to Graphs 121
A. BASIC PRINCIPLES OF GRAPHS
A graph is a diagram illustrating one or more mathematical relationships between two
variables. The following figures show the relationship between the height of a plant and the
time since it was planted. We shall use this to explain some of the basic features of graphs
and their construction.
We shall start by considering the framework of the graph.
Framework of a Graph
Figure 5.1: Sample graph framework
A graph has two axes along which the values of the variables are shown:
 The y-axis is the vertical axis and usually displays the dependent variable – i.e. the
variable which changes according to changes in the other variable. Here this is the
height of the plant (which changes in accordance with the time since planting).
 The x-axis is the horizontal axis and usually displays the independent variable – i.e. the
variable which causes change in the other variable. Here this is the time since planting
(which causes the height of the plant to increase).
The values of the variables are marked along each axis according to a scale (here, 0–9 on
the x-axis and 0–6 on the y-axis). Not all values need to be marked, but the dividing marks
do need to be equally spaced along the axis and the scale used must be clear to the reader.
The scale need not be the same on both axes.
Each axis should be titled to show clearly both what it represents and the units of
measurement.
The area within the axes (known as the plot area) is divided into a grid by reference to the
scale on the axes. This grid may actually be marked on the paper (as is the case with graph
paper, which is why it is much easier to draw graphs using graph paper) or may be invisible.
We have shown the main gridlines here as grey, dotted lines.
Particular points within the graph are identified by a pair of numbers called co-ordinates
which fix a point by reference to the scales along the axes. The first co-ordinate number
always refers to a value on the scale along the x-axis and the second to the scale on the y-
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9
Time since planting (days)
Height
of
plant
(cm)
y-axis
origin

co-ordinate (3,3)
co-ordinate (6,2)
x-axis
122 Introduction to Graphs
axis. We have shown two points on the framework in Figure 5.1 – make sure you understand
their referencing.
The point at which the two axes meet is called the origin. This has the co-ordinate reference
(0,0).
Finally, note that the x and y-axes may extend to include negative numbers. In this case, the
framework of the graph will look like that shown in Figure 5.2.
Figure 5.2: Graph framework including negative scales
Drawing a Graph
When drawing a graph, follow this procedure for constructing the framework.
 Determine the range of values which will need to be shown on each of the axes. You
should start by considering the x-axis, since the range of values for the independent
variable will determine the range of values needed for the dependent variable on the y-
axis.
 Draw the axes and mark off the scales along them. The aim is usually to draw as large
a graph as possible since this will make it easier to plot the co-ordinates accurately,
and also make it easier to read. Remember that the scales must be consistent along
each axis, but do not need to be the same on both. If you are using graph paper, use
the thick lines as your main divisions. If you are not using graph paper, use a ruler to
mark the scale and select easy-to-identify measurements for the divisions (such as one
cm for each main unit).
 Give each axis a title, number the main divisions along each scale and give the whole
graph a title.
Now you are ready to start plotting the graph itself.
Points on the graph are identified by their co-ordinates. These are the values given (or
determined) for firstly, the independent variable and, secondly, for the dependent variable.
Thus, you may be given the following data about the growth of a plant:
0
y-axis

x-axis

co-ordinate
(3,1)
3
2
1
3
2
1
1 2 3 1 2 3
co-ordinate (2,2)
co-ordinate (1,2)
Introduction to Graphs 123
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7 8 9
Time since planting (days)
Height
of
pl ant
(cm)
Height of plant (cm)
Day 1 1
Day 3 2
Day 5 3
Day 7 4
The points should be plotted with a small x using a sharp pencil.
To plot the co-ordinate (1,1) you count one unit across (along the x-axis) and then count one
unit up (along the y-axis). Be careful not to get these in the wrong order or when you come to
plot the co-ordinate (7,4) you will have plotted the point (4,7), which is somewhere
completely different!
(Remember that if the x co-ordinate is negative you count to the left and if the y co-ordinate
is negative you count downwards.)
Once you have plotted all the given points, you should join them all up.
If your graph is a linear graph, all the points will be in a straight line and you can join them
using a ruler. (In fact, to draw a linear graph, you only need to plot two points which, when
joined and extended across the whole scale of the x-axis, give the complete graph for all
values of the independent variable.)
However, if your graph is a quadratic graph, then the plot line will form a curve. The points
should be joined with a smooth curve. The easiest way of doing this is to trace through the
points with a pencil, but without touching the paper, to find the shape of the curve. When you
are satisfied that the path of the curve is smooth, draw the curve through the points in one
movement.
The full graph derived from the previous data on the height of a plant and extended over nine
days is given in Figure 5.3.
Figure 5.3: Graph of height of plant against time
Now that we have reviewed the basic principles, we shall move on to consider in detail the
way in which variables and equations are represented on graphs.
124 Introduction to Graphs
B. GRAPHING THE FUNCTIONS OF A VARIABLE
The statement that something is a "function of x" simply means that the value of that
something depends upon the value of a variable quantity, x. It is denoted by f(x). If y  f(x), y
will have a different value for each value of x, so that we can obtain a succession of values
for y by giving successive values to x.
For example, if y  x  3, then when x is 0, y will be equal to 3; when x  1, y  4; when x  2,
y  5; when x  3, y  6, and so on. For convenience, we usually tabulate our values for x
and y, as follows:
x 0 1 2 3 4 5
y  x  3 3 4 5 6 7 8
Let's look at another example. Suppose you are given the equation y  x
2
to find the values of y from x  2 to x  4.
You could start by making the following table:
x 2 1 0 1 2 3 4
x
2
4 1 0 1 4 9 16
3 3 3 3 3 3 3 3
y  x
2
 3 7 4 3 4 7 12 19
Alternatively, as this is a simple problem, you could tabulate just the first and the last lines,
We can also find a single function of x by substituting that value directly into the given
equation, for example:
f(x)  x
2
 2x
Replace x by 3 and find the value:
f(3)  3
2
 2  3  15
Replace x by 2 and find the value
f(2)  (2)
2
 2(2)  0
Introduction to Graphs 125
Questions for Practice 1
1. Complete the table for y  3x  2.
x 0 0.5 1 1.5 2 2.5
y  3x  2
2. Which one of the following functions could be represented by the following table?
y  x  4 y  x
2
 6x  16
y  2x  1 y  x
3
 1
x 3 5 7 9 11
y 7 11 15 19 23
3. If f(x)  5x  2 find:
(a) f(2)
(b) f(3).
4. If f(x)  x
2
 2 find:
(a) f(2)
(b) f(3)
Now check your answers with those given at the end of the unit.
Representation
From the tables we see that as the value of x changes, so does the value of y. One way of
showing how these values change is by means of a graph.
The values in the table can be expressed as a pair of co-ordinates which may be plotted on a
graph. The first figure is the value of x along the scale of the x (horizontal) axis, and the
second figure is the value of y along the scale of the y (vertical) axis. For example, in Figure
5.4, A is the point (2,1), B is (4,2), C is (3,1), D is (5,
2
1
), E is (1,0) and F is (0,1
2
1
).
126 Introduction to Graphs
Figure 5.4: Co-ordinates for y  f(x)
Ordered Pairs
Another way of describing the varying values of x and y is by means of a set of ordered pairs.
This is a very useful and helpful way of noting the points we required in the graphical
representation of a function.
The table in Question 2 of Questions for Practice 1 could be written as a set of ordered pairs
as follows:
(3,7) (5,11) (7,15) (9,19) (11,23)
These are then the points which we would plot if we were going to draw the graph of the
function in question.
When dealing with ordered pairs, you will normally be told the values of x with which you are
concerned. Thus, if we had to write down the ordered pairs of the function 2x  1 for integral
(whole number) values of x from 1 to 2 inclusive, we would write (1,1), (0,1), (1,3), (2,5);
and if the function was x
2
 3 for the same values of x it would be (1,4), (0,3), (1,4), (2,7).
C. GRAPHS OF EQUATIONS
In this section we shall consider graphs of linear, quadratic, exponential, logarithmic and
hyperbolic equations.
 Linear equations – equations of the first degree in x and y (i.e. containing no higher
power of x than the first, such as y  2x  3, y  5x  1). As the name implies, the graph
of expressions like these will always be a straight line.
 Quadratic equations – those of the second degree in x and y (i.e. containing the square
of the independent variable, but no higher power, such as y  2x
2
x  1, y  ½x
2
 3).
Such equations always give a curved graph with a characteristic "U" shape, called a
parabola.
 Hyperbolic equations – we shall look at graphs from the expression y 
x
a
where a is a
constant. Such graphs have a different curved shape – a hyperbola.

2 4 6 2 0
3
1
2
2
1
4 6

A
B
C
D
E
F
Introduction to Graphs 127
 Exponential and logarithmic equations – an exponential or logarithmic function occurs
when the independent variable appears as the exponent or as a logarithm. When
plotted, equations of both exponentials and logarithms produce non-linear lines.
Linear Equations
A linear equation is of the form y  mx  c, where m and c are constants. This will always
give a straight line. (Note that such an equation can be written in different forms, e.g. y  b 
ax.)
To draw a line we must know two points on it. It is even better if we know three points, using
the third as a check. So, the minimum number of points you ought to plot for a linear graph is
three.
If we choose any three values of x we can find the corresponding values of y and this will
give us the co-ordinates of three points which lie on the line.
Example 1:
Draw the graph of y  2x  3.
Suppose we choose x  0, x  2 and x  2.
Taking the given equation, y  2x  3:
when x  0: y  0  3  3
when x  2: y  (4  3)  1
when x  2: y  (4 3)  7
This is better tabulated as:
x 0 2 2
y  2x  3 3 1 7
We now have three points which we can plot and, on joining them, we have the graph of y 
2x  3.
Figure 5.5: Graph of y  2x – 3
-8
-6
-4
-2
0
2
4
-3 -2 -1 0 1 2 3
x
y
128 Introduction to Graphs
Example 2:
Try this one yourself. Draw the graph of y  3x  2, for values of x from 3 to 2.
First choose three values of x, say 3, 0 and 2. Then draw up a table of values – fill in the
spaces for yourself to find the co-ordinates of the points to be plotted.
x 3 0 2
y  3x  2
You should have found the points: (3, 7), (0, 2), (2, 8).
Now draw your two axes to cover the range of values for x specified (although it is good
practice to extend them beyond this – leaving some spare room at each end of each axis),
mark the scale, plot your points and join them up. Then compare your result with Figure 5.6.
Figure 5.6: Graph of y  2x + 2
If you look carefully at Figure 5.6 you will notice that the line y  2x  2 meets the y-axis at
the point where y  2. This is because the constant c in the general equation y  mx  c is
always the point where the line passes through the y-axis (i.e. where x  0). A further fact
which emerges is that when x goes from 0 to 2 (i.e. increases by two), y rises from 2 to 8 (i.e.
increases by four). The change in the value of y for a change of one in the value of x is given
in the equation y  mx  c by the constant m.
We have thus found that in the equation y  mx  c:
the constant c  the point where the line crosses the y-axis
the constant m  the slope of the line.
The point at which the line crosses an axis is called the intercept. The y intercept is where
the line crosses the y-axis and the x intercept is where the line crosses the x-axis.
-8
-6
-4
-2
0
2
4
6
8
-4 -2 0 2 4
x
y
Introduction to Graphs 129
Example 3:
Suppose we are asked to draw the graph of 3y  2x  1, from x  4 to 2.
Before we can work out the table of values in this case, we must change the equation to the
form y  mx  c. So we rewrite the equation, having divided both sides of it by 3:
3
1 2x
y

Now choose some values of x – let's say 4, 1, 2. Working these out gives us the following
table of values:
x 4 1 2
y 3 1 1
Now plot these points and join them. Note that, as in the title to Figure 5.7, this is still a graph
of the original equation.
Figure 5.7: Graph of 3y  2x + 1
-3
-1
1
3
-2 -1 0 1 2 3 4
x
y
130 Introduction to Graphs
Example 4:
We are not limited to having just one line plotted on a graph. We can draw two or more
graphs using the same axes – a very useful method of comparing different equations of the
same variables.
For example, we could draw the graphs of y 
3
1
x and y  8  x on the same axes.
Taking values of x between 4 and 10, we get the following tables of values:
For y 
3
1
x
x 3 0 3
y 1 0 1
For y  8  x x 2 4 10
y 10 4 2
The graphs are shown in Figure 5.8.
Figure 5.8: Graphs of y  ⅓ x and y  8 – x
You will notice that the lines intersect (i.e. cross) at A (6,2)
-2
0
2
4
6
8
10
-4 -2 0 2 4 6 8 10
x
y
A
Introduction to Graphs 131
Questions for Practice 2
1. Draw the graph of y  5x  4, from x  1 to x  2.
2. Draw the graph of y  2x  3, from x  4 to x  2.
3. Draw the graph of y  2  x, for values of x from 4 to 6.
4. On the same axes, plot the following pairs of points on squared paper, and draw the
line through them:
(i) (3,0) (0,6)
(ii) (1,0) (0,5)
(iii) (3,2) (2,3)
(iv) (4,3) (1,2)
Write down the co-ordinates of the points where (i) and (ii) each cut the lines (iii) and
(iv).
5. For values of x from 3 to 6, draw on the same axes the graphs of:
(i) y  5 (ii) x  5 (iii) y  5x (iv) y  
1
5
x
Describe briefly the resultant lines.
6. For values of x from 3 to 6, draw on the same axes the graphs of:
(i) y  2x (ii) y  2x  3 (iii) y  2x  4
Describe briefly the resultant lines and their intercepts.
7. For values of x from 3 to 6, draw on the same axes the graphs of:
(i) y  x  2 (ii) 3y  6  x
Identify the point of intersection.
Now check your answers with those given at the end of the unit.
Characteristics of Straight-Line Graphs
If you consider the graphs we have drawn as examples, and those you drew in the Questions
for Practice, we can start to identify a number of characteristics of straight-line graphs.
The general equation for a straight line is y  mx  c where m and c are constants which may
have any numerical value, including 0. The value of m is known as the coefficient of x, and
plays a very important part in calculations concerning a straight-line graph.
 Lines parallel to an axis
Consider the case where m  0. This means y  c.
Whatever value x has, y is always equal to c. The graph of y  c must, then, be a line
parallel to the x-axis at a distance c from it. Note that y  0 is the x-axis itself.
In the same way, x  c is a line parallel to the y-axis at a distance c from it. x  0 is the
y-axis.
132 Introduction to Graphs
 Lines through the origin
If c  0, then the equation of the line is y  mx. When x  0 it is clear that y  0, so the
graph of y  mx is a line through the origin (0,0).
 Parallel lines
When the coefficient of x is the same for several lines, the lines are parallel. Thus, if m
is kept the same and c has different values, we get a set of parallel straight lines.
 Direction of slope
If the coefficient of x is negative (for example, y  4  3x) the line will slope down from
left to right instead of up, as it does when the coefficient of x is positive. Thus, in y  mx
 c, if m is positive the line slopes up from left to right; if m is negative the line slopes
down.
 Intercepts
The distances from the origin to the point where the line cuts the axes are called the
intercepts on the axes. They are found by putting x  0 and y  0 in the equation of the
graph. In y  mx  c, the intercept on the y-axis is always at c.
A quadratic equation is of the form y  ax
2
 bx  c, where a, b and c are constants. This will
always give you a curved graph, and the part that is usually asked for is the part that does a
"U" turn.
The general shape of a quadratic equation will be as shown in Figure 5.9.
Figure 5.9: General shape of a quadratic equation
In contrast to linear graphs, in order to draw a quadratic graph accurately, you need as many
points as possible – especially around the dip.
The maximum and minimum values of the quadratic equation can be found by careful
inspection of the graph.
or
where a is negative
y
x
where a is negative
y
x
Introduction to Graphs 133
Example 1:
Draw the graph of y  x
2
taking whole number values of x from x  3 to x  3 inclusive. What
are the co-ordinates at which the function y  x
2
is minimised?
As ever, the first step is to draw up a table of values for the given range of x.
x 3 2 1 0 1 2 3
y  x
2
9 4 1 0 1 4 9
Plot these seven points on the graph and then join the points with a smooth curve, as shown
in Figure 5.10. From the graph you will see that function y  x
2
is minimised where y  0 and
x  0 (i.e. the co-ordinates where y  x
2
is minimised are (0,0)).
Figure 5.10: Graph of y  x
2
-2
0
2
4
6
8
10
-3 -2 -1 0 1 2 3
x
y
y  x
2
134 Introduction to Graphs
Example 2:
Draw the graph of y  2x
2
from x  3 to 3. Then from the graph find:
(a) the values of x when y  10
(b) the value of y when x  1.25
(c) the values of x and y where the function y  2x
2
is minimised.
Table of values:
x 3 2 1 0 1 2 3
x
2
9 4 1 0 1 4 9
y  2x
2
18 8 2 0 2 8 18
Plot the points and draw the graph as in Figure 5.11.
Figure 5.11: Graph of y  2x
2
(a) when y  10, there are two values of x, at points A and B: x  2.25 and x  2.25
(b) when x  1.25, the value of y is at the point C: y  3.3
(c) when the function y  2x
2
is minimised, x  0 and y  0.
0
2
4
6
8
10
12
14
16
18
-3 -2 -1 0 1 2 3
x
y
y  2x
2
y  10
y  1.25
A B
C
Introduction to Graphs 135
Example 3:
Draw the graph of y  x
2
 2x  1 from x  2 to 3.
At what value of x does the minimum value of y occur?
Table of values:
x 2 1 0 1 2 3
x
2
4 1 0 1 4 9
2x 4 2 0 2 4 6
1 1 1 1 1 1 1
y  x
2
 2x  1 9 4 1 0 1 4
Plot the points and draw the graph as in Figure 5.12.
Figure 5.12: Graph of y  x
2
– 2x + 1
Reading from the graph, we can see that the minimum value of y occurs when x  1.
0
2
4
6
8
10
-2 -1 0 1 2 3
x
y
136 Introduction to Graphs
Questions for Practice 3
1. Draw the graph of y 
2
1
x
2
from x  4 to 4.
From the graph find:
(a) the value of y when x  3.2
(b) the values of x when y  7.
2. Draw the graph of y  x
2
from x  3 to 3.
3. Draw the graph of y  4x
2
from x  2 to 2.
Find the values of x when y  9.
4. Draw the graph of the profit (£000s) function x
2
 8x  1 from x  0 to 8, where x ('000)
represents output.
From the graph find:
(a) the output at which profit is maximised
(b) the output range at which profits generated amount to at least £8,000.
Now check your answers with those given at the end of the unit.
Equations with x as the Denominator
These equations (hyperbolic equations) take the form of y 
x
a
where a is a constant.
They result in a characteristic curve, the direction of which may vary.
Example 1:
Let's look at the case where a  1:
y 
x
1
We'll take x to have values from x  1 to x  6, which gives the table following and the graph
in Figure 5.13.
x 1 2 3 4 5 6
y 
x
1
1 0.5 0.33 0.25 0.2 0.17
Note that we have avoided taking x  0, since the expression
0
1
is not defined (i.e. it has no
meaning).
Introduction to Graphs 137
0
0.2
0.4
0.6
0.8
1
1.2
0 2 4 6
0
1
2
3
4
0 2 4 6 8
Figure 5.13: Graph of y 
x
1
This graph illustrates the general shape of graphs representing y 
x
a
. In each case, as x
gets bigger, y gets smaller. There is also no point corresponding to x  0.
Example 2:
Draw the graph of y 
x
4
for values of x from x  1 to x  8.
From the graph, find the value of x when y  2.5.
Table of values:
x 1 2 3 4 5 6 7 8
y 
x
4
4 2 1.33 1 0.8 0.67 0.57 0.5
Figure 5.14: Graph of y 
x
4
Reading from the graph, when y  2.5, x  1.7.
138 Introduction to Graphs
Questions for Practice 4
1. Draw the graph of y 
x
2
from x  1 to x  6.
From your graph, find the value of x when y  1.5.
Now check your answers with those given at the end of the unit.
Just as we saw that the shape of the quadratic graphs changed according to whether the
coefficient of x
2
is positive or negative, so the shape of graphs developed from equations
where x is the denominator change:

x
a
(i.e. where both the a and x are positive) gives a graph which slopes down from left
to right

x
a

gives a graph which slopes down from left to right

x
a

gives a graph which slopes up from right to left

x
a 
gives a graph which slopes up from right to left.
These are illustrated in Figure 5.15.
Figure 5.15: Shape of graphs from the equation
x
a
x
0
y
x
a 
x
a

x
a

x
a
Introduction to Graphs 139
0
20
40
60
80
100
120
140
160
-2 -1 0 1 2 3 4 5
Exponential and Logarithmic Equations
So far in this unit we plotted and interpreted graphs for linear and quadratic equations (also
known as polynomial functions). In the final part of this section we will now plot graphs for
exponential and logarithmic equations. The mathematics behind exponentials and logarithms
was discussed in Study Unit 4.
Example 1:
Let us begin by plotting the graph of an exponential function. Using a scientific calculator, or
otherwise, calculate the value of y  e
x
for the values of x from x  2 to x  5. Plot a graph of
this exponential function and from the graph find the value of x when y  10.
Table of values:
x -2 -1 0 1 2 3 4 5
y  e
x
0.14 0.37 1.00 2.72 7.39 20.1 54.60 148.41
Figure 5.16: Graph of equation y  e
x
Reading from the graph, when y  10, x  2.5 (approximately).
140 Introduction to Graphs
-1
-0.5
0
0.5
1
1.5
0 1 2 3 4
Example 2:
In this example we will plot a graph of the natural logarithmic function. Using a scientific
calculator, calculate the values of y  ln(x) for the values of x from x  0.5 to x  4.0. Plot a
graph of this logarithmic function and from the graph find the value of x when y  0.
Table of values:
x 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
ln (x) 0.693 0.00 0.405 0.693 0.916 1.099 1.253 1.386
Figure 5.17: Graph of equation ln(x)
Reading from the graph, when y  0, x  1.
Questions for Practice 5
1. Use a scientific calculator to construct a table of values for y  log
10
(x) when x  0,5,
10, 15, 20 and 25 (give your answers to 2 decimal places). Plot these values on a
graph.
2. The sales S (in thousands of units) of a product after it has been on the market for t
months is given by the equation: S  100(1  e
0.1625t
).
(a) Construct a table of values for S for t  0,5, 10, 15, 20 and 25 (give S values to
the nearest whole number).
(b) Use the table to draw a graph of S against t and from your graph estimate the
value of S when t is very large.
Now check your answers with those given at the end of the unit.
Introduction to Graphs 141
The gradient of a line is a measure of how much it slopes.
 A horizontal line has a gradient of zero:
 A vertical line has an infinitely large gradient:
 In between, the gradient increases:
or decreases
Now, consider the five slopes illustrated in Figure 5.18.
In the five shapes, the slope of the line(s) increases as we go from (i) through to (v).
Figure 5.18: Gradient of a slope
In (i), the gradient (s) is zero (no slope).
In the other cases, the gradient is defined as the height of the right-angled
triangle divided by the base. Therefore
in (ii), the gradient of s is 1/2  0.5
in (iii), the gradient of s is 2/2  1
in (iv), the gradient of s is 4/2  2
in (v), the gradient of s is 8/2  4
2
(ii)
s
2
(i)
s
1
2
(iii)
s
2
2
(iv)
s 4
2
(v)
s 8
142 Introduction to Graphs
If a line slopes down from left to right (instead of up), then the gradient is negative.
This is denoted by introducing a negative sign, as illustrated by Figure 5.19.
The gradient of the line s is 
2
2
 1.
We can easily calculate the gradient of a straight line joining two points from the co-ordinates
of those points.
Firstly, we shall examine the method by reference to plotting the line on a graph.
Example 1:
Plot the points (1,3) and (4,8) and find the gradient of the line joining them.
The procedure would be as follows:
(a) Plot the two points on a graph and join them to give a sloping line, as shown in Figure
5.20.
Figure 5.20: Gradient of the line connecting points (1, 3) and (4, 8)
(b) Make this line the hypotenuse of a right-angled triangle by drawing in a horizontal
"base" and a vertical "height".
2
s
2
x
y
5
3
1 2 3 4 5
9
4
6
5
3
2
7
1
8
Introduction to Graphs 143
(c) Work out the lengths of the base and the height from the graph:
base  3 height  5.
As we have seen, the gradient is given by the equation
base
height

3
5

3
2
1
Note that the slope is positive.
Therefore, the gradient of the straight line joining (1, 3) and 4, 8) is
3
2
1
Note that, at step (c), we could have worked out the lengths of the base and the height from
the co-ordinates of the points:
length of the base  difference between x co-ordinates  4  1  3
length of the height  difference between y co-ordinates  8  3  5
Example 2:
Plot the points (1,4) and (2,3) and find the gradient of the line joining them.
Plot the points, join them and construct a right-angled triangle with the line forming the
hypotenuse:
Figure 5.21: Gradient of the line connecting points (–1, 4) and (2, –3)
This time we shall work out the lengths of the base and the height from the co-ordinates of
the points.
base  2  (1)  3
height  4  (3)  7
We can check that this is correct by reference to the plotted triangle on the graph.
Note that the slope is negative.
base
height
 
3
7
 2
3
1
2 1 2 3 1
x
y
2
1
1
2
3
3
4
7
3
144 Introduction to Graphs
General formula
We can now develop a general formula for the gradient of a straight line by reference to its
co-ordinates.
To find the gradient of a line, take any two points A and B on the line. DrawAC parallel to the
x-axis, and CB parallel to the y-axis. The gradient is then given by the ratio CB : AC.
Figure 5.22: Gradient of a straight line
Note that these lengths are directed lengths: AC is positive if the movement fromA to C is in
the same direction as O to x, and negative if the movement fromA to C is in the opposite
direction to O to x. In the same way, CB is positive if the movement from point C to B is in the
same direction as O to y. (The point at O is the origin, i.e. (0,0).)
Taking the lines in Figure 5.22, we find the following:
 In line (i), AC and CB are both positive, hence the gradient
AC
CB
is positive.
 In line (ii), AC is negative, CB is positive, and the gradient is negative.
We can further see that AC is the difference between the co-ordinates along the x-axis and
CB is the difference between the co-ordinates along the y-axis. Therefore, we could calculate
the gradient where the points are, say, (5,1) and (7,2) as follows:
5 7
1 2

2
1
For points (1,2) and (2,3), the gradient would be:
) 1 ( 2
) 2 ( 3
 
  

1 2
2 3

 
 
3
1
B
x
y
A
C C
A
B
(i) (ii)
O
Introduction to Graphs 145
Equation of a Straight Line and Gradient
We can now go back to consider aspects of the equation of a straight line which we have met
before.
Consider a general straight line drawn on a graph. D is the point on that line where it crosses
the y-axis, and P is another point along the line. This is shown in Figure 5.23.
Figure 5.23: Proving the equation of a straight line
OAPB is a rectangle and OAGD is a rectangle.
Let: OA  x, OB  y, the intercept OD  c and the gradient of DP  m.
Then:
m 
DG
PG

OA
PG

OA
BD

OA
DO BO

x
c y 
Rearranging the equation m 
x
c y 
we get:
mx  y  c
y  mx  c
This is the general equation of any straight line. Note that c is the intercept that the line
makes with the y-axis. If the line cuts the y-axis below the origin, its intercept is negative, c.
Using this method, given any straight-line graph, its equation can be found immediately.
Also, given any linear equation, we can find the gradient of the line (m) immediately. For
example:
Find the gradient of a line with the equation 6x  3y  19.
We need to put the equation into the general form y  mx  c:
y 
3
19 x 6  
P
x
y
A
G
B
O
D
146 Introduction to Graphs
0
2
4
6
8
10
0 2 4 6 8
3
6 
 2
Finally, consider the graph in Figure 5.24.
Figure 5.24: Deriving the equation of a straight line
The gradient of the line, m 
6
5 10 

6
5
The intercept, c  5.
Therefore, the equation of this line is:
y  5 x
6
5

or
6y  5x  30
Questions for Practice 6
1. Find the gradients of the straight lines joining the following points.
(a) (1,3) and (3,7)
(b) (2,1) and (5,6)
(c) (2,6) and (1,1)
(d) (4,5) and (2,0)
(e) (7,0) and (1,4)
(f) (0,2) and (4,7)
2. Find the gradients of the lines with the following equations:
(a) 2y  3x  6  0
(b) 7y  4x
(c) 7y  1  2x
Introduction to Graphs 147
The next three questions bring together aspects of the whole unit so far. They are actual
examination questions.
3. Draw an x-axis from 3 to 5 and a y-axis from 4 to 4. On them draw the graphs of y 
x  1 and 2y  4  x. Write down the co-ordinates of the point where the two lines
intersect.
4. Draw the graph of y 
2
2
x 3
for values of x from 3 to 3.
On the same axes draw the line y  5. Write down the co-ordinates of the points where
the line and the curve intersect.
5. Plot the points (6,5) and (3,3). Find (a) the gradient and (b) the equation of the straight
line joining them.
Now check your answers with those given at the end of the unit.
E. USING GRAPHS IN BUSINESS PROBLEMS
In this final part of the unit, we shall consider how we can use the properties of graphs we
have been examining to analyse and solve business problems.
All businesses incur costs in the production and supply of the goods and/or services which
they provide. These are an important consideration in many aspects of business finance, not
least in calculating profit, which at heart is simply:
revenue from sales less costs of production and supply (or cost of sales)
The analysis of costs is a subject of its own and, within that, there are many ways of looking
at costs. One of these is to divide them into fixed and variable costs.
 Fixed costs are those which the business must always meet, regardless of the number
of units produced – such as rent on buildings.
 Variable costs are those which vary with output – such as the cost of raw materials
used as part of the production process.
Thus, if we take the case of a printing company, its fixed costs would include the rent on its
offices, printing room and warehouse, and the variable costs would include paper and ink.
Whether they print six or six hundred books a week, the fixed costs remain the same, but the
variable costs rise with the number of books produced. (Note that fixed costs are only fixed in
the short term – over a long period, they may change, for example by acquiring different
premises which may be cheaper.)
If we consider variable costs, we can say that they are a function of the level of production.
This means that, as the level of production changes, so do the variable costs.
We can turn this observation into an algebraic expression:
let: x  number of units produced (level of production)
b  variable costs
148 Introduction to Graphs
Then:
b  f(x)
This is where we started the unit!
We know that we can draw a graph of this relationship if we know some of the values for x
and y.
Let us suppose that we do have such values and can produce such a graph (Figure 5.25):
Figure 5.25: Graph of relationship between variable costs and level of production
We can work out the gradient of the line on this graph, based on two co-ordinates. One of
those co-ordinates must be (0, 0), since when there is no production, no variable costs will
be incurred. If we know that one other point on the graph is (9, 3), then the gradient (which is
m) is:
m 
0 9
0 3

9
3

3
1
We also know that the intercept (c) of the line is 0, so we can express the relationship of
variable costs to units of production by the following equation in the form of y  mx  c:
a 
3
1
x  0
or
3a  x
Now consider fixed costs. These do not vary with the level of production. So, if we know the
level of fixed costs, we can show this on the same graph as in Figure 5.26.
Variable costs
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7 8 9
Costs
(£000s)
Production (thousands of units)
Introduction to Graphs 149
Figure 5.26: Graph of fixed costs and level of production
We can see that this line is parallel to the x-axis, with an intercept at 2. This is represented by
the equation:
b  2 (where b is fixed costs)
Businesses are not only interested in costs as either fixed or variable, but in their total costs.
We can say that:
total costs  variable costs  fixed costs
If total costs  y, and using the notation from above, we get the following equation:
y  a  b
Substituting from the equations established for a and b, we get:
y 
3
1
x  2
This is a standard linear equation. We could work out a table of values for it (which we shall
skip here) and draw a graph. This has been done, again on the same graph as in Figure
5.26, in Figure 5.27.
Note that the line for total costs is in fact a combination of the lines for variable and fixed
costs. Its intercept is at the level of fixed costs and its gradient is that of variable costs.
Variable costs
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7 8 9
Costs
(£000s)
Production (thousands of units)
Fixed costs
150 Introduction to Graphs
Figure 5.27: Graph of total costs and level of production
Note that the chart is only a model, which is why we can show the variable and total costs
lines as linear. In real life, variable costs do not necessarily rise at a constant rate with
increases in production. Despite this, it is a useful way of looking at costs.
Consider the following situation.
A business wishes to use a courier company on a regular basis. Company A charge a call out
charge of £5.00 plus £1.00 per mile, whilst company B charge £2.00 per mile but no call out
charge. Which one is more economical?
We can develop equations for these charges and then use a graph to show which company
would be more economical at different journey lengths:
Let: x  number of miles
y  total cost (£)
Then, for company A, total cost may be defined as :
y  x  5
For company B, the costs is:
y  2x
The graph is shown in Figure 5.28.
The point at which the two lines intersect means that the values of x and y are the same for
each line. Thus at that point, the cost of using each company is the same. This takes place at
5 miles. For any distance above this, company A is cheaper, whereas for distances less than
five miles, it is cheaper to use company B. For example:
 at 7 miles, A charges £12 and B charges £14
 at 2 miles, A charges £7 and B charges £4.
Some of these situations are shown by dotted lines on the graph.
Variable costs
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7 8 9
Costs
(£000s)
Production (thousands of units)
Fixed costs
Total costs
Introduction to Graphs 151
Figure 5.28: Comparison of courier company charges
Break-even Analysis
We can further develop the costs graph by including revenue on it.
Revenue is the total income from sales of the units produced. If we assume that each unit
sells for £1, we can say that y  x where y is the income from sales (in £) and x is the number
of units sold.
Adding the graph of this to the costs graph shown in Figure 5.27, we get Figure 5.29. (Note
that we are assuming that all units produced are sold.)
Figure 5.29: Costs and revenue graph
The point at which the revenue and cost lines intersect is known as the break-even point. At
that point, revenue exactly equals costs. Here, this takes place at sales of 3,000 units, with
both costs and revenue being £3,000.
0
5
10
15
20
0 2 4 6 8 10
Costs
(£)
Journey length (miles)
Company A
Company B
0
2
4
6
8
0 1 2 3 4 5 6 7 8 9
Costs
(£000s)
Sales(thousands of units)
Total costs
Revenue
Break-even point
152 Introduction to Graphs
Any sales above the break-even point will mean that revenue exceeds costs, and there will
be a profit. At sales levels below the break-even point though, costs exceed revenue and
there will be a loss. For example:
 at sales of 2,000 units, costs are £2,667 whilst revenue is only £2,000 – there will a
loss of £667
 at sales of 6,000 units, costs are £4,000 whilst revenue is £6,000 – there will be a profit
of £2,000.
The break-even point represents the minimum position for the long term survival of a
Questions for Practice 7
1. Teapots R U Ltd makes and sells a small range of teapots. It has worked out that the
following costs for its best selling teapot "Standard" are:
fixed costs  £20,150
variable costs  £5.50 per Standard teapot.
If the price of a Standard teapot is £12.00, how many must Teapots R U Ltd. sell each
week to break even? Use a fixed and variable costs chart (break-even chart) to find the
2. A company makes bicycles and sells them for £80.00. The fixed cost of producing them
is £80,000 with a variable cost per bicycle of £36.00. Using a fixed and variable costs
chart (break-even chart), find the number of bicycles needed to be sold for the
company to break even. Give your answer to the nearest 100 bicycles.
Now check your answers with those given at the end of the unit.
Economic Ordering Quantities
We shall now look at the role of graphs in analysing and helping to solve a purchasing
problem. We shall not work through the problem in detail, but just review the general
approach and the way in which graphs can be used.
One of the central issues in purchasing materials to be used as part of the production
process is how much to order and when. Many producers and most retailers must purchase
the materials they need for their business in larger quantities than those needed immediately.
However, there are costs in the ordering of materials – for example, each order and delivery
might incur large administrative and handling costs, including transportation costs, so buying-
in stock on a daily basis might not be appropriate. On the other hand, there are also costs
involved in holding a stock of unused materials – very large stocks can incur large
warehousing costs. Set against this, buying in larger quantities has the effect of spreading
the fixed costs of purchase over many items, and can also mean getting cheaper prices for
buying in bulk. The purchaser must determine the optimum point.
By "economic order quantity" we mean the quantity of materials that should be ordered in
order to minimise the total cost. In order to do this, we need to look at the relationships
between the usage, supply and costs involved.
Introduction to Graphs 153
Suppose we have the following problem:
 We want to order 3,600 units over one year and these units are used at a constant rate.
 As soon as we order, we receive the whole order instantaneously – unrealistic but let
us keep the problem simple.
 We order as soon as existing stock runs out.
 The cost of placing the order is £30.
 The cost of keeping one unit in stock for a year is £2.
We have several options as to how we could tackle this:
(a) Place a single order for the whole year.
(b) Make 12 monthly orders.
(c) Make three orders at four-monthly intervals.
(d) Make six orders bimonthly.
(e) Make 36 orders at 10-day intervals.
There are many other possibilities but these will serve our purpose.
The relevant costs for each option above are:
(a) £30 for one order  storage costs of £2  1,800  £3,630.
(As we place an order for 3,600 units and use them evenly through the year, the
average stock held in storage will be 1,800 units.)
(b) (12  £30) for 12 orders  (£2  150) storage  £660
(In this case, the orders are for 300 each, so the average holding is 150.)
(c) (3  £30)  (£2  600)  £1,290.
(600 is the average holding on orders of 1,200 each.)
(d) (6  £30)  (£2  300)  £780.
(Six orders of 600 each, giving the average holding of 300.)
(e) (36  £30)  (£2  50)  £1,180
(36 orders of 100 each, giving the average holding of 50.)
We can see that the cheapest option is (b), i.e. monthly orders of 300. But is it really the
cheapest? We would need to try many other options before we could be sure.
Let us consider another problem.
A company called Oily Engineers Ltd uses 1,000 sprockets every day. It could purchase daily,
but considers this to be unacceptable as the cost of raising an order and sending a van to a
cash-and-carry to collect the goods has been estimated at £100.
If it buys in lots of 1,000Q sprockets, then the stock will last for Q days. The graph of stock
level is a saw-tooth. There will be 1,000Q sprockets in stock immediately after a purchase
has been made. Q days later this will be down to zero, when another purchase will be made.
The average stock level is 500Q (see Figure 5.30).
154 Introduction to Graphs
Figure 5.30: Stock levels over time
The warehousing cost is linear – if the amount purchased in each order is doubled, then the
warehousing costs are doubled as there are, on average, twice as many items in the
warehouse. This shown in Figure 5.31.
Figure 5.31: Warehousing costs
The cost of purchasing is not linear. The cost per annum depends on the number of orders
placed, as it is a fixed amount per order.
For example, assume that there are 250 working days each year:
(a) Sprockets purchased in batches of 1,000, i.e. a batch every day:
cost  £100  250  £25,000
(b) Sprockets purchased in batches of 2,000, i.e. a batch every other day:
cost  £100  125  £12,500
(c) Sprockets purchased in batches of 4,000, i.e. a batch every fourth day:
cost  £100  62.5  £6,250.
Each time the amount purchased is doubled, the cost per annum halves. The graph is shown
in Figure 5.32.
Size of purchased lot
Cost
Stock
level
0
2Q
Days
3Q Q
1,000Q
Introduction to Graphs 155
Figure 5.32: Cost of purchasing (per annum)
These costs can be added to the warehouse costs to give a total cost per annum. The graph
of this is shown in Figure 5.33.
Figure 5.33: Total costs of purchasing
Thus, we can identify the economic order quantity from the minimum cost position on the
graph.
Plotting Time Series Data
Businesses and governments statistically analyse information and data collected at regular
intervals over extensive periods of time, to plan future strategies and policies. For example,
sales values or unemployment levels recorded at yearly, quarterly or monthly intervals could
be examined in an attempt to predict their future behaviour. Such sets of values observed at
regular intervals over a period of time are called time series.
The analysis of this data is a complex problem as many variable factors may influence the
changes which take place. The first step in analysing the data is to plot the observations on a
scattergram, with time evenly spaced across the x-axis and measures of the dependent
variable forming the y-axis. This can give us a good visual guide of the trend in the time
series data.
Cost
Size of purchased lot
Cost
Size of purchased lot
Total cost
cost
Storage or
warehousing
cost
Minimum
cost
156 Introduction to Graphs
Let us consider a factory employing a number of people in producing a particular commodity,
say thermometers. Naturally, at such a factory during the course of a year some employees
will be absent for various reasons. The following table shows the number of days lost through
sickness over the last five years. Each year has been broken down into four quarters of three
months. We have assumed that the number of employees at the factory remained constant
over the five years.
Year Quarter Days Lost
2003 1 30
2 20
3 15
4 35
2004 1 40
2 25
3 18
4 45
2005 1 45
2 30
3 22
4 55
2006 1 50
2 32
3 28
4 60
2007 1 60
2 35
3 30
4 70
Table 5.1: Days lost (per quarter) through sickness at a thermometer factory
We can plot this information using a particular type of scattergram, as shown in Figure 5.34.
The scattergram of a time series is often called a historigram. (Do not confuse this with a
histogram, which is a type of bar chart.) Note the following characteristics of a historigram:
 It is usual to join the points by straight lines. The only function of these lines is to help
your eyes to see the pattern formed by the points.
 Intermediate values of the variables cannot be read from the historigram.
 A historigram is simpler than other scattergrams since no time value can have more
than one corresponding value of the dependent variable.
 It is possible to change the scale of the y-axis to highlight smaller changes.
Every historigram will look similar to the one in Figure 5.34, depending on the scale of the y-
axis.
Introduction to Graphs 157
Figure 5.34: Historigram (scattergram) of days lost (per quarter) through sickness
(2003–2007)
There are four factors that influence the changes in a time series over time. These are the
trend, seasonal variations, cyclical fluctuations, and irregular or random fluctuations.
However, you only need to concern yourself with the trend.
The trend is the change in general level over the whole time period and is often referred to as
the secular trend. You can see in Figure 5.34 that the trend is definitely upwards, in spite of
the obvious fluctuations from one quarter to the next. A trend can thus be defined as a clear
tendency for the time series data to travel in a particular direction in spite of inter-period
fluctuations (be they large or small).
If we were to draw a simple linear trend line through the time series data shown in Figure
5.34, then we would get a trend line similar to that shown in Figure 5.35. Based on this trend
line, we can clearly see that lost days from work (per quarter) have increased over the
period, from approximately 25 days lost per quarter in 2003 to approximately 50 days lost per
quarter in 2007.
Lost
days
2003 2004 2005 2006 2007
0
10
20
30
40
50
60
70
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
158 Introduction to Graphs
Figure 5.35: Historigram (scattergram) of days lost (per quarter) through sickness
(2003–2007), with linear trend line
Lost
days
2003 2004 2005 2006 2007
0
10
20
30
40
50
60
70
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
Introduction to Graphs 159
Questions for Practice 8
The following table shows sales of premium bonds at a large post office over five years:
Year and
Quarter
Sales
(£000)
2004 1 320
2 200
3 150
4 400
2005 1 400
2 210
3 180
4 450
2006 1 800
2 400
3 340
4 900
2007 1 950
2 500
3 400
4 980
1. Plot the time series on a historigram.
2. Using a ruler, draw a linear trend line through the time series plotted in Question 1 and
comment on the trend.
Now check your answers with those given at the end of the unit.
160 Introduction to Graphs
Questions for Practice 1
1.
x 0 0.5 1 1.5 2 2.5
y  3x  2 2 0.5 1 2.5 4 5.5
2. y  2x  1 is the only function which satisfies all values in the table.
3. (a) f(2)  10  2  8
(b) f(3)  15  2  17
4. (a) f(2)  4  2  6
(b) f(3)  9  2  11
Questions for Practice 2
1. Table of values:
x 1 0 2
y  5x  4 9 4 6
Graph of y  5x – 4
-10
-8
-6
-4
-2
0
2
4
6
-2 -1 0 1 2 3
x
y
Introduction to Graphs 161
2. Table of values:
x 4 0 2
y  2x  3 5 3 7
Graph of y  2x + 3
3. Table of values:
x 4 1 6
y  2  x 6 1 4
Graph of y  2 – x
-6
-4
-2
0
2
4
6
8
-4 -3 -2 -1 0 1 2
x
y
-4
-2
0
2
4
6
-4 -2 0 2 4 6
x
y
162 Introduction to Graphs
Note that for the next four questions, we do not show the graphs themselves.
4. (i) cuts (iii) at (
3
2
1 ,
3
2
2 ) or (1.7, 2.7) to 1 decimal place.
(i) cuts (iv) at (2.6, 0.8).
(ii) cuts (iii) at (1, 0).
(ii) cuts (iv) at (1.4, 2.1) to 1 decimal place.
5. (i) y  5 is a line parallel to the x-axis and 5 units above it.
(ii) x  5 is a line parallel to the y-axis and 5 units to the left of it.
(iii) y  5x is a line through the origin, sloping up from left to right with a gradient 5 in
1 (i.e. if you go along 1 unit you go up 5 units).
(iv) y  
5
1
x is a line through the origin, sloping down from left to right with a gradient
of 1 in 5 (i.e. along 5 and down 1).
6. These three graphs are parallel lines:
(i) passes through the origin.
(ii) makes an intercept of 3 on the y-axis and 1½ on the x-axis (i.e. crosses the y-
axis at 3 and the x-axis at 1½).
(iii) makes an intercept of 4 on the y-axis and 2 on the x-axis.
7. The lines intersect at (3, 1).
Questions for Practice 3
1. Table of values:
x 4 3 2 1 0 1 2 3 4
x
2
16 9 4 1 0 1 4 9 16
y 
2
1
x
2
8 4½ 2 ½ 0 ½ 2 4½ 8
The graph is shown on the next page. Reading off from it, we get:
(a) when x  3.2, y  5.1
(b) when y  7, x  3.7 and 3.7
Introduction to Graphs 163
0
2
4
6
8
-4 -2 0 2 4
Graph of y 
2
1
x
2
2. Table of values:
x 3 2 1 0 1 2 3
y  x
2
9 4 1 0 1 4 9
Graph of y  –x
2
-10
-8
-6
-4
-2
0
-3 -2 -1 0 1 2 3
164 Introduction to Graphs
3. Table of values:
x 2 1 0 1 2
x
2
4 1 0 1 4
4x
2
16 4 0 4 16
y  4x
2
16 4 0 4 16
Graph of y  –4x
2
Reading from the graph, when y  9, x  1.5 and 1.5.
4. Table of values:
x 0 1 2 3 4 5 6 7 8
-x
2
0 -1 -4 -9 -16 -25 -36 -49 -64
8x 0 8 16 24 32 40 48 56 64
1 1 1 1 1 1 1 1 1 1
y  x
2
 8x  1 1 8 13 16 17 16 13 8 1
The graph is shown on the next page. Reading off from it, we get:
(a) Profit is maximised (£17,000) at 4,000 units of output, i.e. when x  4.
(b) At an output of between 1,000 and 7,000 units (i.e. between x  1and x  7),
profits generated are in excess of £8,000 (i.e. y  8 or above).
-16
-12
-8
-4
0
-2 -1 0 1 2
Introduction to Graphs 165
Graph of y  x
2
+ 8x + 1
Questions for Practice 4
1. Table of values:
x 1 2 3 4 5 6
y 
x
2
2 1 0.67 0.5 0.4 0.33
Graph of y 
x
2
When y  1.5, x  1.3.
0
0.5
1
1.5
2
0 1 2 3 4 5 6
0
2
4
6
8
10
12
14
16
18
0 1 2 3 4 5 6 7 8
Profit
(£000s)
Output (£000s)
166 Introduction to Graphs
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 5 10 15 20 25
0
20
40
60
80
100
120
0 5 10 15 20 25
Questions for Practice 5
1. Table of values for y  log
10
(x):
x 0 5 10 15 20 25
y  log
10
(x) 0.00 0.70 1.00 1.18 1.30 1.40
Graph of y  log
10
(x)
2. Table of values for S 100(1  e
0.1625t
):
t  0 5 10 15 20 25
S  0 56 80 91 96 98
Graph of S 100(1  e
0.1625t
)
From the graph, when t is very large S will be approximately 100,000.
Introduction to Graphs 167
Questions for Practice 6
1. (a)
1 3
3 7

2
4
 2
(b)
3 5
) 1 ( 6

 

3
7

3
1
2
(c)
1 ) 2 (
) 1 ( 6
 
 

3
7

 
3
1
2
(d)
2 ) 4 (
0 5
 

6
5

 
6
5
(e)
) 1 ( 7
) 4 ( 0
 
 

8
4

2
1
(f)
0 4
) 2 ( 7

 

4
9

2
1
2
2. (a) y 
2
6 x 3 
2
3
(b) y 
7
4
7
4
(c) y 
7
1 x 2 
7
2
3. Tables of values:
x 3 1 5
y  x  1 4 0 4
2y  4  x
y  2 
2
1
x
x 2 0 4
2 2 2 2
2
1
x 1 0 2
y  2 
2
1
x
3 2 0
The graph is shown on the following page.
168 Introduction to Graphs
Graph of y  x – 1 and 2y  4 – x
Reading from this, the lines intersect at point (2, 1)
4. Table of values:
x 3 2 1 0 1 2 3
y
2
x 3
2
 13.5 6 1.5 0 1.5 6 13.5
Graph of y 
2
3x
2
and y  5
From the graph, you will see that the line and curve intersect at (1.8, 5) and (1.8, 5).
-4
-3
-2
-1
0
1
2
3
4
-3 -2 -1 0 1 2 3 4 5
y  x  1
2y  4  x
0
2
4
6
8
10
12
14
-3 -2 -1 0 1 2 3
y  5
y
2
x 3
2

Introduction to Graphs 169
0
1
2
3
4
5
0 1 2 3 4 5 6
5.
3 6
3 5

3
2
(b) The line cuts the y-axis at (0,1).
Therefore, the equation is: y 
3
2
x  1 or 3y  2x  3
Questions for Practice 7
1. Break-even chart:
The break-even point is at sales of 3,100 Standard teapots.
0
10
20
30
40
50
0 1 2 3 4
Revenue/
Costs
(£000s)
Sales (000s)
170 Introduction to Graphs
2. Break-even chart
The break-even point is at sales of 1,800 bicycles (to the nearest 100).
Questions for Practice 8
1. Time series historigram:
2. Linear trend is shown on the historigram above. The historigram clearly shows that
sales of premium bonds have increased over the period, from approximately £250,000
per quarter in 2003 to approximately £700,000 per quarter in 2007.
0
50
100
150
200
250
0 1 2 3
Revenue/
Costs
(£000s)
Sales (000s)
Sales
(£000s)
2004 2005 2006 2007
0
200
400
600
800
1000
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
171
Study Unit 6
Introduction to Statistics and Data Analysis
Contents Page
Introduction 172
A. Data for Statistics 173
Preliminary Considerations 173
Types of Data 174
Requirements of Statistical Data 174
B. Methods of Collecting Data 175
Published Statistics 175
Personal Investigation/Interview 176
Delegated Personal Investigation/Interview 176
Questionnaire 176
C. Classification and Tabulation of Data 177
Forms of Tabulation 179
Secondary Statistical Tabulation 181
Rules for Tabulation 183
D. Frequency Distributions 185
Simple Frequency Distributions 186
Grouped Frequency Distributions 186
Choice of Class Interval 188
Cumulative Frequency Distributions 189
Relative Frequency Distributions 189
Answers to Questions for Practice 191
172 Introduction to Statistics and Data Analysis
INTRODUCTION
This unit is the first of five concerned specifically with statistics.
"Statistics" is a word which is used in a variety of ways and with a variety of meanings, but, in
whatever way it is used, it is always concerned with numerical information. There are two
particular meanings of the word which concern us, namely:
 The numerical facts themselves – for example, we talk of the "statistics" of steel
production.
 The methods of analysing the facts – in this sense, "statistics" is the title of a subject
like "arithmetic" or "chemistry" or "physics". Sometimes the subject is called "statistical
method".
Thus, statistics is concerned with numerical facts about a problem or issue – the facts
themselves, and with methods of analysing these facts in order to solve or shed light on the
problem or issue.
The numerical facts with which statistics are concerned are commonly called data. We
therefore talk about the collection and analysis of data in respect of a particular problem.
The methods used for analysing data can be divided into two categories – descriptive and
inferential.
 Descriptive methods simply manipulate the available data in order to convey
information about it, often in the form of summaries or graphical displays, and usually
for the purpose of simplifying comprehension or appreciation of the facts. These
methods are concerned only with the available data, even if that data is only a part of a
(possibly much larger) whole.
 Inferential methods use the available data in order to make general statements about
the whole from which that data is only a part. In other words, they allow us to infer
something about the whole mass of data from a study of just a part of it. Predictions
and estimates are typical of the general conclusions which may be made by using
inferential methods.
For example, a statistical analysis may be made of the ages of the eleven players in the
England football team. The ages of the individual players form the data. If we determine the
average age of the team from that data, we are applying descriptive methods to tell us
something about the eleven players as a whole. However, the players are also a part of a
much larger group – for example, all professional English football players – and clearly their
average age will not necessarily be an accurate estimate of the average age of this larger
group. If we wanted to draw conclusions about this larger group from the study of the eleven
players in the England team, we would need to use inferential methods.
In this course we shall only be concerned with descriptive statistical methods. In this study
unit we shall review the basis of the statistical method – the collection of facts (data) about a
problem or issue – and then consider initial ways of analysing this data so that it has
meaning in relation to the problem.
Objectives
When you have completed this study unit you will be able to:
 explain the concept of data in statistics
 distinguish between primary and secondary data
 explain the differences between quantitative and qualitative variables and between
continuous and discrete variables
Introduction to Statistics and Data Analysis 173
 explain the reasons for classifying and tabulating data, and prepare tables from given
data
 explain the concept of a frequency distribution and define the terms class frequency,
class boundary, class limits and class interval
 prepare simple, grouped, cumulative and relative frequency distributions from given
data.
A. DATA FOR STATISTICS
A statistical investigation involves a number of stages:
 definition of the problem or issue
 collection of relevant data
 classification and analysis of the collected data
 presentation of the results.
So even before the collection of data starts, there are some important points to consider
when planning a statistical investigation.
Preliminary Considerations
It is important to be aware of these issues as they impact on the data which is to be collected
and analysed.
 Exact definition of the problem
This is necessary in order to ensure that nothing important is omitted from the enquiry,
and that effort is not wasted by collecting irrelevant data. The problem as originally put
to the statistician is often of a very general type and it needs to be specified precisely
before work can begin.
 Definition of the units
The results must appear in comparable units for any analysis to be valid. If the analysis
is going to involve comparisons, then the data must all be in the same units. It is no use
just asking for "output" from several factories – some may give their answers in
numbers of items, some in weight of items, some in number of inspected batches and
so on.
 Scope of the enquiry
No investigation should be commenced without defining the field to be covered. Are we
interested in all departments of our business, or only some? Are we to concern
ourselves with our own business only, or with others of the same kind?
 Accuracy of the data
To what degree of accuracy is data to be recorded? For example, are ages of
individuals to be given to the nearest year or to the nearest month or as the number of
completed years? If some of the data is to come from measurements, then the
accuracy of the measuring instrument will determine the accuracy of the results. The
degree of precision required in an estimate might affect the amount of data we need to
collect. In general, the more precisely we wish to estimate a value, the more readings
we need to take.
174 Introduction to Statistics and Data Analysis
Types of Data
There are a number of different ways in which data may be categorised.
 Primary and secondary data
In its strictest sense, primary data is data which is both original and has been obtained
in order to solve the specific problem in hand. Primary data is therefore raw data, and
has to be classified and processed using appropriate statistical methods in order to
reach a solution to the problem.
Secondary data is any data other than primary data. Thus it includes any data which
has been subject to the processes of classification or tabulation or which has resulted
from the application of statistical methods to primary data; it includes all published
statistics.
 Quantitative and qualitative data
Variables may be either quantitative or qualitative. Quantitative variables, to which we
shall restrict discussion here, are those for which observations are numerical in nature
(for example a person's height or the number of children in a school). Qualitative
variables relate to non-numeric observations, such as colour of hair. However each
possible non-numeric value (hair colour say) may be associated with a numeric
frequency of occurrence.
 Continuous and discrete data
Quantitative variables may be either continuous or discrete. A continuous variable may
take any value between two stated limits (which may possibly be minus and plus
infinity). Height, for example, is a continuous variable. This is because a person's
height may (with appropriately accurate equipment) be measured to any minute
fraction of a millimetre. A discrete variable however can take only certain values
occurring at intervals between stated limits. For most (but not all) discrete variables,
these interval values are the set of integers (whole numbers).
For example, if the variable is the number of children per family, then the only possible
values are 0, 1, 2, etc. because it is impossible to have other than a whole number of
children. However, in Britain, shoe sizes are stated in half-units, and so here we have
an example of a discrete variable which can take the values 1, 1½, 2, 2½, etc.
Requirements of Statistical Data
Having decided upon the preliminary matters about the investigation, the statistician must
look in more detail at the actual data to be collected. The desirable qualities of statistical data
are as follows.
 Homogeneity
The data must be in properly comparable units. "Five houses" means little since five
slum hovels are very different from five ancestral mansions. Houses cannot be
compared unless they are of a similar size or value. If the data is found not to be
homogeneous, there are two methods of adjustment possible.
(a) Break down a non-homogeneous group into smaller component groups which are
homogeneous, and study them separately.
(b) Standardise the data. Use units such as "output per person-hour" to compare the
output of two factories of very different size. Alternatively, determine a
relationship between the different units so that all may be expressed in terms of
one – in food consumption surveys, for example, a child may be considered
Introduction to Statistics and Data Analysis 175
 Completeness
Great care must be taken to ensure that no important aspect is omitted from the
enquiry.
 Accurate definition
Each term used in an investigation must be carefully defined; it is so easy to be slack
different things to the injured party, the police and the insurance company! Watch out
also, when using other people's statistics, for changes in definition. Acts of Parliament
may, for example, alter the definition of an "indictable offence" or of an "unemployed
person".
 Uniformity
The circumstances of the data must remain the same throughout the whole
investigation. It is no use comparing the average age of employees in an industry at
two different times if the age structure has changed markedly. Likewise, it is not much
use comparing a firm's profits at two different times if the working capital has changed.
B. METHODS OF COLLECTING DATA
When all the foregoing matters have been considered, we come to the question of how to
collect the data we require. The methods usually available are as follows:
 use of published statistics
 personal investigation/interview
 delegated personal investigation/interview
 questionnaire.
Published Statistics
Sometimes we may be attempting to solve a problem that does not require us to collect new
information, but only to reassemble and reanalyse data which has already been collected by
someone else for some other purpose.
We can often make good use of the great amount of statistical data published by
governments, the United Nations, nationalised industries, chambers of trade and commerce
and so on. When using this method, it is particularly important to be clear on the definition of
terms and units and on the accuracy of the data. The source must be reliable and the
information up-to-date.
This type of data is sometimes referred to as secondary data (as we have mentioned), in that
the investigator has not been personally responsible for collecting it, and it thus came to the
investigator "second-hand". By contrast, data which has been collected by the investigator for
the particular survey in hand is called primary data.
The information you require may not be found in its entirety in one source, but parts of it may
appear in several different sources. Although the search through these may be time-
consuming, it can lead to data being obtained relatively cheaply; this is one of the
advantages of this type of data collection. Of course, the disadvantage is that you could
spend a considerable amount of time looking for information which may not be available.
Another disadvantage of using data from published sources is that the definitions used for
variables and units may not be the same as those you wish to use. It is sometimes difficult to
establish the definitions from published information, but, before using the data, you must
establish what it represents.
176 Introduction to Statistics and Data Analysis
Personal Investigation/Interview
In this method the investigator collects the data personally. This method has the advantage
that the data will be collected in a uniform manner and with the subsequent analysis in mind.
However the field that may be covered by personal investigation may be limited. Additionally,
there is sometimes a danger that the investigator may be tempted to select data that accords
with preconceived notions.
The personal investigation method is also useful if a pilot survey is carried out prior to the
main survey, as personal investigation will reveal the problems that are likely to occur.
Delegated Personal Investigation/Interview
When the field to be covered is extensive, the task of collecting information may be too great
for one person. Then a team of selected and trained investigators or interviewers may be
used. The people employed should be properly trained and informed of the purposes of the
investigation. Their instructions must be very carefully prepared to ensure that the results are
in accordance with the requirements described in the previous section of this study unit. If
there are many investigators, personal biases may tend to cancel out.
Care in allocating the duties to the investigators can reduce the risks of bias. For example,
suppose you were investigating the public attitude to a new drug in two towns. Do not put
investigator A to explore town X and investigator B to explore town Y. If you do, any
difference that is revealed might be due to the towns being different, or it might be due to
different personal biases on the part of the two investigators. In such a case, you would try to
get both people to do part of each town.
Questionnaire
In some enquiries the data consists of information which must be supplied by a large number
of people. The most convenient way to collect such data is to issue questionnaires to the
people concerned, asking them to fill in the answers to a set of printed questions. This
method is usually cheaper than delegated personal investigation and can cover a wider field.
(A carefully thought out questionnaire is also often also used in personal or delegated
interview in order to reduce the effect of personal bias.)
The distribution and collection of questionnaires by post suffers from two main drawbacks:
 The forms are completed by people who may be unaware of some of the requirements
and who may place different interpretations on the questions – even the most carefully
worded ones!
 There may be a large number of forms not returned, and these may be mainly by
people who are not interested in the subject or who are hostile to the enquiry. The
result is that we end up with completed forms only from a certain kind of person and
thus have a biased sample. For this reason, it is essential to include a reply-paid
envelope to encourage people to respond.
If the forms are distributed and collected by interviewers, a greater response is likely and
queries can be answered. (This is the method used, for example, in the UK Population
Census.) However care must be taken that the interviewers do not lead respondents in any
way.
Introduction to Statistics and Data Analysis 177
C. CLASSIFICATION AND TABULATION OF DATA
Having completed the survey and collected the data, we need to organise it so that we can
extract useful information and then present our results.
Very often the data will consist of a mass of figures in no special order. For example, we may
have a card index of the 3,000 employees in a large factory. The cards are probably kept in
alphabetical order of names, but they will contain a large amount of other data such as wage
rates, age, sex, type of work, technical qualifications and so on. If we are required to present
to the factory management a statement about the age structure of the workforce (both male
and female), then the alphabetical arrangement does not help us, and no one could possibly
gain any idea about the topic from merely looking through the cards as they are.
What is needed is to classify the cards according to the age and sex of the employee and
then present the results of the classification as a tabulation.
The data in its original form, before classification, is usually known as raw data (i.e. it is
unprocessed).
Example:
Of course we cannot give an example involving 3,000 cards, so we will look at a shortened
version, involving only a small number.
Let us assume that our raw data about employees consists of 15 cards in alphabetical order,
as follows. The age of each employee is stated in years.
Archer, L Mr 39
Black, W Mrs 20
Brown, A Mr 22
Brown, W Miss 22
Carter, S Miss 32
Dawson, T Mr 30
Dee, C Mrs 37
Edwards, D. Mr 33
Edwards, R Mr 45
Gray, J Mrs 42
Green, F Miss 24
Gunn, W Mr 27
Jackson, J Miss 28
Johnson, J Mr 44
Jones, L Mr 39
As it stands, this data does not allow us to draw any useful conclusions about the group of
employees. In order to do this, we need to classify the data in some way. One way would be
to classify the individual cases by sex:
178 Introduction to Statistics and Data Analysis
Male Female
Archer, L Mr 39 Black, W Mrs 20
Brown, A Mr 22 Brown, W Miss 22
Dawson, T Mr 30 Carter, S Miss 32
Edwards, D Mr 33 Dee, C Mrs 37
Edwards, R Mr 45 Gray, J Mrs 42
Gunn, W Mr 27 Green, F Miss 24
Johnson, J Mr 44 Jackson, J Miss 28
Jones, L Mr 39
An alternative way of classifying the group would be by age, splitting them into three groups
as follows:
Age between 20 and 29 Age between 30 and 39
Black, W Mrs 20 Archer, L Mr 39
Brown, A Mr 22 Carter, S Miss 32
Brown, W Miss 22 Dawson, T Mr 30
Green, F Miss 24 Dee, C Mrs 37
Gunn, W Mr 27 Edwards, D Mr 33
Jackson, J Miss 28 Jones, L Mr 39
Age over 40
Edwards, R Mr 45
Gray, J Mrs 42
Johnson, J Mr 44
Again, although the data has now acquired some of the characteristics of useful information,
it is still a collection of data about individuals. In statistics though, we are concerned with
things in groups rather than with individuals. In comparing the height of Frenchmen with the
height of Englishmen, we are concerned with Frenchmen in general and Englishmen in
general, but not with Marcel and John as individuals. An insurance company, to give another
example, is interested in the proportion of men or women who die at certain ages, but it is not
concerned with the age at which John Smith or Mary Brown, as individuals, will die.
We need to transform the data into information about the group, and we can do this by
tabulation. The number of cards in each group, after classification, is counted and the results
presented in a table.
Introduction to Statistics and Data Analysis 179
Table 6.1: Employees by age and sex
Sex
Age group
Male Female
Total
20 – 29 2 4 6
30 – 39 4 2 6
40 – 49 2 1 3
Total 8 7 15
You should look through this example again to make quite sure that you understand what has
been done. Of particular importance to note is the presentation of the original age data,
which is given to the nearest year (and not broken down further into months, days, etc.). This
has allowed the use of the age groups 20 – 29, 30 – 39 and 40 – 49. As we learnt in section
A of this unit, age is a continuous variable, which means it is possible to measure a person’s
age broken down into months, days, minutes and seconds. If the age of employees had been
presented more precisely in the original data, then it is possible that Mr Archer and Mr Jones,
for example, may be slightly older than 39 years. Let us assume that Mr Archer is 39 years, 4
months and 3 days old and Mr Jones is 39 years, 2 months and 20 days old. In which age
group in Table 6.1 would Mr Archer and Mr Jones now be classified? Clearly they are both
older than (exactly) 39 years, therefore they should not be included in the age group 30 – 39,
but they are not as old as (exactly) 40 years, therefore they should not be included in the age
group 40 – 49!
Thus, when tabulating continuous variables such as a person’s age, we need to ensure that
the data groups chosen are appropriate. In most cases (i.e. where a more precise breakdown
of a person’s age is given), we could use the age groups 20 to < 30 (i.e. 20 to less than 30),
30 to < 40 and 40 to < 50. Mr Archer and Mr Jones would now clearly fall into the 30 to less
than 40 years age group (i.e. 30 to < 40). You should note that we did not use the age groups
20 – 30, 30 – 40 and 40 – 50 years. This is because an employee that is, for example,
exactly 40 years old, could not be classified into a single group, as this person would fall into
both the 30 – 40 and 40 – 50 age groups! So choice of date grouping (often referred to as
"class intervals") is an important aspect of the classification and tabulation of data. We shall
look in more detail at the use of class intervals in section D of this unit when we look
"frequency distribution".
Having understood this, you are now in a position to appreciate the purpose behind
classification and tabulation –- it is to condense an unwieldy mass of raw data to manageable
proportions and then to present the results in a readily understandable form.
Forms of Tabulation
We can classify the process of tabulation into simple tabulation and complex or matrix
tabulation.
(a) Simple tabulation
This covers only one aspect of a set of figures. The idea is best conveyed by an
example.
Consider the card index of 3,000 employees mentioned earlier. Assume that each card
carries the name of the workshop in which the person works. A question as to how the
workforce is distributed can be answered by sorting the cards and preparing a simple
table as in Table 6.2.
180 Introduction to Statistics and Data Analysis
Table 6.2: Distribution of employees by workshop
Workshop Number employed
A 600
B 360
C 660
D 840
E 540
Total 3,000
Another question might have been "What is the wage distribution of the employees?"
The answer can be given in another simple table, Table 6.3.
Table 6.3: Wage distribution
Wages group Number of employees
£40 but less than £60 105
£60 but less than £80 510
£80 but less than £100 920
£100 but less than £120 1,015
£120 but less than £140 300
£140 but less than £160 150
Total 3,000
Note that such simple tables do not tell us very much – although it may be enough to
answer the initial question at the moment.
(b) Complex tabulation
This deals with two or more aspects of a problem at the same time. In the problem just
studied, it is very likely that the two questions presented in Tables 6.2 and 6.3 would be
asked at the same time. We could therefore present the answers in a single, although
more complex, table or matrix (as shown in Table 6.4).
Introduction to Statistics and Data Analysis 181
Table 6.4: Distribution of employees and wages by workshop
Wages group (£ per week)
Workshop
40 –
59.99*
60 –
79.99
80 –
99.99
100 –
119.99
120 –
139.99
140 –
159.99
Total no.
employed
A 20 101 202 219 29 29 600
B 11 52 90 120 29 58 360
C 19 103 210 200 88 40 660
D 34 167 303 317 18 1 840
E 21 87 115 159 136 22 540
Total 105 510 920 1,015 300 150 3,000
* Note: 40 – 59.99 is the same as "40 but less than 60" and similarly for the other columns.
This table is much more informative than are the two previous simple tables, although it
is more complicated. We could have complicated it further by dividing the groups into
male and female employees, or into age groups.
Later in this unit, we will look at a list of the rules you should try to follow in compiling
statistical tables. At the end of that list you will find a table relating to our 3,000
employees, which you should study as you read the rules.
Secondary Statistical Tabulation
So far, our tables have merely classified the already available figures – the primary statistics.
However, we can go further than this and do some simple calculations to produce other
figures – secondary statistics.
As an example consider again Table 6.2, and this time calculate how many employees there
are on average per workshop. This is obtained by dividing the total (3,000) by the number of
workshops (5), and the table appears as follows:
Table 6.5: Distribution of employees by workshop
Workshop Number employed
A 600
B 360
C 660
D 840
E 540
Total 3,000
Average number of
employees per workshop: 600
The average is a secondary statistic.
182 Introduction to Statistics and Data Analysis
For another example, we may take Table 6.3 and calculate the proportion of employees in
each wage group, as follows:
Table 6.6: Wage distribution
Wages group Number of employees Proportion of employees
£40.00 – £59.99 105 0.035
£60.00 – £79.99 510 0.170
£80.00 – £99.99 920 0.307
£100.00 – £119.99 1,015 0.338
£120.00 – £139.99 300 0.100
£140.00 – £159.99 150 0.050
Total 3,000 1.000
The proportions given here are also secondary statistics. In commercial and business
statistics it is more usual to use percentages than proportions, and so in Table 6.6 these
would be 3.5%, 17%, 30.7%, 33.8%, 10% and 5%.
Of course secondary statistics are not confined to simple tables. They are used in complex
tables too, as in the example presented in Table 6.7.
Table 6.7: Inspection results for a factory product in two successive years
Year 1 Year 2
Machine no.
Output No. of
rejects
% of
rejects
Output No. of
rejects
% of
rejects
1 800 40 5.0 1,000 100 10.0
2 600 30 5.0 500 100 20.0
3 300 12 4.0 900 45 5.0
4 500 10 2.0 400 20 5.0
Total 2,200 92 4.2 2,800 265 9.5
Average per
machine 550 23 4.2 700 66.2 9.5
The percentage columns and the average row show secondary statistics. All the other figures
are primary statistics.
Note carefully that percentages cannot be added or averaged to get the percentage of a total
or of an average. You must work out such percentages on the totals or averages themselves.
Another danger in the use of percentages has to be watched: you must not forget the size of
the original numbers. Take the case of two doctors treating a particular disease. One doctor
has only one patient and the doctor cures the patient – 100% success! The other doctor has
100 patients of whom 80 are cured – only 80% success! You can see how very unfair it would
be to make a comparison based on the percentages alone.
Introduction to Statistics and Data Analysis 183
Rules for Tabulation
There are no absolute rules for drawing up statistical tables, but there are a few general
principles which, if borne in mind, will help you to present your data in the best possible way.
 Try not to include too many features in any one table (say, not more than four or five)
as otherwise it becomes rather clumsy. It is better to use two or more separate tables.
 Each table should have a clear and concise title to indicate its purpose.
 It should be very clear what units are being used in the table (tonnes, £, people, £000,
etc.).
 Appropriate data groups (or class intervals) for continuous and discrete data should be
chosen to ensure that all data can be included within a single group (or class).
 Blank spaces and long numbers should be avoided, the latter by a sensible degree of
approximation.
 Columns should be numbered to facilitate reference.
 Try to have some order to the table, for example using size, time, geographical location
or alphabetical order.
 Figures to be compared or contrasted should be placed as close together as possible.
 Percentages should be placed near to the numbers on which they are based.
 Use a ruler to draw the tables neatly – scribbled tables with freehand lines nearly
always result in mistakes and are difficult to follow. However, it is useful to draw a rough
sketch first so that you can choose the best layout and decide on the widths of the
columns.
 Insert totals where these are meaningful, but avoid "nonsense totals". Ask yourself
what the total will tell you before you decide to include it. An example of such a
"nonsense total" is given in Table 6.8.
Table 6.8: Election results
Party
Year 1
Seats
Year 2
Seats
Total
Seats
A 350 200 550
B 120 270 390
Total 470 470 940
The totals (470) at the foot of the two columns make sense because they tell us the
total number of seats being contested, but the totals in the final column (550, 390, 940)
are nonsense totals for they tell us nothing of value.
 If numbers need to be totalled, try to place them in a column rather than along a row for
easier computation.
 If you need to emphasise particular numbers, then underlining, significant spacing or
bold type can be used. If data is lacking in a particular instance, then insert an asterisk
in the empty space and give the reasons for the lack of data in a footnote.
 Footnotes can also be used to indicate, for example, the source of secondary data, a
change in the way the data has been recorded, or any special circumstances which
make the data seem odd.
184 Introduction to Statistics and Data Analysis
It is not always possible to obey all of these rules on any one occasion, and there may be
times when you have a good reason for disregarding some of them. But only do so if the
reason is really good – not just to save you the bother of thinking!
Now study the layout of the following table (based on our previous example of 3,000
employees) and check through the list of rules to see how they have been applied.
Table 6.9: ABC & Co. – Wage structure of labour force
Numbers of persons in specified categories
Wage group (£ per week)
Workshop
40 –
59.99
60 –
79.99
80 –
99.99
100 –
119.99
120 –
139.99
140 -
159.99
Total no.
employed
%
See note
(a)
(1) (2) (3) (4) (5) (6) (7) (8) (9)
A 20 101 202 219 29 29 600 20.0
B 11 52 90 120 29 58 360 12.0
C 19 103 210 200 88 40 660 22.0
D 34 167 303 317 18 1 840 28.0
E 21 87 115 159 136 22 540 18.0
Total no. in
wage group
105 510 920 1,015 300 150 3,000 100.0
%
See note (b)
3.5 17.0 30.7 33.8 10.0 5.0 100.0 –
Note (a): Total number employed in each workshop as a percentage of the total workforce.
Note (b): Total number in each wage group as a percentage of the total workforce.
This table can be called a "twofold" table as the workforce is broken down by wage and
workshop.
Questions for Practice 1
These questions are typical of those likely to be set in an examination.
1. The human resources/personnel manager of a firm submits the following report on
labour and staff to the managing director:
"The total number of employees is 136, of whom 62 are male. Of these
men, 43 are industrial labourers and the rest are staff. The age distribution
of the 43 is 'under 18' (seven), '18–45' (22) and the others are over 45. Our
total industrial labour force is 113, of whom 25 are 'under 18' and 19 are
'over 45'. The total number of 'under-18s' on the site is 29 (10 male, 19
female). There are 23 'over-45s', of whom only 4 are male staff."
The managing director, quite rightly, thinks that this is gibberish and instructs you to
present it to him in a tabular form which he can readily understand. Show the table you
would produce.
Introduction to Statistics and Data Analysis 185
2. Consider the following report.
"In the period Year 1 – Year 6 there were 1,775 stoppages of work arising from
industrial disputes on construction sites. Of these, 677 arose from pay disputes, 111
from conflict over demarcation, 144 on working conditions and 843 were unofficial
walk-outs. 10% of all stoppages lasted only one day. Of these, the corresponding
proportions were: 11% for pay disputes, 6% in the case of demarcation and 12½% in
the case of working conditions.
In the subsequent period Year 7 – Year 12, the number of stoppages was 2,807 higher.
But, whereas the number of stoppages arising from pay disputes fell by 351 and those
concerning working conditions by 35, demarcation disputes rose by 748. The number
of pay disputes resolved within one day decreased by 45 compared with the earlier
period, but only by 2 in the case of working conditions. During the later period 52
stoppages over demarcation were settled within the day, but the number of stoppages
from all causes which lasted only one day was 64 greater in the later period than in the
earlier."
You are required to:
(a) Tabulate the information in the above passage, making whatever calculations are
needed to complete the entries in the table.
(b) Write a brief comment on the main features disclosed by the completed table.
Now check your answers with those given at the end of the unit.
D. FREQUENCY DISTRIBUTIONS
A frequency distribution is a tabulation which shows the number of times (i.e. the frequency)
each different value occurs.
For example, the following figures are the times (in minutes) taken by a shop-floor employee
to perform a given repetitive task on 20 specified occasions during the working day:
3.5 3.8 3.8 3.4 3.6
3.6 3.8 3.9 3.7 3.5
3.4 3.7 3.6 3.8 3.6
3.7 3.7 3.7 3.5 3.9
If we now assemble and tabulate these figures, we can obtain a frequency distribution as
follows:
186 Introduction to Statistics and Data Analysis
Table 6.10: Time taken for one employee to perform task
Length of time
(minutes)
Frequency
3.4 2
3.5 3
3.6 4
3.7 5
3.8 4
3.9 2
Total 20
Simple Frequency Distributions
A useful way of preparing a frequency distribution from raw data is to go through the records
as they stand and mark off the items by the "tally mark" or "five-bar gate" method. First look
at the figures to see the highest and lowest values so as to decide the range to be covered
and then prepare a blank table. Then mark the items on your table by means of a tally mark.
To illustrate the procedure, Table 6.11 shows the procedure for the data in Table 6.10 after all
20 items have been entered.
Table 6.11: Time taken for one employee to perform task
Length of time
(minutes)
Tally marks Frequency
3.4 | | 2
3.5 | | | 3
3.6 | | | | 4
3.7 | | | | 5
3.8 | | | | 4
3.9 | | 2
Total 20
Grouped Frequency Distributions
Sometimes the data is so extensive that a simple frequency distribution is too cumbersome
and, perhaps, uninformative. Then we make use of a grouped frequency distribution.
In this case, the "length of time" column consists not of separate values, but of groups of
values:
Introduction to Statistics and Data Analysis 187
Table 6.12: Time taken for one employee to perform task
Length of time
(minutes)
Tally marks Frequency
3.4 to 3.5 | | | | 5
3.6 to 3.7 | | | | | | | | 9
3.8 to 3.9 | | | | | 6
Total 20
Grouped frequency distributions are only needed when there is a large number of values
and, in practice, would not have been required for the small amount of data in our example.
Table 6.13 shows a grouped frequency distribution used in a more realistic situation, when an
ungrouped table would not have been of much use.
Table 6.13: Age distribution of employees in an office
Age group
(years)
Number
of employees
16 to < 21 10
21 to < 26 17
26 to < 31 28
31 to < 36 42
36 to < 41 38
41 to < 46 30
46 to < 51 25
51 to < 56 20
56 to < 60 10
Total 220
The various groups (for example, "26 to < 31") are called "classes" and the range of values
covered by a class (five years in this example) is called the "class interval". The number of
items in each class (for example, 28 in the 26 to < 31 class) is called the "class frequency"
and the total number of items (in this example, 220) is called the "total frequency".
The term "class boundary" is used to denote the dividing line between adjacent classes, so in
the age group example (see Table 6.13) the class boundaries are 16, 21, 26, ... years. In the
length of time example (see table 6.12), as grouped earlier in this section, the class
boundaries are 3.35, 3.55, 3.75, 3.95 minutes. This needs some explanation. As the original
readings were given correct to one decimal place, we assume that is the precision to which
they were measured. If we had had a more precise stopwatch, the times could have been
measured more precisely. In the first group of 3.4 to 3.5 we put times which could in fact be
anywhere between 3.35 and 3.55 if we had been able to measure them more precisely. A
time such as 3.57 minutes would not have been in this group as it equals 3.6 minutes when
corrected to one decimal place and it goes in the 3.6 to 3.7 group.
The term "class limits" is used to stand for the lowest and highest values that can actually
occur in a class. In the age group example, these would be 16 years and 20 years 364 days
188 Introduction to Statistics and Data Analysis
for the first class, 21 years and 25 years 364 days for the second class and so on, assuming
that the ages were measured correct to the nearest day. In the length of time example, the
class limits are 3.4 and 3.5 minutes for the first class and 3.6 and 3.7 minutes for the second
class.
You should make yourself quite familiar with these terms.
Choice of Class Interval
When compiling a frequency distribution you should, if possible, make the length of the class
interval equal for all classes. This is so that fair comparisons can be made between one class
and another. Sometimes, however, this rule has to be broken – for example, some grouped
frequency distributions bring together the first few classes and the last few classes because
there are few instances in each of the individual classes.
In such cases, before using the information, it is as well to make the classes comparable by
calculating a column showing "frequency per interval of so much", as in the example for
wage statistics in Table 6.14.
Table 6.14: Annual income statistics
Annual income
to the nearest £
No. of persons
Frequency per
£200 interval
4,801 – 5,000 50,000 50,000
5,001 – 5,200 40,000 40,000
5,201 – 5,600 55,000 27,500
5,601 – 6,000 23,000 11,500
6,001 – 6,400 12,000 6,000
6,401 – 7,200 12,000 3,000
Total 192,000 –
Notice that the intervals in the first column are:
200, 200, 400, 400, 400, 800.
These intervals let you see how the last column was compiled.
A superficial look at the original table (first two columns only) might have suggested that the
most frequent incomes were at the middle of the scale, because of the appearance of the
figure 55,000. But this apparent preponderance of the middle class is due solely to the
change in the length of the class interval, and column three shows that, in fact, the most
frequent incomes are at the bottom end of the scale, i.e. the top of the table.
You should remember that the purpose of compiling a grouped frequency distribution is to
make sense of an otherwise troublesome mass of figures. It therefore follows that we do not
want to have too many groups or we will be little better off; nor do we want too few groups or
we will fail to see the significant features of the distribution. As a practical guide, you will find
that somewhere between about five and 10 groups will usually be suitable.
Introduction to Statistics and Data Analysis 189
Cumulative Frequency Distributions
Very often we are not especially interested in the separate class frequencies, but in the
number of items above or below a certain value. When this is the case, we form a cumulative
frequency distribution as illustrated in column three of Table 6.15.
Table 6.15: Time taken for one employee to perform task
Length of time
(minutes)
Frequency
Cumulative
frequency
3.4 2 2
3.5 3 5
3.6 4 9
3.7 5 14
3.8 4 18
3.9 2 20
The cumulative frequency tells us the number of items equal to or less than the specified
value, and it is formed by the successive addition of the separate frequencies. A cumulative
frequency column may also be formed for a grouped distribution.
The example in Table 6.15 gives us the number of items "less than" a certain amount.
However we may wish to know, for example, the number of persons having more than some
quantity. This can easily be done by doing the cumulative additions from the bottom of the
table instead of the top, and as an exercise you should now compile the "more than"
cumulative frequency column for Table 6.15.
Relative Frequency Distributions
All the frequency distributions which we have looked at so far in this study unit have had their
class frequencies expressed simply as numbers of items. However, remember that
proportions or percentages are useful secondary statistics. When the frequency in each class
of a frequency distribution is given as a proportion or percentage of the total frequency, the
result is known as a "relative frequency distribution" and the separate proportions or
percentages are the "relative frequencies".
The total relative frequency is, of course, always 1.0 (or 100%). Cumulative relative
frequency distributions may be compiled in the same way as ordinary cumulative frequency
distributions.
As an example, the distribution used in Table 6.14 is set out as a relative frequency
distribution in Table 6.16 for you to study.
190 Introduction to Statistics and Data Analysis
Table 6.16: Annual income statistics
Annual income
to the nearest £
Frequency
Relative
frequency (%)
Cumulative
frequency
Relative
cumulative
frequency (%)
4,801 – 5,000 50,000 26 50,000 26
5,001 – 5,200 40,000 21 90,000 47
5,201 – 5,600 55,000 29 145,000 76
5,601 – 6,000 23,000 12 168,000 88
6,001 – 6,400 12,000 6 180,000 94
6,401 – 7,200 12,000 26 192,000 100
Total 192,000 100 – –
This example is in the "less than" form, and you should now compile the "more than" form in
the same way as you did for the non-relative distribution.
Questions for Practice 2
The following figures show the number of faulty items made on successive days by a
machine in an engineering workshop:
2 0 2 4 3 2 11 4 0 1
1 4 1 2 1 4 1 7 3 2
4 5 5 6 5 2 4 2 1 2
2 4 0 6 1 0 1 5 1 2
3 1 5 3 2 2 1 3 5 3
1 1 10 1 2 3 4 3 10 6
2 6 5 3 1 2 5 7 2 3
4 4 1 3 7 4 1 7 3 7
1 1 2 6 1 0 8 5 8 1
5 6 1 3 2 2 1 2 6 4
Tabulate these figures as a frequency distribution, giving columns containing the frequency,
the relative frequency, the cumulative frequency and the relative cumulative frequency.
Now check your answers with those given at the end of the unit.
Introduction to Statistics and Data Analysis 191
Questions for Practice 1
1. XYZ Company: Age of Staff and Labour
Industrial Staff Total
Age group
Male Female Total Male Female Total Male Female Total
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Under 18 7 18 25 3 1 4 10 19 29
18 – 45 22 47 69 12 3 15 34 50 84
Over 45 14 5 19 4 0 4 18 5 23
TOTAL 43 70 113 19 4 23 62 74 136
2. (a) The complete table is as follows:
Number of Stoppages of Work Arising from
Industrial Disputeson Construction Sites
Year 1 – Year 6 Year 7 – Year 12
Cause
1 day
More than
1 day
Total 1 day
More than
1 day
Total
Pay 74.47 602.53 677 29.47 296.53 326
Demarcation 6.66 104.34 111 52 807 859
Working
conditions
18 126 144 16 93 109
Unofficial* 78.37 764.63 843 144.03 3,143.97 3,288
All 177.5 1,597.5 1,775 241.5 4,340.5 4,582
* Note that we have given the figures as worked out from the information
supplied. Obviously, though, fractional numbers of stoppages cannot occur, and
you should round off each figure to the nearest whole number.
The subtotals of the columns will then be 177; 1,598; 241; and 4,341.
(b) The overall number of stoppages in the later period was more than 2½ times
greater than in the earlier period. In both periods, the number of unofficial
stoppages was the highest category. This was particularly remarkable in the later
period, when they amounted to more than twice the number of all other
stoppages put together. Demarcation disputes rose in number dramatically,
although stoppages over pay and working conditions both declined. About 95% of
stoppages in the later period lasted more than one day, compared with 90% in
the previous period, with those concerning working conditions being less likely to
last more than one day.
192 Introduction to Statistics and Data Analysis
Questions for Practice 2
Faulty items made on successive days by a machine in an engineering workshop
Number of
faulty items
Number of days
with these faulty
items
Relative
frequency
Cumulative
frequency
Relative
cumulative
frequency
0 5 0.05 5 0.05
1 23 0.23 28 0.28
2 20 0.20 48 0.48
3 13 0.13 61 0.61
4 12 0.12 73 0.73
5 10 0.10 83 0.83
6 7 0.07 90 0.90
7 5 0.05 95 0.95
8 2 0.02 97 0.97
9 0 0.00 97 0.97
10 2 0.02 99 0.99
11 1 0.01 100 1.00
Total 100 1.00 – –
193
Study Unit 7
Graphical Representation of Information
Contents Page
Introduction 194
A. Graphical Representation of Frequency Distributions 194
Frequency Dot Diagram 195
Frequency Bar Chart 195
Frequency Polygon 196
Histogram 197
The Ogive 200
Stem and Leaf Diagram 203
B. Pictograms 204
Limited Form 204
Accurate Form 205
C. Pie Charts 206
D. Bar Charts 207
Component Bar Chart 207
Horizontal Bar Chart 208
E. Scatter Diagrams 209
F. General Rules for Graphical Presentation 210
Answers to Questions for Practice 215
194 Graphical Representation of Information
INTRODUCTION
In this study unit we shall move on and look at further methods of summarising and
presenting data as a means of providing useful information.
Diagrams are particularly effective as a way of conveying quite complex information.
Presenting information visually is easy to understand, and enables broad distributions and
trends to be taken in quickly. However, diagrams are rarely as accurate as the figures
themselves, and details are likely to be lost.
In the first part of this unit we shall examine the specific graphical forms used for presenting
frequency distributions of the type considered in the previous study unit. We then move on to
discuss a number of popular graphical presentations that are used in everyday life,
presentations that are not necessarily used by statisticians (e.g. those used in business
journals and the mass media of newspapers and television).
Objectives
When you have completed this study unit you will be able to:
 use frequency dot diagrams, frequency bar charts, frequency polygons, histograms,
ogives and stem and leaf diagrams to represent frequency distributions
 prepare pictograms, pie charts, bar charts and scatter diagrams from given data, and
discuss the circumstances in which each form of presentation is most effective
 describe the general rules and principles for using graphical representations of data.
A. GRAPHICAL REPRESENTATION OF FREQUENCY
DISTRIBUTIONS
Tabulated frequency distributions are sometimes more readily understood if represented by a
diagram. Graphs and charts are normally much superior to tables (especially lengthy
complex tables) for showing general states and trends.
The methods of presenting frequency distributions graphically are as follows:
 frequency dot diagram
 frequency bar chart
 frequency polygon
 histogram
 ogive.
However, such graphs and charts cannot usually be used for accurate analysis of data. The
following is a diagram that to some extent overcomes this limitation:
 stem and leaf diagram.
We will now examine each of these graphical methods in turn, initially using the data on the
times (in minutes) taken by an employee to perform a given task on 20 specified occasions
during the working day which we previously considered in Study Unit 6 (Table 6.10). The
tabulated frequency distribution for this data is as follows:
Graphical Representation of Information 195
Table 7.1: Time taken for one employee to perform task
Length of time
(minutes)
Frequency
3.4 2
3.5 3
3.6 4
3.7 5
3.8 4
3.9 2
Total 20
Frequency Dot Diagram
This is a simple form of graphical representation for the frequency distribution of a discrete
variate. (In statistics, the term "variate" is used when you might have expected the term
"variable". Whilst the two terms do not mean exactly the same thing, roughly speaking, at this
level, they can be used interchangeably. Loosely, a variate is a quantity that can have any of
a set of numerical values. Do not worry too much about the difference between these two
terms.)
A horizontal scale is used for the variate and a vertical scale for the frequency. Above each
value on the variate scale we mark a dot for each occasion on which that value occurs.
Thus, a frequency dot diagram of the information in Table 7.1 would be as shown in Figure
7.1.
Figure 7.1: Frequency dot diagram to show length
of time taken by operator to complete a given task
Frequency
Length of time (minutes)
3.4 3.5 3.6 3.7 3.8 3.9
196 Graphical Representation of Information
Frequency Bar Chart
We can avoid the business of marking every dot in such a diagram by drawing instead a
vertical line, the length of which represents the number of dots which should be there. The
frequency dot diagram in Figure 7.1 now becomes a frequency bar chart, as in Figure 7.2.
Figure 7.2: Frequency bar chart to show length
of time taken by operator to complete a given task
Frequency Polygon
Instead of drawing vertical bars as we do for a frequency bar chart, we could merely mark
the position of the top end of each bar and then join up these points with straight lines. When
we do this, the result is a frequency polygon, as in Figure 7.3.
Figure 7.3: Frequency polygon to show length
of time taken by operator to complete a given task
Note that we have added two fictitious classes at each end of the distribution, i.e. we have
marked in groups with zero frequency at 3.3 and 4.0. This is done to ensure that the area
enclosed by the polygon and the horizontal axis is the same as the area under the
corresponding histogram which we shall consider in the next section.
Frequency
Length of time (minutes)
0
1
2
3
4
5
6
3.4 3.5 3.6 3.7 3.8 3.9
Frequency
Length of time (minutes)
0
1
2
3
4
5
6
3.3 3.4 3.5 3.6 3.7 3.8 3.9 4
Graphical Representation of Information 197
These three kinds of diagram are all commonly used as a means of making frequency
distributions more readily comprehensible. They are mostly used in those cases where the
variate is discrete and where the values are not grouped. Sometimes, frequency bar charts
and polygons are used with grouped data, by drawing the vertical line (or marking its top end)
at the centre point of the group.
Histogram
A histogram is the best way of showing a grouped frequency distribution.
To illustrate this, we shall use an example of a grouped frequency distribution similar to one
from Study Unit 6 (Table 6.13), as set out in Table 7.2. The information in the table may be
shown graphically by the histogram in Figure 7.4.
Table 7.2: Age distribution of employees in an office
Age group
(years)
Number
of employees
15 to < 20 10
20 to < 25 17
25 to < 30 28
30 to < 35 42
35 to < 40 38
40 to < 45 30
45 to < 50 25
50 to < 55 20
55 to < 60 10
Total 220
Figure 7.4: Histogram showing age distribution of employees in an office
Frequency
Age (years)
15 20 25 30 35 40 45 50 60
0
10
20
30
40
55
198 Graphical Representation of Information
In a histogram, the frequency in each group is represented by a rectangle and – this is a very
important point – it is the area of the rectangle, not its height, which represents the
frequency.
When the lengths of the class intervals are all equal, then the heights of the rectangles
represent the frequencies in the same way as do the areas (this is why the vertical scale has
been marked in this diagram). If however the lengths of the class intervals are not all equal,
then you must remember that the heights of the rectangles have to be adjusted to give the
correct areas. (Do not stop at this point if you have not quite grasped the idea, because it will
become clearer as you read on.)
Look once again at the histogram of ages given in Figure 7.4. Note particularly how it
illustrates the fact that the frequency falls off towards the higher age groups – any form of
graph which did not reveal this fact would be misleading. Now let us imagine that the original
table had not used equal class intervals but, for some reason or other, had given the last few
groups as those in Table 7.3.
Table 7.3: Age distribution of employees in an office (amended extract)
Age group
(years)
Number
of employees
40 to < 45 30
45 to < 50 25
50 to < 60 30
The last two groups have been lumped together as one. A wrong form of histogram, using
heights instead of areas, for the amended data would look like Figure 7.5.
Figure 7.5: Histogram showing age distribution of employees in an office
(amended data)
Frequency
Age (years)
15 20 25 30 35 40 45 50 60
0
10
20
30
40
This
column
is the
wrong
size
Graphical Representation of Information 199
Now, this clearly gives an entirely wrong impression of the distribution with respect to the
higher age groups. In the correct form of the histogram, the height of the last group (50 to <
60) would be halved because the class interval is double all the other class intervals. The
histogram in Figure 7.6 gives the right impression of the falling off of frequency in the higher
age groups. The vertical axis has been labelled "Frequency density per 5-year interval" as
five years is the "standard" interval on which we have based the heights of our rectangles.
Figure 7.6: Amended histogram showing age distribution of employees in an office
(amended data)
Often it happens, in published statistics, that the last group in a frequency table is not
completely specified. For example, the last few groups in our age distribution table may have
been specified as follows:
Table 7.4: Age distribution of employees in an office (further amended extract)
Age group
(years)
Number
of employees
40 to < 45 30
45 to < 50 25
50 and over 30
A distribution of this kind is called an "open-ended" distribution.
How would we draw this on a histogram?
If the last (open-ended) group has a very small frequency compared with the total frequency
(say, less than about 1% or 2%), then nothing much is lost by leaving it off the histogram
altogether. If the last group has a larger frequency than about 1% or 2%, then you should try
to judge from the general shape of the histogram how many class intervals to spread the last
frequency over, in order not to create a false impression of the extent of the distribution. In
Frequency
density per
5-year
interval
Age (years)
15 20 25 30 35 40 45 50 60
0
10
20
30
40
200 Graphical Representation of Information
the example given, you would probably spread the last 30 people over two or three class
intervals, but it is often simpler to assume that an open-ended class has the same length as
its neighbour.
Whatever procedure is adopted, it is essential to state clearly what you have done and why.
The Ogive
This is the name given to the graph of the cumulative frequency. It can be drawn in either the
"less than" or the "or more" form, but the "less than" form is the usual one. Ogives for two of
the distributions already considered are now given as examples: Figure 7.7 is for ungrouped
data (as per Table 7.5) and Figure 7.8 is for grouped data (as per Table 7.6).
Study these two diagrams so that you are quite sure that you know how to draw them. There
is only one point which you might be tempted to overlook in the case of the grouped
distribution – the points are plotted at the ends of the class intervals and not at the centre
point. Look at the example Table 7.6 and Figure 7.8 and see how the 168,000 point is plotted
against the upper end of the 5,601 – 6,000 group (or class interval) and not against the
midpoint, 5,750. If we had been plotting a "or more" ogive, the plotting would have been
against the lower end of the group (or class interval).
Table 7.5: Time taken for one employee to perform task
Length of time
(minutes)
Frequency
Cumulative
frequency
3.4 2 2
3.5 3 5
3.6 4 9
3.7 5 14
3.8 4 18
3.9 2 20
Table 7.6: Annual income statistics
Annual income
to nearest £
Frequency
Cumulative
frequency
4,801 – 5,000 50,000 50,000
5,001 – 5,200 40,000 90,000
5,201 – 5,600 55,000 145,000
5,601 – 6,000 23,000 168,000
6,001 – 6,400 12,000 180,000
6,401 – 7,200 12,000 192,000
Total 192,000 –
Graphical Representation of Information 201
Figure 7.7: Ogive to show length of time taken by operator to complete a given task
Figure 7.8: Ogive to show annual income distribution
0
5
10
15
20
3.2 3.4 3.6 3.8 4
Cumulative
frequency
(less than or
equal to)
Length of time (mins)
0
50
100
150
200
4.8 5.2 5.6 6.0 6.4 6.8 7.2
Less than
cumulative
frequency
(000s)
Annual income (£000s)
202 Graphical Representation of Information
As an example of an "or more" ogive, we can compile the data from our example of annual
income statistics as a "more than" cumulative frequency, shown in Table 7.7.
Table 7.7: Annual income statistics
Annual income
to nearest £
Frequency
Cumulative
frequency
(more than)
4,801 – 5,000 50,000 192,000
5,001 – 5,200 40,000 142,000
5,201 – 5,600 55,000 102,000
5,601 – 6,000 23,000 47,000
6,001 – 6,400 12,000 24,000
6,401 – 7,200 12,000 12,000
Total 192,000 –
The ogive for this now appears as shown in Figure 7.9.
Figure 7.9: Ogive to show annual income distribution
Check that you see how the plotting has been made against the lower end of the group and
notice how the ogive has a reversed shape.
In each of Figures 7.8 and 7.9 we have added a fictitious group of zero frequency at one end
of the distribution.
0
50
100
150
200
4.8 5.2 5.6 6.0 6.4 6.8 7.2
More than
cumulative
frequency
(000s)
Annual income (£000s)
Graphical Representation of Information 203
It is common practice to call the cumulative frequency graph a "cumulative frequency
polygon" if the points are joined by straight lines, and a "cumulative frequency curve" if the
points are joined by a smooth curve.
(NB. Unless you are told otherwise, always compile a "less than" cumulative frequency.)
Of course all of these diagrams may be drawn from the original figures or on the basis of
relative frequencies. In more advanced statistical work the latter are used almost exclusively,
and you should practise using relative frequencies whenever possible.
Stem and Leaf Diagram
A stem and leaf diagram is the final method of presenting frequency distributions graphically
that we will study in this section. It differs from the frequency dot diagram, frequency bar
chart, frequency polygon, histogram and ogive that we have already studied in that it has the
advantage of preserving the original data. Thus, in addition to providing a visual
representation of the data, stem and leaf diagrams can also allow further analysis of the data
as it is possible to obtain the original data points from the diagram (i.e. there is no loss of
information).
Similar to the histogram, a stem and leaf diagram is a way of grouping your original data into
classes. Like the histogram, it provides a partial sorting of the original data to allow a visual
identification of the distribution pattern of the data. Unlike the histogram, it provides extra
detail regarding individual values.
In a stem and leaf diagram, each data value is split into a stem and a leaf. The leaf is usually
the last digit of the number and the other digits to the left of the leaf form the stem. The stem
and leaf diagram is plotted by listing the first digits (i.e. the "stems") in a column (and sorted
in numerical order), then listing the second digit (i.e. the "leaves") in a second column to the
right of the appropriate "stem" (and again sorted in numerical order).
It is probably easiest to understand the stem and leaf diagram through an example, so let us
construct a stem and leaf diagram using the following data of a student's exam marks:
79 80 82 68 65 59
61 57 50 62 61 70
69 64 67 70 62 65
65 73 76 87 80 82
83 79 77 71 80 77
Before you begin, you might want to arrange the data in numerical order, but is not a required
step as ordering can be done later. But what is necessary is that you separate each number
into a stem and a leaf. Since these are two digit numbers, the tens digit is the stem and the
units digit is the leaf (e.g. for the number 79, the stem is "7" and the leaf is "9").
Next, group the numbers with the same stems (e.g. 79 and 77 are in the same group as both
have a stem of 7). Then, list the stems in numerical order (in our example, the stems range
from 5 to 8 as the lowest value is 50 and the highest value is 87) in a column. In a second
column list the leaves to the right of the appropriate stem values. If your leaf values are not in
increasing order, order them now.) Finally, add a title to your diagram. If you have
successfully followed these steps, then you should have finished up with a stem and leaf
diagram identical to the one shown in Figure 7.10. As you will see, for the stem value of 5 we
have leaf values of 0, 7 and 9, which correspond to the original values of 50, 57 and 59.
204 Graphical Representation of Information
Figure 7.10: Stem and leaf diagram to show the distribution of exam marks
Stem Leaf
5 0 7 9
6 1 1 2 2 4 5 5 5 7 8 9
7 0 0 1 3 6 7 7 9 9
8 0 0 0 2 2 3 7
Key: 50 means 50 marks
B. PICTOGRAMS
Pictograms are the simplest method of presenting information visually. These diagrams are
variously called "pictograms", "ideograms", or "picturegrams" – the words all refer to the
same thing. Their use is confined to the simplified presentation of statistical data for the
general public.
Pictograms consist of simple pictures which represent quantities. There are two types and
these are illustrated in the following examples. The data we will use is shown in Table 7.8.
Table 7.8: Cruises organised by a shipping line between Year 1 and Year 3
Year Number of
cruises
No. of passengers
carried
1 200 100,000
2 250 140,000
3 300 180,000
Limited Form
We could represent the number of cruises by showing pictures of ships of varying size, as in
Figure 7.11.
Figure 7.11: Number of cruises Years 1-3
(Source: Table 7.8)
Year 1
Year 2
Year 3
200 cruises
250 cruises
300 cruises
Graphical Representation of Information 205
Although these diagrams show that the number of cruises has increased each year, they can
give false impressions of the actual increases. The reader can become confused as to
whether the quantity is represented by the length or height of the pictograms, their area on
the paper, or the volume of the object they represent. It is difficult to judge what increase has
taken place. (Sometimes you will even find pictograms in which the sizes shown are actually
wrong in relation to the real increases.)
Accurate Form
If the purpose of the pictogram is to covey anything but the most minimal of information, to
avoid confusion it is better to try and give it a degree of accuracy in representing the
numerical data. This is usually done by representing specific quantities of the variable with
pictures or graphics of some sort – usually related to the subject of the variable.
So, we could represent the numbers of passengers carried by using pictures of people as in
Figure 7.12.
Figure 7.12: Number of passengers carried Years 1-3
(Source: Table 7.8)
Each matchstick man is the same height and represents 20,000 passengers, so there can be
no confusion over size.
These diagrams have no purpose other than generally presenting statistics in a simple way.
They are not capable of providing real accurate detail. Consider Figure 7.13 – it is difficult to
represent a quantity of less than 10m barrels – so we are left wondering what "[" means.
Figure 7.13: Imports of crude oil
Year 1:
Year 2:
Year 3
Key:  20,000 passengers
Year 2:
Year 1:
Key: represents 10m barrels of oil
206 Graphical Representation of Information
C. PIE CHARTS
Pie charts, known also as circular diagrams, are used to show the manner in which various
components add up to a total. Like pictograms, they are only used to display very simple
information to non-expert readers. They are popular in computer graphics.
Suppose that we wish to illustrate the sales of gas in Great Britain in a certain year. The data
is taken from the Annual Abstract of Statistics as shown in Table 7.9.
Table 7.9: Gas sales in Great Britain in one year
(Source: Annual Abstract of Statistics)
Uses Million therms %
Domestic 1,383 51
Industrial 843 31
Commercial 437 16
Public* 60 2
Total 2,723 100
* Central and local government uses, including public lighting
The data is illustrated as a pie chart or circular diagram in Figure 7.14.
Figure 7.14: Example of a pie chart – gas sales in Great Britain
(Source: Table 7.9)
There are a number of simple steps to creating a pie chart.
(a) Tabulate the data and calculate the percentages.
(b) Convert the percentages into degrees, e.g.
51% of 360 
100
51
 360  183.6
(c) Construct the diagram by means of a pair of compasses and a protractor. Don't
overlook this point, because inaccurate and roughly drawn diagrams do not convey
information effectively.
Commercial
437m therms
Domestic
1,383m therms
Industrial
843m therms
Public
60m therms
Graphical Representation of Information 207
0
2
4
6
8
10
12
14
16
Year 1 Year 2
(d) Label the diagram clearly, using a separate legend or key if necessary. (A key is
illustrated in Figure 7.15 below.) You can insert the exact values of the data for each
segment if more detail is required, since it is not possible to read this exactly from the
diagram itself.
(e) If you have the choice, don't use a diagram of this kind with more than four or five
component parts.
The main use of a pie chart is to show the relationship each component part bears to the
whole. Pie charts are sometimes used side by side to provide comparisons, but this is not
really to be recommended. Unless the whole diagram in each case represents exactly the
same total amount, other diagrams such as bar charts (see next section) are much clearer.
D. BAR CHARTS
We have already met one kind of bar chart in the course of our studies of frequency
distributions, namely the frequency bar chart. The bars there were simply thick lines
representing, by their length, the frequencies of different values of the variate. However the
idea of a bar chart can be extended beyond the field of frequency distributions, and we will
now illustrate a number of the types of bar chart in common use.
We say "illustrate" because there are no rigid and fixed types, but only general ideas which
are best studied by means of examples. You can supplement the examples in this study unit
by looking at the commercial pages of newspapers and magazines.
Component Bar Chart
This first type of bar chart serves the same purpose as a circular diagram and, for that
reason, is sometimes called a "component bar diagram" (see Figure 7.15).
Figure 7.15: Component bar chart showing cost of production of ZYX Co. Ltd
Note that the lengths of the components represent the amounts, and that the components
are drawn in the same order so as to facilitate comparison. These bar charts are preferable
to circular diagrams because:
 they are easily read, even when there are many components
 they are more easily drawn
 it is easier to compare several bars side by side than several circles.
Wages
Materials
Royalties
Key:
208 Graphical Representation of Information
Bar charts with vertical bars are sometimes called "column charts", to distinguish them from
those in which the bars are horizontal (see Figure 7.16).
Figure 7.16: Horizontal bar chart of visitors arriving in the UK in one year
This figure is also an example of a percentage component bar chart, i.e. the information is
expressed in percentages rather than in actual numbers of visitors.
If you compare several percentage component bar charts, you must be careful. Each bar
chart will be the same length, as they each represent 100%, but they will not necessarily
represent the same actual quantities – for example, 50% might have been 1 million, whereas
in another year it may have been nearer to 4 million and in another to 8 million.
Horizontal Bar Chart
A typical case of presentation by a horizontal bar chart is shown in Figure 7.17. Note how a
loss is shown by drawing the bar on the other side of the zero line.
Figure 7.17: Horizontal bar chart for the So & So Company Ltd to show profits made
by branches in Year 1 and Year 2
Pie charts and bar charts are especially useful for "categorical" variables as well as for
numerical variables. The example in Figure 7.17 shows a categorical variable – i.e. the
different branches form the different categories, whereas in Figure 7.15 we have a numerical
variable, namely, time. Figure 7.17 is also an example of a multiple or compound bar chart as
there is more than one bar for each category.
European
51%
Commonwealth
23%
USA
22%
O
t
h
e
r
s
4
%
0 15 10 20 25 5 5
Profit (£000s) Loss (£000s)
London B
London A
Newcastle
Bristol
Year 1
Year 1
Key:
Graphical Representation of Information 209
E. SCATTER DIAGRAMS
So far we have learnt about various methods used to present data graphically. The final
graphical method that we shall learn about here is called a scatter diagram.
A scatter diagram (also known as a scattergraph, scattergram or scatter plot) differs
fundamentally from the other graphical forms of presenting data that we have considered so
far. Scatter diagrams are used as a visual aid to examine potential relationships between two
sets of (quantitative) variables, rather than simply providing a visual summary of one (or
more) set of variables (as in the case of frequency distribution graphs, pictograms, pie charts
and bar charts). Specifically, scatter diagrams are the appropriate method of presenting data
when you want to examine the relationship between two continuous variables, such as the
relationship that may exist between sales revenue and the level of advertising expenditure.
Scatter diagrams are plotted with one variable (e.g. sales revenue) on the x-axis (the
horizontal axis) and the other variable (e.g. level of advertising expenditure) on the y-axis
(vertical axis). It is convention to plot the independent variable on the x-axis and the
dependent variable on the y-axis. So using our example of sales revenue and the level of
advertising expenditure, we are examining whether sales revenue (the dependent variable) is
dependent on the level of advertising (the independent variable): does an increase in the
Let us look at an example. Table 7.10 presents sales data for a manufacturing company
against its expenditure on advertising over a given period. Looking at the data, it is not
immediately obviously whether a relationship exists or not and if so, whether this relationship
is particularly strong.
Table 7.10: Advertising expenditure and sales for a manufacturing company
Advertising expenditure (£000) Sales revenue (£000)
23 123
25 134
28 150
31 179
31 187
33 191
36 200
37 202
43 203
49 210
20 85
12 62
8 49
19 106
17 91
15 89
210 Graphical Representation of Information
However, if we plot this data on a scatter diagram, the relationship between the level of
advertising expenditure and sales revenue for this manufacturing company becomes
immediately clear. The scatter diagram presented in Figure 7.18 confirms that there appears
to be a positive relationship. As the company spends more on advertising, its sales revenue
increases (although sales revenue does seem to flatten out after advertising reaches
£350,000).
Figure 7.18 Scatter diagram of advertising expenditure against sales revenue for a
manufacturing company
While scatter diagrams are used to study possible relationships between two variables, it is
important to stress that these diagrams cannot prove that one variable causes the other.
However, they may indicate the existence and strength of an association between two
variables. This association is known as correlation.
Thus, the relationship shown in Figure 7.18, where both variables are increasing together, is
described as positive correlation (both variables are positively correlated with each other).
In contrast, if one variable increases while the other decreases, the relationship is described
as negative correlation. For example, we would expect the efficiency of the machines in this
manufacturing company to decline as they get older, suggesting that the efficiency and age
of machinery are negatively correlated.
If the scatter diagram does not show any relationship between the variables then no
correlation exists. For example, we would not expect any correlation between the cost of raw
materials and employee output.
These three types of correlation are shown in Figure 7.19.
0
50
100
150
200
250
0 10 20 30 40 50 60
Sales
revenue
(£000)
Graphical Representation of Information 211
Figure 7.19: Scatter diagram examples showing different types of correlation
Positive correlation
Negative correlation
No correlation
 

Years of post school education
1 2
Annual
Income
(£000s)
10
20
3 4
50
40
30
0
0
250

  

Age of machinery (months)
10 20
Machinery
efficiency
(output per
day)
50
100
30 40
200
150
0
0

Employee output (items per day)
20 40
Cost of raw
materials
(£ per tonne)
20
40
60 80
100
80
60
0
0
212 Graphical Representation of Information
F. GENERAL RULES FOR GRAPHICAL PRESENTATION
There are a number of general rules which must be borne in mind when planning and using
the graphical methods covered in this unit.
 Graphs and charts must be given clear but brief titles.
 The axes of graphs must be clearly labelled, and the scales of values clearly marked.
 Diagrams should be accompanied by the original data, or at least by a reference to the
source of the data.
 Avoid excessive detail, as this defeats the object of diagrams.
 Wherever necessary, guidelines should be inserted to facilitate reading.
 Try to include the origins of scales. Obeying this rule sometimes leads to rather a waste
of paper space. In such a case the graph could be "broken" as shown in Figure 7.20,
but take care not to distort the graph by overemphasising small variations.
Figure 7.20: Sales of LMN Factory
A
Months
Sales
(£000s)
M F J M J J A S O N D
50
60
70

 

 
  
Graphical Representation of Information 213
Questions for Practice
These questions are typical of those likely to be set in an examination.
1. The following figures show the number of faulty items made on successive days by a
machine in an engineering workshop:
2 0 2 4 3 2 11 4 0 1
1 4 1 2 1 4 1 7 3 2
4 5 5 6 5 2 4 2 1 2
2 4 0 6 1 0 1 5 1 2
3 1 5 3 2 2 1 3 5 3
1 1 10 1 2 3 4 3 10 6
2 6 5 3 1 2 5 7 2 3
4 4 1 3 7 4 1 7 3 7
1 1 2 6 1 0 8 5 8 1
5 6 1 3 2 2 1 2 6 4
Tabulate these figures as a frequency distribution, giving columns containing the
frequency, the relative frequency, the cumulative frequency and the relative cumulative
frequency. Illustrate the distribution by means of a frequency bar chart of the relative
figures.
2. Illustrate the following data by means of a histogram, a frequency polygon and an
ogive:
Weekly wage (£) No. of employees
31.01 to 36.00 6
36.01 to 41.00 8
41.01 to 46.00 12
46.01 to 51.00 18
51.01 to 56.00 25
56.01 to 61.00 30
61.01 to 66.00 24
66.01 to 71.00 14
3. Construct a stem and leaf diagram to illustrate the age distribution of customers
purchasing clothing from a retail store over a one-hour period given the following data
on customer ages:
8 13 16 25 26 29 30 32 37
38 40 41 44 47 49 51 54 55
58 61 63 67 75 78 82 86 95
214 Graphical Representation of Information
4. Construct a circular diagram to illustrate the following figures:
Source Amount (£000)
Income Tax 1,863,400
Estate Duties 164,500
Customs & Excise 1,762,500
Miscellaneous 99,450
Sources of revenue in Ruritania for year ending 31 March
5. Illustrate by means of a bar chart the following data relating to the number of units sold
of a particular kind of machine:
Year Home sales Commonwealth sales Foreign sales
1 6,500 5,000 5,800
2 6,250 5,400 5,700
3 5,850 6,600 5,500
6. The unit price of a commodity and its demand over a period of time are shown in the
following table:
Price
(£ per item)
Number of items sold
(000)
1.20 41
2.39 38
3.30 34
4.31 29
5.27 26
6.29 30
7.26 25
8.24 22
9.18 19
10.16 15
Draw a scatter diagram of the data and comment on the relationship between price and
the amount sold.
Now check your answers with those given at the end of the unit.
Graphical Representation of Information 215
1.
Number of
faulty items
Number of days
with these items
Relative
frequency
Cumulative
frequency
Relative
cumulative
frequency
0 5 0.05 5 0.05
1 23 0.23 28 0.28
2 20 0.20 48 0.48
3 13 0.13 61 0.61
4 12 0.12 73 0.73
5 10 0.10 83 0.83
6 7 0.07 90 0.90
7 5 0.05 95 0.95
8 2 0.02 97 0.97
9 0 0.00 97 0.97
10 2 0.02 99 0.99
11 1 0.01 100 1.00
Total 100 1.00 – –
Frequency bar chart – faulty items produced by machine in workshop
Relative
frequency
No. of faulty items per day
0
0.05
0.1
0.15
0.2
0.25
0 1 2 3 4 5 6 7 8 9 10 11
216 Graphical Representation of Information
2. Combined histogram and frequency polygon
Workings for ogive:
Wage Group (£) Frequency Cumulative
frequency
31.01 – 36.00 6 6
36.01 – 41.00 8 14
41.01 – 46.00 12 26
46.01 – 51.00 18 44
51.01 – 56.00 25 69
56.01 – 61.00 30 99
61.01 – 66.00 24 123
66.01 – 71.00 14 137
The ogive follows:
Weekly wage (£)
No. of
employees
31 36 41 46 51 56 61 66
0
10
20
30
71
 

Graphical Representation of Information 217
Ogive for wages distribution
3. Stem and leaf diagram to show the age distribution of retail customers for a
clothing store over a one-hour period
Stem Leaf
0 8
1 3 6
2 5 6 9
3 0 2 7 8
4 0 1 4 7 9
5 1 4 5 8
6 1 3 7
7 5 8
8 2 6
9 5
Key: 08 means 8 years old
Cumulative
frequency
Weekly wage (£)
0
20
40
60
80
100
120
140
31 36 41 46 51 56 61 66 71
218 Graphical Representation of Information
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
Year 1 Year 2 Year3
4. The amounts must first be turned into percentages and degrees.
Income tax 47.9% 172
Estate duties 4.2% 15
Customs & excise 45.3% 163
Miscellaneous 2.6% 9
Total 100.0% 359
Note: The angles do not add up to 360 exactly because of rounding errors.
Sources of revenue in Ruritania
5. The bar chart is shown in below. Shading has been added for clarity. You can, of
course, draw bar charts on a percentage basis.
Bar chart of machine sales
Income tax
Customs & excise
Estate duties
Miscellaneous
Foreign sales
Home sales
Commonwealth
sales
Key:
Graphical Representation of Information 219
6. The scatter diagram showing the relationship between product price and the number of
items sold for a company follows. In general, the scatter diagram shows that as price
increases demand for the product (measured in terms of number of items sold) falls.
This suggests that there is a negative correlation between the price of this particular
item and product demand.
0
5
10
15
20
25
30
35
40
45
0 2 4 6 8 10 12
Sales
(000s)
Price (£ per item)
220 Graphical Representation of Information
221
Study Unit 8
Measures of Location
Contents Page
Introduction 222
A. The Arithmetic Mean 223
The Mean of a Simple Frequency Distribution 224
The Mean of a Grouped Frequency Distribution 225
Simplified Calculation 226
Calculating the Mean from a Stem and Leaf Diagram 230
Characteristics of the Arithmetic Mean 231
B. The Mode 232
Mode of a Simple Frequency Distribution 232
Mode of a Grouped Frequency Distribution 232
Calculating the Mode from a Stem and Leaf Diagram 234
Characteristics of the Mode 235
C. The Median 236
Median of a Simple Frequency Distribution 236
Median of a Grouped Frequency Distribution 237
Calculating the Median from a Stem and Leaf Diagram 238
Answers to Questions for Practice 243
222 Measures of Location
INTRODUCTION
We have looked at frequency distributions in detail in Study Unit 6 and you should, by means
of a quick revision, make sure that you have understood them well before proceeding.
A frequency distribution may be used to give us concise information about its variate, but
often we wish to compare two or more distributions. Consider for example the distribution of
the weights of eggs from two different breeds of poultry, which is a topic in which you would
be interested if you worked for an egg marketing company. Having weighed a large number
of eggs from each breed, we might have compiled frequency distributions and graphed the
results. The two frequency polygons might well look something like Figure 8.1.
Figure 8.1: Frequency polygon showing weight of eggs from two breeds of hen
Examining these distributions you will see that they look much alike except for one thing –
they are located on different parts of the scale. In this case the distributions overlap and,
although some eggs from Breed A are of less weight than some eggs from Breed B, eggs
from Breed A are, in general, heavier than those from Breed B.
Remember that one of the objects of statistical analysis is to condense unwieldy data so as
to make it more readily understood. Drawing the frequency curves has enabled us to make
an important general statement concerning the relative egg weights of the two breeds of
poultry, but we would now like to take the matter further. We would like to calculate some
figure which will serve to indicate the general level of the variate under discussion.
In everyday life, we commonly use such a figure when we talk about the "average" value of
something or other. We might have said, in reference to the two kinds of egg, that those from
Breed A had a higher average weight than those from Breed B. Distributions with different
averages indicate that there is a different general level of the variate in the two groups.
The single value which we use to describe the general level of the variate is called a
"measure of location" or a "measure of central tendency". There are three such measures
with which you need to be familiar:
 the arithmetic mean
 the mode
 the median.
Weight of eggs
Frequency
 

 

Breed A
Breed B
Measures of Location 223
Objectives
When you have completed this study unit you will be able to:
 calculate the arithmetic mean and mode of grouped and ungrouped distributions,
including stem and leaf diagrams
 find the median of grouped and ungrouped data (including stem and leaf diagrams) by
calculation and, where appropriate, graphical methods
 utilise the assumed mean method and the "class interval" method in calculating the
arithmetic mean
 list the relative advantages and disadvantages of the arithmetic mean, median and
mode, and so decide on the appropriate measure of location to use in common
situations.
A. THE ARITHMETIC MEAN
The arithmetic mean is what we normally think of as the "average" of a set of values. It is
obtained by adding together all the values and then dividing the total by the number of values
involved.
Take, for example, the following set of values which are the heights, in centimetres, of seven
men:
Table 8.1: Height of seven men
Man Height (cms)
A 187
B 160
C 163
D 180
E 180
F 168
G 187
Total 1225
The arithmetic mean of these heights is 1225  7  175 cms. Notice that some values occur
more than once, but we still add them all.
At this point we must introduce a little algebra. We don't always want to specify what
particular items we are discussing, whether it is heights, egg weights, wages, etc. For this
reason in general discussion we will use (as you will recall from algebra) some general letter,
usually x. Also, we indicate the sum of a number of values of x by the Greek letter sigma,
written Σ.
Thus, in our example, we may write:
Σx  1225
We shall be using sigma notation extensively in this unit, so make sure you understand the
meaning of the expression.
224 Measures of Location
We indicate the arithmetic mean by the symbol x (called "x bar") and the number of items by
the letter n. The calculation of the arithmetic mean can then be described by formula:
n
Σx
x 
or
Σx
n
1
x 
The latter expression is customary in statistical work. Applying it to the example above, we
have:
175 1225
7
1
x   ) ( cms.
You will often find the arithmetic mean simply referred to as "the mean" when there is no
chance of confusion with other means (which we are not concerned with here).
The Mean of a Simple Frequency Distribution
When there are many items (that is when n is large), the arithmetic can be eased somewhat
by forming a frequency distribution. Consider the following example.
Example 1:
Table 8.2: Height distribution data
Height in cms
(x)
No. of men at
this height
(f)
Product
(fx)
160 1 160
163 1 163
168 1 168
180 2 360
187 2 374
Total Σf  7 Σ(fx)  1225
Indicating the frequency of each value by the letter f, you can see that Σf  n and that, when
the x's are not all the separate values but only the different ones, the formula becomes:
Σf
fx Σ
x
) (

which is 1225/7  175 cms as before.
Of course, with only seven items it would not be necessary in practice to use this method.
However if we had a much larger number of items the method would save a lot of additions.
Example 2:
Consider the following table and, as practice, complete the (fx) column and calculate the
value of the arithmetic mean, x . Do this before moving on.
Measures of Location 225
Table 8.3: Data for calculation of arithmetic mean
Value of x Number of items
(f)
Product
(fx)
3 1 3
4 4 16
5 7 35
6 15
7 35
8 24
9 8
10 6
Total
You should have obtained the following answers:
the total numbers of items, Σf  100
the total product, Σ(fx)  713
the arithmetic mean,
100
713
x   7.13
Make sure that you understand everything we have covered so far. Revise it if necessary,
before going on to the next paragraph. It is most important that you do not get muddled about
calculating arithmetic means.
The Mean of a Grouped Frequency Distribution
Suppose now that you have a grouped frequency distribution. In this case, you will
remember, we do not know the actual individual values, only the groups in which they lie
(unless you have a stem and leaf diagram, which we will discuss later). How then can we
calculate the arithmetic mean? The answer is that we cannot calculate the exact value of x ,
but we can make an approximation sufficiently accurate for most statistical purposes. We do
this by assuming that all the values in any group are equal to the midpoint (or mid-value) of
that group.
The procedure is very similar to that for a simple frequency distribution (which is why we
stressed the need for revision) and is shown in this example:
226 Measures of Location
Table 8.4: Data for calculation of arithmetic mean of a grouped frequency distribution
Group Mid-value
(x)
Frequency
(f)
Product
(fx)
0 to < 10 5 3 15
10 to < 20 15 6 90
20 to < 30 25 10 250
30 to < 40 35 16 560
40 to < 50 45 9 405
50 to < 60 55 5 275
60 to < 70 65 1 65
Total Σf  50 Σ(fx)  1,660
    1,660
50
1
Σf
Σfx
x 33.2
Provided that Σf is not less than about 50 and the number of groups is not less than about
12, the arithmetic mean thus calculated is sufficiently accurate for all practical purposes.
There is one pitfall to be avoided when using this method – if all the groups should not have
the same class interval, be sure that you get the correct mid-values! Table 8.5 is part of a
larger table with varying class intervals, to illustrate the point:
Table 8.5: Data for grouped frequency distribution
Group Mid-value
(x)
0 to < 10 5
10 to < 20 15
20 to< 40 30
40 to < 60 50
60 to < 100 80
You will remember that in discussing the drawing of histograms we had to deal with the case
where the last group was not exactly specified. The same rules for drawing the histogram
apply to the calculation of the arithmetic mean.
Simplified Calculation
It is possible to simplify the arithmetic involved in calculating an arithmetic mean still further
by the following two devices:
 working from an assumed mean in the middle of one convenient class
 working in class intervals instead of in the original units.
Let us consider the first device.
Measures of Location 227
(a) Working from an assumed mean in the middle of one convenient class
If you go back to our earlier examples, you will discover after some arithmetic that if
you add up the differences in value between each reading and the true mean, then
these differences add up to zero.
Take first the height distribution (Table 8.1) considered at the start of this section:
Table 8.6: Height distribution data, x  175 cms
Man Height, x (cms) (x – x ) (cms)
A 187 12
B 160 15
C 163 12
D 180 5
E 180 5
F 168 7
G 187 12
) x Σ(x   34  34  0
That is, 0 ) x Σ(x   .
Second, consider the grouped frequency distribution (Tables 8.3, 8.4) given earlier in
this section:
Table 8.7: Data for grouped frequency distribution, x  33.2
Group Mid-value
(x)
Frequency
(f)
(x – x )  f(x – x ) 
0 to < 10 5 3 28.2 84.6
10 to < 20 15 6 18.2 109.2
20 to < 30 25 10 8.2 82
30 to < 40 35 16 1.8 28.8
40 to < 50 45 9 11.8 106.2
50 to < 60 55 5 21.8 109
60 to < 70 65 1 31.8 31.8
Total Σf  50 Σ(f(x – x ))  275.8  275.8  0
That is, 0 ) x Σ(x   .
If we take any value other than x and follow the same procedure, the sum of the
differences (sometimes called deviations) will not be zero. In our first example, let us
assume the mean to be 170 cms and label the assumed mean x
o
. The differences
between each reading and this assumed value are:
228 Measures of Location
Table 8.8: Differences between x and x
o
for height distribution data
Man Height, x (cms) d = (x – x
o
) (cms)
A 187 17
B 160 10
C 163 7
D 180 10
E 180 10
F 168 2
G 187 17
Σ(x  x
o
)  54  19  35
That is Σ(x – x
o
)  35 cms or Σd  35 cms.
We can make use of this property and use it as a short cut for finding x .
We start by choosing some value of x as an assumed mean. We try to choose it near to
where we think the true mean ( x ) will lie, and we always choose it as the midpoint of
one of the groups when we are dealing with a grouped frequency distribution.
In the example just given, the total deviation (d) does not equal zero, so 170 cannot be
the true mean. As the total deviation is positive, we must have underestimated in our
choice of x
o
, so the true mean is higher than 170. As there are seven readings, we
o
upwards by one seventh of the total deviation, i.e. by (35)  7  5.
Therefore the true value of x is:
175 5 170
7
) 35 (
170   

 cms.
We know this to be the correct answer from our earlier work.
Let us now illustrate the short cut method for the grouped frequency distribution. We
shall take x
o
as 35 as this is the mid-value in the centre of the distribution.
Table 8.9: Differences between x and x
o
for grouped frequency distribution data
Group Mid-value
(x)
Frequency
(f)
d = (x – x
o
) f(x – x
o
) = fd
0 to < 10 5 3 30 90
10 to < 20 15 6 20 120
20 to < 30 25 10 10 100
30 to < 40 35 16 0 0
40 to < 50 45 9 10 90
50 to < 60 55 5 20 100
60 to < 70 65 1 30 30
Total Σf  50 Σfd  310  220  90
This time Σfd  90.
Measures of Location 229
This means we must have overestimated x
o
, as the total deviation (Σfd) is negative. As
there are 50 readings altogether, the true mean must be th
50
1
of (90) lower than 35,
i.e.
2 . 33 8 . 1 35
50
) 90 (
35 x   

 
which is as we found previously.
(b) Working in class intervals instead of in the original units
This method of calculation can be used with a grouped frequency distribution to work in
units of the class interval instead of in the original units. In the fourth column of Table
8.9, you can see that all the deviations are multiples of 10, so we could have worked in
units of 10 throughout and then compensated for this at the end of the calculation.
Let us repeat the calculation using this method. The result (with x
o
 35) is set out in
Table 8.10.
Note that here the symbol used for the length of the class interval is c, although you
may also come across the symbol i used for this purpose.
Table 8.10: Differences between x and x
o
for grouped data using class intervals
Group Mid-value
(x)
Frequency
(f)
(x – x
o
)
d =
c
x x
o
 fd
0 to < 10 5 3 –30 3 9
10 to < 20 15 6 –20 2 12
20 to < 30 25 10 10 1 10
30 to < 40 35 16 0 0 0
40 to < 50 45 9 10 1 9
50 to < 60 55 5 20 2 10
60 to < 70 65 1 30 3 3
Total Σf  50 9 31 22
c
x x
Σ
o
    

Remembering that we need to compensate, by multiplying by 10, for working in class
interval units in the table:
10
50
) 9 (
35 x 

 
33.2 8 . 1 35
50
90
35 x      as before.
The general formula for x that applies for all grouped frequency distributions having
equal class intervals can be written as:
c
n
Σfd
x x
o
  
where d 
c
x x
o

230 Measures of Location
As we mentioned at an earlier stage, you have to be very careful if the class intervals
are unequal, because you can only use one such interval as your working unit. Table
8.11 shows how to deal with this situation.
Table 8.11: Differences between x and x
o
using unequal class intervals
Group Mid-value
(x)
d =
c
x x
o
 Frequency
(f)
Product
(fd)
0 to < 10 5 3 3 9
10 to < 20 15 2 6 12
20 to < 30 25 1 10 10
30 to < 40 35 0 16 0
40 to < 50 45 1 9 9
50 to < 70 60 2½ 6 15
Total 50 24  (31)  7
The assumed mean is 35 as before, and the working unit is a class interval of 10.
Notice how d for the last group is worked out: the midpoint is 60, which is 2½ times 10
above the assumed mean. The required arithmetic mean is, therefore:
6 . 33 4 . 1 35
50
70
35
50
10 7
35 x     

 
We have reached a slightly different figure from before because of the error introduced
by the coarser grouping in the "50 to < 70" region.
The method just described is of great importance. By using it correctly, you can often
do the calculations for very complicated-looking distributions by using mental arithmetic
and pencil and paper. With the advent of electronic calculators, the time saving on
calculations of the arithmetic mean is not great, but the method is still preferable
because:
 The numbers involved are smaller and thus you are less likely to make slips in
arithmetic.
 The method can be extended to enable us to find easily the standard deviation of
a frequency distribution (which we shall examine in Study Unit 9).
Calculating the Mean from a Stem and Leaf Diagram
You will recall that a stem and leaf diagram presents frequency distributions of grouped data
graphically. However because these diagrams preserve the original data, they can be used
for further analysis to calculate the mean relatively straightforwardly, as if the data was
ungrouped.
Consider the following example of a stem and leaf diagram in Figure 8.2, which shows the
distribution of marks awarded to an ice skater in a competition. Note that the stem values
represent the units of a mark awarded and the leaf values represent the tenths of a mark
awarded (i.e. a mark of 5.8 is expressed in the stem and leaf diagram as 5|8).
Measures of Location 231
Figure 8.2: Stem and leaf diagram to show the marks awarded to a competition skater
Stem Leaf
0 0
1
2
3 5 9
4 0 1 1 3
5 0 5 8
Note: 0|0  0.0 marks
Although the stem and leaf diagram presents grouped data, the individual data is preserved
so that the mean can be calculated using the simple formula:
n
Σx
x 
Accordingly:
10
) 8 . 5 5 . 5 0 . 5 3 . 4 1 . 4 1 . 4 0 . 4 9 . 3 5 . 3 0 . 0 ( Σ
x
        

0 . 4
10
40.2
x  
Characteristics of the Arithmetic Mean
There are a number of characteristics of the arithmetic mean which you must be aware of:
 It is not necessary to know the value of every item in a distribution in order to calculate
the arithmetic mean. Only the total and the number of items are needed. For example,
if you know the total wages bill and the number of employees, you can calculate the
arithmetic mean wage without knowing the wages of each person.
 The arithmetic mean is fully representative because it is based on all, and not just
some, of the items in the distribution.
 One or two extreme values can make the arithmetic mean somewhat unrepresentative
by their influence on it. For example, if a millionaire came to live in a country village, the
inclusion of his or her income in the arithmetic mean for the village might make the
place seem very much better off than it really was!
 The arithmetic mean is reasonably easy to calculate and to understand.
 In more advanced statistical work, it has the advantage of being amenable to algebraic
manipulation.
232 Measures of Location
B. THE MODE
The first alternative to the mean which we will discuss is the mode. This is the name given to
the most frequently occurring value.
Mode of a Simple Frequency Distribution
Consider the following frequency distribution:
Table 8.12 Accident distribution data
Number of accidents per day
(x)
Number of days with the stated
number of accidents
(f)
0 27
1 39
2 30
3 20
4 7
Total 123
In this case the most frequently occurring value is 1 (it occurred 39 times) and so the mode
of this distribution is 1.
Note that the mode, like the mean, is a value of the variate x, not the frequency of that value.
A common error is to say that the mode of the distribution in Table 8.12 is 39. This is wrong.
The mode is 1. Watch out, and do not fall into this trap!
For comparison, calculate the arithmetic mean of the distribution – it works out at 1.52.
The mode is used in those cases where it is essential for the measure of location to be an
actually occurring value. An example is the case of a survey carried out by a clothing store to
determine what size of garment to stock in the greatest quantity. The average size of
garment in demand might turn out to be, let us say, 9.3724. This is not an actually occurring
value and doesn't help us to answer our problem. However, the mode of the distribution
obtained from the survey would be an actual value (perhaps size 8) and it would provide the
Mode of a Grouped Frequency Distribution
When the data is given in the form of a grouped frequency distribution, it is not quite so easy
to determine the mode (unless you have a stem and leaf diagram, which we will discuss
later). What, you might ask, is the mode of the following distribution shown in Table 8.13?
Measures of Location 233
Table 8.13: Data for determining the mode of a grouped frequency distribution
Group Frequency
0 to < 10 4
10 to < 20 6
20 to < 30 10
30 to < 40 16
40 to < 50 24
50 to < 60 32
60 to < 70 38
70 to < 80 40
80 to < 90 20
90 to < 100 10
100 to < 110 5
110 to < 120 1
All we can really say is that "70 to < 80" is the modal group (the group with the largest
frequency).
You may be tempted to say that the mode is 75, but this is not true, nor even a useful
approximation in most cases. The reason is that the modal group depends on the method of
grouping, which can be chosen quite arbitrarily to suit our convenience. The distribution could
have been set out with class intervals of five instead of 10, and would then have appeared as
follows (only the middle part is shown, to illustrate the point):
Table 8.14: Revised data for determining the mode of a grouped frequency distribution
Group Frequency
60 to < 65 16
65 to < 70 22
70 to < 75 21
75 to < 80 19
80 to < 85 12
85 to < 90 8
The modal group is now "65 to < 70". Likewise, we will get different modal groups if the
grouping is by 15 or by 20 or by any other class interval, and so the midpoint of the modal
group is not a good way of estimating the mode.
In practical work, this determination of the modal group is usually sufficient, but you may
occasionally be asked for the mode to be determined from a grouped distribution.
A number of procedures based on the frequencies in the groups adjacent to the modal group
can be used, and we shall now describe one procedure. You should note, however, that
these procedures are only mathematical devices for finding the most likely position of the
mode – it is not possible to calculate an exact and true value in a grouped distribution.
234 Measures of Location
We saw that the modal group of our original distribution was "70 to < 80". Now examine the
groups on each side of the modal group. The group below ("60 to < 70") has a frequency of
38, and the one above ("80 to < 90") has a frequency of 20. This suggests to us that the
mode may be some way towards the lower end of the modal group rather than at the centre.
A graphical method for estimating the mode is shown in Figure 8.3.
This method can be used when the distribution has equal class intervals. Draw that part of
the histogram which covers the modal class and the adjacent classes on either side.
Draw in the diagonals AB and CD as shown in Figure 8.3. From the point of intersection draw
a vertical line downwards. Where this line crosses the horizontal axis is the mode. In our
example the mode is just less than 71.
Figure 8.3: Using a histogram to determine the most likely value of the mode
Calculating the Mode from a Stem and Leaf Diagram
Although a stem and leaf diagram presents frequency distributions of grouped data
graphically, as we have remarked the original data is preserved. This means that these
diagrams can be used to calculate the mode relatively straightforwardly as if the data was
ungrouped.
Let us consider our previous example of a stem and leaf diagram which shows the
distribution of marks awarded to an ice skater in a competition (Figure 8.4, previously as
Figure 8.2)). Remember that the stem values represent the units of a mark awarded and the
leaf values represent the tenths of a mark awarded.
Variate
Class
frequency
A
20
30
40
60 80 90 70
B
C
D
Mode
Measures of Location 235
Figure 8.4: Stem and leaf diagram to show the marks awarded to a competition skater
Stem Leaf
0 0
1
2
3 5 9
4 0 1 1 3
5 0 5 8
Note: 0|0  0.0 marks
As the mode is defined as the data value that occurs most often, all we need to do is look for
the leaf (number) that occurs the most often on one stem of the diagram. In our example,
there are two '1' leaves on the '4' stem (i.e. two data points of value 4.1), so the mode is 4.1.
Characteristics of the Mode
Some of the characteristics of the mode are worth noting, particularly as they compare with
those of the arithmetic mean.
 The mode is very easy to find with ungrouped distributions, since no calculation is
required.
 It can only be determined approximately with grouped distributions.
 It is not affected by the occurrence of extreme values.
 Unlike the arithmetic mean, it is not based on all the items in the distribution, but only
on those near its own value.
 In ungrouped distributions the mode is an actually occurring value.
 It is not amenable to the algebraic manipulation needed in advanced statistical work.
 It is not unique, i.e. there can be more than one mode. For example, in the set of
numbers, 6, 7, 7, 7, 8, 8, 9, 10, 10, 10, 12, 13, there are two modes, namely 7 and 10.
This set of numbers would be referred to as having a bimodal distribution.
 The mode may not exist. For example, in the set of numbers 7, 8, 10, 11, 12 each
number occurs only once so this distribution has no mode.
236 Measures of Location
C. THE MEDIAN
The desirable feature of any measure of location is that it should be near the middle of the
distribution to which it refers. Now, if a value is near the middle of the distribution, then we
expect about half of the distribution to have larger values, and the other half to have smaller
values. This suggests to us that a possible measure of location might be that value which is
such that exactly half (50%) of the distribution has larger values and exactly half has lower
values. The value which divides the distribution into equal parts is called the median.
Look at the following set of values:
6, 7, 7, 8, 8, 9, 10, 10, 10, 12, 13
The total of these eleven numbers is 100 and so the arithmetic mean is therefore 100/11 
9.091 (three decimal places), while the mode is 10 because that is the number which occurs
most often (three times). The median however is 9, because there are five values above and
five values below 9.
Our first rule for determining the median is, therefore, as follows:
Arrange all the values in order of magnitude and the median is then the middle value.
Note that all the values are to be used – even though some of them may be repeated, they
must all be put separately into the list. In the example just dealt with, it was easy to pick out
the middle value because there was an odd number of values. But what if there is an even
number? Then, by convention, the median is taken to be the arithmetic mean of the two
values in the middle. For example, take the following set of values:
6, 7, 7, 8, 8, 9, 10, 10, 11, 12
The two values in the middle are 8 and 9, so the median is 8½.
Median of a Simple Frequency Distribution
Of course statistical data is rarely in such small groups and, as you have already learned, we
usually deal with frequency distributions. How then do we find the median if our data is in the
form of a distribution?
Let us take the example of the frequency distribution of accidents already used in discussing
the mode (Table 8.13). The total number of values is 123, and so when those values are
arranged in order of magnitude, the median will be the 62
nd
item because that will be the
middle item. To see what the value of the 62
nd
item will be, let us draw up the distribution
again. Table 8.15 repeats the data from Table 8.12, but adds a column setting out the
cumulative frequency.
Table 8.15: Accident distribution data
Number of accidents
per day
(x)
Number of days with the
stated number of accidents
(f)
Cumulative
frequency
(F)
0 27 27
1 39 66
2 30 96
3 20 116
4 7 123
Total 123
Measures of Location 237
You can see from the last column that, if we were to list all the separate values in order, the
first 27 would all be 0's and from then up to the 66
th
would be 1's. It therefore follows that the
62
nd
item would be a 1 and that the median of this distribution is 1.
Median of a Grouped Frequency Distribution
The final problem connected with the median is how to find it when our data is in the form of
a grouped distribution. The solution to the problem, as you might expect, is very similar to the
solution for an ungrouped distribution – we halve the total frequency and then find, from the
cumulative frequency column, the corresponding value of the variate.
Because a grouped frequency distribution nearly always has a large total frequency, and
because we do not know the exact values of the items in each group, it is not necessary to
find the two middle items when the total frequency is even – just halve the total frequency
and use the answer (whether it is a whole number or not) for the subsequent calculation.
Table 8.16: Data for determining the median of a grouped frequency distribution
Group Frequency
(f)
Cumulative frequency
(F)
0 to < 10 4 4
10 to < 20 6 10
20 to < 30 10 20
30 to < 40 16 36
40 to < 50 24 60
50 to < 60 32 92
60 to < 70 38 130
70 to < 80 40 170
80 to < 90 20 190
90 to < 100 10 200
100 to < 110 5 205
110 to < 120 1 206
Total 206
The total frequency is 206 and therefore the median is the 103
rd
item which, from the
cumulative frequency column, must lie in the "60 to < 70" group. But exactly where in the "60
to < 70" group? Well, there are 92 items before we get to that group and we need the 103
rd
item, so we obviously need to move into that group by 11 items. Altogether in our "60 to < 70"
group there are 38 items, so we need to move 11/38 of the way into that group, that is 11/38
of 10 above 60. Our median is therefore:
60 
38
110
 60  2.89  62.89
The use of the cumulative frequency distribution will no doubt remind you of its graphical
representation, the ogive. In practice, a convenient way to find the median of a grouped
distribution is to draw the ogive. Then, against a cumulative frequency of half the total
frequency, read off the median. In our example the median would be read against 103 on the
cumulative frequency scale (see Figure 8.5). If the ogive is drawn with relative frequencies,
then the median is always read off against 50%.
238 Measures of Location
Figure 8.5: Reading the median from an ogive
Calculating the Median from a Stem and Leaf Diagram
Similarly, we can use a stem and leaf diagram to calculate the median value of a distribution
of grouped data as the original data is preserved. Again, let us consider our previous
example of a stem and leaf diagram which shows the distribution of marks awarded to an ice
skater in a competition (Figure 8.6). Again, remember that the stem values represent the
units of a mark awarded and the leaf values represent the tenths of a mark awarded.
Figure 8.6: Stem and leaf diagram to show the marks awarded to a competition skater
Stem Leaf
0 0
1
2
3 5 9
4 0 1 1 3
5 0 5 8
Note: 0|0  0.0 marks
As noted, the stem and leaf diagram presents grouped data, but the individual data is
preserved. This means that the median can be calculated relatively straightforwardly. As the
median is the middle value in the set, we can simply go through the data presented in our
stem and leaf diagram and cross off pairs of high and low leaves. If we start with the
0
20
40
60
80
100
120
140
160
180
200
220
0 20 40 60 80 100 120
Value of the variate
Cumulative
frequency
Median  63
103
Measures of Location 239
rightmost leaf on the bottom stem and the leftmost leaf on the top stem, we can work our way
towards the middle of the values crossing off the pairs until we have only one (or two)
leave(s) left that have not been crossed out. That value (or the average of the two values) is
the thus median.
In our example, we are left with two middle values, namely the two '1' leaves on the '4' stem.
The average of these values, and hence the median, is thus 4.1.
Characteristic features of the median, which you should compare with those of the mean and
the mode, are as follows:
 It is fairly easily obtained in most cases, and is readily understood as being the "half-
way point".
 It is less affected by extreme values than the mean. The millionaire in the country
village might alter considerably the mean income of the village but he or she would
have almost no effect at all on the median.
 It can be obtained without actually having all the values. For example, if we want to
know the median height of a group of 21 men, we do not have to measure the height of
every single one. It is only necessary to stand the men in order of their heights and
then only the middle one (number 11) need be measured, for his height will be the
median height. Thus median is of value when we have open-ended classes at the
edges of the distribution, as its calculation does not depend on the precise values of
the variate in these classes, whereas the value of the arithmetic mean does.
 The median is not very amenable to further algebraic manipulation.
240 Measures of Location
Questions for Practice
1. The following table shows the consumption of electricity of 100 householders during a
particular week. Calculate the arithmetic mean consumption of the 100 householders.
Consumption of electricity of 100 householders (one week)
Consumption
(kilowatt hours)
Number of
householders
0 – under 10 5
10 – under 20 8
20 – under 30 20
30 – under 40 29
40 – under 50 20
50 – under 60 11
60 – under 70 6
70 – under 80 1
80 or over 0
2. As part of a health study the heights of several dozen men in a particular location are
measured as a sample. After this has been done, it is found that the sample contains
two brothers who are dwarfs. What sort of "average" would you calculate from the
sample, and why?
3. The rate of effluent discharge from two plants based on different systems was sampled
at four-hourly intervals and the results are given below:
System A
10 12 10 11 9 12 14 8 9 12 7 13 10 12 11
13 8 10 9 12 13 10 16 12 7 6 5 10 11 14
System B
10 17 18 6 5 10 11 13 6 15 14 10 12 9 8
7 4 18 19 10 12 13 16 9 12 3 4 12 4 9
Calculate the mean rate of discharge for each system.
4. Using the short-cut methods given in this study unit, calculate the arithmetic mean (i.e.
the average) size of shareholding, from the grouped distribution in the following table.
Measures of Location 241
Distribution of shareholding
Number of
shares
Number of persons with
this size of holding
Less than 50 23
50 – 99 55
100 – 199 122
200 – 299 136
300 – 399 100
400 – 499 95
500 – 599 81
600 – 699 63
700 – 799 17
800 – 1,000 8
5. A garage keeps two cars for daily hire. The number of cars asked for in a day has
varied as follows:
Cars wanted in a day Proportion of days
0 0.36
1 0.38
2 0.18
3 0.06
4 0.02
5 or more 0
(a) What is the mean daily demand?
(b) What is the mean number supplied (i.e. hired out) per day?
(c) On what fraction of days is some demand turned away?
6. The following distribution gives the weight in kilograms of 100 schoolboys. Find the
median weight.
Weight in kg No. of schoolboys
45 – 49 2
50 – 54 15
55 – 59 37
60 – 64 31
65 – 69 11
70 – 74 4
100
242 Measures of Location
7. The holdings of shares (nominal value £1) in DEF plc are given in the following table.
Calculate (a) the arithmetic mean holding, (b) the median holding and (c) the modal
holding.
No. of shares held No. of persons
with these shares
No. of shares held No. of persons
with these shares
Under 50 80 300 – 349 28
50 – 99 142 350 – 399 15
100 – 149 110 400 – 449 9
150 – 199 75 450 – 499 7
200 – 249 52 500 – 549 3
250 – 299 38 550 – 599 2
8. The age distribution of employees working for a manufacturing company is given in the
following stem and leaf diagram.
Stem Leaf
1 8 8 9
2 0 3 5 5 7 9
3 1 3 8
4 2 3 5
Note: 18  18 years old
Based on the data presented in this diagram, calculate the:
(a) mean
(b) median
(c) mode.
Now check your answers with those given at the end of the unit.
Measures of Location 243
1. Consumption of electricity of 100 householders (one week)
Consumption
(kwH)
Mid-value
(x)
Number of
householders
(f)
fx
0 – under 10 5 5 25
10 – under 20 15 8 120
20 – under 30 25 20 500
30 – under 40 35 29 1,015
40 – under 50 45 20 900
50 – under 60 55 11 605
60 – under 70 65 6 390
70 – under 80 75 1 75
80 or over 0 0
100 3,630
x  3 36
100
630 3
f Σ
fx Σ
.
,
 
i.e. the arithmetic mean of the householders' consumption is 36.3 kilowatt hours.
2. In order to give the correct impression of the mean in the sample location, it would be
best to use the median. Normally one thinks of the arithmetic mean as the best
measure of average, but in this case it would be too much affected by the two
extremely low values to be representative. The mode would not be so affected, but with
such a small sample (a dozen or so) it would be erratic, or might not exist if all the
heights were different.
3. You can group the data first if you wish, but it is probably quicker just to add the
numbers as they stand.
System A:
total number of readings  30
total of rates of discharge  316
mean rate of discharge  53 . 10
30
316

System B:
total number of readings  30
total of rates of discharge  316
mean rate of discharge  53 . 10
30
316

244 Measures of Location
4. Take c  100, the most common class interval, and x
o
 250, the middle of the modal
group.
Distribution of shareholding
No. of shares
Mid-value
(x)
No. of persons with
a holding of x
(f)
d =
c
x x
o

fd
Less than 50 25.0 23
2.3 -51.8
50 – 99 74.5 55 1.8 -96.5
100 – 199 149.5 122 1.0 -122.6
200 – 299 249.5 136 0.0 -0.7
300 – 399 349.5 100 1.0 99.5
400 – 499 449.5 95 2.0 189.5
500 – 599 549.5 81 3.0 242.6
600 – 699 649.5 63 4.0 251.7
700 – 799 749.5 17 5.0 85.0
800 – 1,000 900.0 8 6.5 52.0
Total 700 648.7
100
700
7 . 648
250 x   
 250  92.7  343 to the nearest whole number.
Therefore, the mean shareholding is 343 shares.
5. (a)
Demand
(x
1
)
Frequency
(as a proportion)
(f)
fx
1
0 0.36 0
1 0.38 0.38
2 0.18 0.36
3 0.06 0.18
4 0.02 0.08
5 or more 0 0
1.00 1.00
Mean demand  1
00 . 1
00 . 1
Σf
Σfx
 
1
car per day
Measures of Location 245
(b)
Supply
(x
2
)
Frequency
(as a proportion)
(f)
fx
2
0 0.36 0
1 0.38 0.38
2 (0.18  0.06  0.02) 0.26 0.52
1.00 0.90
Mean supply  90 . 0
00 . 1
90 . 0
Σf
Σfx
2
  cars per day
(c) Demand is turned away whenever three or more cars are requested in one day.
This proportion of days is 0.06  0.02  0.08. Therefore, on eight days in every
100 demand is turned away.
6.
Weight in kg No. of schoolboys Cumulative frequency
45 – 49 2 2
50 – 54 15 17
55 – 59 37 54
60 – 64 31 85
65 – 69 11 96
70 – 74 4 100
100
Half the total frequency is 50 so the median lies in the "55 – 59 kg" group, and in the
(50  17)  33
rd
item within it.
Median weight  5
37
33
55  
 55  4.46
 59.46
 60 kg to the nearest kg.
246 Measures of Location
7. Take c  50, x
o
 224.5
Distribution of the sizes of shareholdings in DEF plc
Number of
shares held
Mid-value
No. of persons
with these
shares
d =
c
x x
o

fd
Cumulative
frequency
Under 50 24.5 80 4 320 80
50 – 99 74.5 142 3 426 222
100 – 149 124.5 110 2 220 332
150 – 199 174.5 75 1 75 407
200 – 249 224.5 52 0 0 459
250 – 299 274.5 38 1 38 497
300 – 349 324.5 28 2 56 525
350 – 399 374.5 15 3 45 540
400 – 449 424.5 9 4 36 549
450 – 499 474.5 7 5 35 556
500 – 549 524.5 3 6 18 559
550 – 599 574.5 2 7 14 561
Total 561 1,041
242
 799
(a) x  224.5  50
561
799

 224.5  (1.424  50)
 224.5  71.2  153.3  153 shares to the nearest whole number.
(N.B. You may have chosen a different x
o
, but the result will be the same.)
(b) Half of the total frequency is 280.5 and therefore the median lies in the "100 –
149" group and it is the 58.5
th
(280.5  222) item within it.
Median  99.5  1 . 126
110
50 5 . 58

 126 shares to the nearest whole number.
(c) The modal group is the "50 – 99" group and the estimated mode is given by
drawing this group, together with the surrounding class groups, on a histogram.
This is shown on the next page.
Reading from this, the estimated mode  83 shares.
Measures of Location 247
8. (a) x 
15
45) 43 42 38 33 31 29 27 25 25 23 20 19 18 (18
Σf
Σfx              

 29.1
i.e. the arithmetic mean age of manufacturing employees is 29.1 years.
(b) The median age of manufacturing employees is 27 years.
(c) In our example, there are two 8 leaves on the 1 stem (i.e. two data points of value
18) and two 5 leaves on the 2 stem (i.e. two data points of value 25). Thus, the
data set is bimodal with the mode age of manufacturing employees being 18
years and 25 years.
Shareholding
Frequency
(no of
persons)
80
140
100 150
50
Under
50
group
120
100
100 – 50
group
Mode  83
248 Measures of Location
249
Study Unit 9
Measures of Dispersion
Contents Page
Introduction 250
A. Range 251
B. The Quartile Deviation, Deciles and Percentiles 251
The Quartile Deviation 251
Calculation of the Quartile Deviation 253
Deciles and Percentiles 254
C. Standard Deviation 255
The Variance and Standard Deviation 255
Alternative Formulae for Standard Deviation and Variance 257
Standard Deviation of a Simple Frequency Distribution 258
Standard Deviation of a Grouped Frequency Distribution 259
Characteristics of the Standard Deviation 262
D. The Coefficient of Variation 262
E. Skewness 263
Answers to Questions for Practice 268
250 Measures of Dispersion
INTRODUCTION
In order to get an idea of the general level of values in a frequency distribution, we have
studied the various measures of location that are available. However, the figures which go to
make up a distribution may all be very close to the central value, or they may be widely
dispersed about it – for example, the mean of 49 and 51 is 50, but the mean of 0 and 100 is
also 50. You can see that these two distributions may have the same mean, but the individual
When applying statistical methods to practical problems, knowledge of this spread, which we
call "dispersion" or "variation", is of great importance. Examine the figures in the following
table:
Table 9.1: Output from two factories over 10 weeks
Weekly output
Week
Factory A Factory B
1 94 136
2 100 92
3 106 110
4 100 36
5 90 102
6 101 57
7 107 108
8 98 81
9 101 156
10 98 117
Total 995 995
Mean output 99.5 99.5
Although the two factories have the same mean output, they are very different in their week-
to-week consistency. Factory A achieves its mean production with only very little variation
from week to week, whereas Factory B achieves the same mean by erratic ups and downs
from week to week. This example shows that a mean (or other measure of location) does not
by itself tell the whole story. We therefore need to supplement it with a measure of
dispersion.
As was the case with measures of location, there are several different measures of
dispersion in use by statisticians. Each has its own particular merits and demerits, which will
be discussed later. The measures in common use are:
 range
 quartile deviation
 standard deviation.
We will discuss each of these in this unit.
Measures of Dispersion 251
Objectives
When you have completed this study unit you will be able to:
 calculate the range of a distribution from given data
 find, by both calculation and graphical methods, the quartile deviation, deciles and
percentiles of grouped data
 calculate the standard deviation and variance of grouped and ungrouped distributions,
using the appropriate formulae
 calculate the standard deviation using the "corrected/approximate variance" method
 calculate the coefficient of variation and explain its use
 calculate Pearson's first and second coefficient of skewness from the appropriate data
 explain the nature of skewness and describe the relationship between the mean,
median and mode in a skewed distribution
 identify the advantages and disadvantages of the range, quartile deviation and
standard deviation as measures of dispersion.
A. RANGE
The range is the simplest measure of dispersion: it is simply the difference between the
largest and the smallest. In the example just given, we can see that the lowest weekly output
for Factory A was 90 and the highest was 107; the range is therefore 107 ÷90, i.e.17. For
Factory B the range is 156 ÷ 36 = 120. The larger range for Factory B shows that it performs
less consistently than Factory A.
The advantage of the range as a measure of the dispersion of a distribution is that it is very
easy to calculate and its meaning is easy to understand. For these reasons it is used a great
deal in industrial quality control work. Its disadvantage is that it is based on only two of the
individual values, and takes no account of all those in between. As a result, one or two
extreme results can make it quite unrepresentative. Consequently, the range is not much
used in statistical analysis except in the case just mentioned.
B. THE QUARTILE DEVIATION, DECILES AND
PERCENTILES
The Quartile Deviation
This measure of dispersion is sometimes called the "semi-interquartile range".
To understand it, you must cast your mind back to the method of obtaining the median from
the ogive. The median, you remember, is the value which divides the total frequency into two
halves. The values which divide the total frequency into quarters are called quartiles and they
can also be found from the ogive, as shown in Figure 9.1.
252 Measures of Dispersion
Figure 9.1: Identifying quartiles from an ogive
This is the same ogive that we drew in Study Unit 8 when finding the median of a grouped
frequency distribution.
You will notice that we have added the relative cumulative frequency scale to the right of the
graph. 100% corresponds to 206, i.e. the total frequency. It is then easy to read off the values
of the variate corresponding to 25%, 50% and 75% of the cumulative frequency, giving the
lower quartile (Q
1
), the median and the upper quartile (Q
3
) respectively.
Q
1
= 46.5
median = 63 (as found previously)
Q
3
= 76
The difference between the two quartiles is the interquartile range and half of the difference
is the semi-interquartile range or quartile deviation:
Quartile deviation =
2
Q Q
1 3
÷
=
2
5 . 46 76 ÷
=
2
5 . 29
= 14.75
0
20
40
60
80
100
120
140
160
180
200
220
0 20 40 60 80 100 120
Value of the variate
Cumulative
frequency
Relative
cumulative
frequency
(%)
100
75
50
25
Measures of Dispersion 253
Alternatively, you can work out:
25% of the total frequency, i.e. , 5 . 51
4
206
= and
75% of the total frequency, i.e. 154.5
and then read from the ogive the values of the variate corresponding to 51.5 and 154.5 on
the cumulative frequency scale (i.e. the left-hand scale). The end result is the same.
Calculation of the Quartile Deviation
The quartile deviation is not difficult to calculate in situations where the graphical method is
not acceptable. Remember that graphical methods are never quite as accurate as
calculations.
We shall again use the same example, and the table of values is reproduced for convenience
as Table 9.2:
Table 9.2: Data for finding the quartile deviation of a grouped frequency distribution
Group Frequency
(f)
Cumulative
frequency
(F)
0 to < 10 4 4
10 to < 20 6 10
20 to < 30 10 20
30 to < 40 16 36
40 to < 50 24 60
50 to < 60 32 92
60 to < 70 38 130
70 to < 80 40 170
80 to < 90 20 190
90 to < 100 10 200
100 to < 110 5 205
110 to < 120 1 206
Total 206
The lower quartile is the 25% point in the total distribution and the upper quartile the 75%
point.
As the total frequency is 206:
the 25% point =
4
206
= 51½ and
the 75% point = 206
4
3
× = 154½.
We can make the calculations in exactly the same manner as we used for calculating the
median.
254 Measures of Dispersion
Looking at Table 9.2, the 51½
th
item comes in the "40 to < 50" group and will be the (51½ ÷
36) = 15½
th
item within it.
So, the lower quartile = 40 + 10
24
½ 15
×
= 40 + 6.458
= 46.458
Similarly, the upper quartile will be the 154½
th
item which is in the "70 to < 80" group and is
the (154½ ÷ 130) = 24½
th
item within it.
So, the upper quartile = 70 + 10
40
½ 24
×
= 70 + 6.125
= 76.125
Remember that the units of the quartiles and of the median are the same as those of the
variate.
The quartile deviation =
2
Q Q
1 3
÷
=
2
458 . 46 125 . 76 ÷
=
2
667 . 29
= 14.8335
= 14.8 to 1 decimal place.
The quartile deviation is unaffected by an occasional extreme value. However it is not based
on the values of all the items in the distribution and, to this extent, it is less representative
than the standard deviation (which we discuss later). In general, when a median is the
appropriate measure of location, then the quartile deviation should be used as the measure
of dispersion.
Deciles and Percentiles
It is sometimes convenient, particularly when dealing with wages and employment statistics,
to consider values similar to the quartiles, but which divide the distribution more finely. Such
partition values are deciles and percentiles. From their names you will probably have
guessed that the deciles are the values which divide the total frequency into tenths and the
percentiles are the values which divide the total frequency into hundredths. Obviously it is
only meaningful to consider such values when we have a large total frequency.
The deciles are labelled D
1
, D
2
... D
9
. For instance, the second decile (D
2
) is the value below
which 20% of the data lies, and the sixth decile (D
6
) is the value below which 60% of the data
lies.
The percentiles are labelled P
1
, P
2
... P
99
. Again we can say that, for example, P
5
is the value
below which 5% of the data lies and P
64
is the value below which 64% of the data lies.
Using the same example as previously, let us calculate as an illustration the third decile D
3
.
The method follows exactly the same principles as the calculation of the median and
quartiles.
Measures of Dispersion 255
The total frequency is 206 and D
3
is the value below which 30% of the data lies:
30% of 206 is 8 . 61 206
100
30
= ×
We are therefore looking for the value of the 61.8
th
item. A glance at the cumulative
frequency column shows that the 61.8
th
item lies in the "50 to < 60" group, and is the (61.8 ÷
60) = 1.8
th
item within it. So:
D
3
= 50 + 10
32
8 . 1
×
= 50 +
32
18
= 50.6 to one decimal place.
Therefore 30% of our data lies below 50.6.
We could also have found this result graphically – check that you agree with the calculation
3
from the graph in Figure 9.1. You will see that the calculation method enables
us to give a more precise answer than is obtainable graphically.
C. STANDARD DEVIATION
The most important of the measures of dispersion is the standard deviation. Except for the
use of the range in statistical quality control and the use of the quartile deviation in wages
statistics, the standard deviation is used almost exclusively in statistical practice.
The standard deviation is defined as the square root of the variance, and so we need to know
how to calculate the variance first.
The Variance and Standard Deviation
We start by finding the deviations from the mean, and then squaring them, which removes
the negative signs in a mathematically acceptable fashion. The variance is then defined as
the mean of the deviations from the mean, squared:
variance =
n
) x x ( Σ
2
÷
, where n is the total frequency.
Consider the following example in Table 9.3.
256 Measures of Dispersion
Table 9.3: Output from Factory A over 10 weeks
Week
Weekly output
(x)
Deviation
) ( x x ÷
Deviation squared
2
x x ) ( ÷
1 94 ÷5.5 30.25
2 100 +0.5 0.25
3 106 +6.5 42.25
4 100 +0.5 0.25
5 90 ÷9.5 90.25
6 101 +1.5 2.25
7 107 +7.5 56.25
8 98 ÷1.5 2.25
9 101 +1.5 2.25
10 98 ÷1.5 2.25
Total 995 +18.0
÷18.0
228.50
Mean output 99.5 Variance = 22.85
The mean of the squared deviations is the variance:
variance =
n
x x Σ
2
) ( ÷
=
10
28.50 2
= 22.85
The formula for the standard deviation, which is defined as the square root of the variance,
is:
SD =
n
) x x Σ
2
÷ (
, again where n is the total frequency.
In this example, then:
SD = 85 . 22 = 4.78.
The symbol usually used to denote standard deviation is "σ" (theta), although you may
sometimes find "s" or "sd" used.
Measures of Dispersion 257
Alternative Formulae for Standard Deviation and Variance
The formula we have used to calculate the standard deviation is:
n
) x Σ(x
σ
2
÷
= Formula (a)
where:
x = value of the observations
x = mean of the observations
n = number of observations.
As the standard deviation is defined as the square root of the variance, therefore:
variance = σ
2
=
n
) x x ( Σ
2
÷
By expanding the expression Σ(x ÷ x )
2
we can rewrite the formula for standard deviation in
two alternative and sometimes more useful forms, as follows:
2
2
x
n
Σx
σ ÷ =
Formula (b)
or
2
2
n
x Σ
n
x Σ
σ |
.
|

\
|
÷ =
Formula (c)
The choice of the formula to use depends on the information that you already have and the
information that you are asked to give. In any calculation, you should aim to keep the
arithmetic as simple as possible and the rounding errors as small as possible.
 If you already know x and it is a small integer, there is no reason why you should not
use formula (a).
 If you already know x but it is not an integer (as in the previous example), then formula
(b) is the best to use.
 If you do not know x , then you should use formula (c) – particularly if you are not
When you are dealing with data in the form of a simple or grouped frequency distribution,
then the formula for calculating the standard deviation needs to be expressed slightly
differently:
Σf
) x x f Σ
σ
2
÷
=
(
Formula (d)
or
|
|
.
|

\
|
÷ =
n
) fx Σ (
fx Σ
n
1
σ
2
2
, where n = Σf Formula (e)
Formula (d), like formula (a), is derived directly from the definition of the standard deviation.
Formula (e) is obtained by using the sigma notation as in formula (c) and this is the one to
use in calculations, as shown in the following sections.
258 Measures of Dispersion
Standard Deviation of a Simple Frequency Distribution
If the data in Table 9.3 had been given as a frequency distribution (as is often the case), then
only the different values would appear in the "x" column and we would have to remember to
multiply each result by its frequency. This is shown in Table 9.4 where we set out the
information for calculating the standard deviation by applying formula (d):
Table 9.4: Determining the standard deviation of a simple frequency distribution
Weekly
output
(x)
Frequency
(f)
fx
Deviation
) ( x x ÷
Deviation
squared
2
x x ) ( ÷
f(x – x )
2
90 1 90 ÷9.5 90.25 90.25
94 1 94 ÷5.5 30.25 30.25
98 2 196 ÷1.5 2.25 4.50
100 2 200 +0.5 0.25 0.50
101 2 202 +1.5 2.25 4.50
106 1 106 +6.5 42.25 42.25
107 1 107 +7.5 56.25 56.25
Total 10 995 228.50
Mean 99.5 Variance = 22.85
The standard deviation is then worked out in the same way as previously, as follows:
Σf
) x x f Σ
σ
2
÷
=
(
σ =
10
5 . 228
= 85 . 22 (22.85 is the variance)
σ = 4.78
Measures of Dispersion 259
However, we could have applied formula (e) and reduced the calculation as follows:
Table 9.5: Determining the standard deviation of a simple frequency distribution
Weekly
output
(x)
Frequency
(f)
fx x
2
fx
2
90 1 90 8,100 8,100
94 1 94 8,836 8,836
98 2 196 9,604 19,208
100 2 200 10,000 20,000
101 2 202 10,201 20,402
106 1 106 11,236 11,236
107 1 107 11,449 11,449
10 995 99,231
Therefore:
n = Σf = 10
Σfx = 995
Σfx
2
= 99,231
Calculating the standard deviation proceeds as follows:
|
|
.
|

\
|
÷ =
n
) fx Σ (
fx Σ
n
1
σ
2
2
=
|
|
.
|

\
|
÷
10
995
231 , 99
10
1
2
= |
.
|

\
|
÷
10
025 , 990
231 , 99
10
1
= ( ) 5 . 002 , 99 231 , 99
10
1
÷
= 5 . 228
10
1
×
= 85 . 22 (the variance)
= 4.78.
Standard Deviation of a Grouped Frequency Distribution
When we come to the problem of finding the standard deviation of a grouped frequency
distribution, we again assume that all the readings in a given group fall at the midpoint of the
group. Using formula (e) and applying it to the distribution shown in Table 9.6, the standard
deviation is calculated as follows.
260 Measures of Dispersion
Table 9.6: Determining the standard deviation of a grouped frequency distribution
Class
Mid-value
(x)
Frequency
(f)
fx x
2
fx
2
10 to < 20 15 2 30 225 450
20 to < 30 25 5 125 625 3,125
30 to < 40 35 8 280 1,225 9,800
40 to < 50 45 6 270 2,025 12,150
50 to < 60 55 5 275 3,025 15,125
60 to < 70 65 3 195 4,225 12,675
70 to < 80 75 1 75 5,625 5,625
Total 30 1,250 58,950
Therefore:
n = Σf = 30
Σfx = 1,250
Σfx
2
= 58,950
Calculating the standard deviation proceeds as follows:
|
|
.
|

\
|
÷ =
n
) fx Σ (
fx Σ
n
1
σ
2
2
=
|
|
.
|

\
|
÷
30
250 , 1
950 , 58
30
1
2
= |
.
|

\
|
÷
30
500 , 562 , 1
950 , 58
30
1
= ( ) 33 . 083 , 52 950 , 58
30
1
÷
= 67 . 866 , 6
30
1
×
= 89 . 228 (the variance)
= 15.13
The arithmetic for this table is rather tedious even with a calculator, but we can extend the
short-cut method which we used for finding the arithmetic mean of a distribution to find the
standard deviation as well. In that method we:
 worked from an assumed mean
 worked in class intervals
 applied a correction to the assumed mean.
Table 9.7 applies this approach to working out the standard deviation.
Measures of Dispersion 261
Table 9.7: Determining the standard deviation of a grouped frequency distribution
(short cut)
Class
Mid-value
(x)
Frequency
(f)
d =
c
x x
o
÷
fd d
2
fd
2
10 to < 20 15 2 ÷2 ÷4 4 8
20 to < 30 25 5 ÷1 ÷5 1 5
30 to < 40 35 8 0 0 0 0
40 to < 50 45 6 +1 +6 1 6
50 to < 60 55 5 +2 +10 4 20
60 to < 70 65 3 +3 +9 9 27
70 to < 80 75 1 +4 +4 16 16
Total 30 +29
÷9
82
+20
x
o
= 35, c = 10
The standard deviation is calculated in four steps from this table, as follows:
(a) The approximate variance is obtained from
2
Σfd
n
1
× which in our case is equal to:
30
1
× 82 =
30
82
= 2.7333
(b) The correction is ÷
2
Σfd
n
1
|
.
|

\
|
×
It is always subtracted from the approximate variance to get the corrected variance. In
our case the correction is:
÷(20/30)
2
= ÷0.4444.
(c) The corrected variance is thus:
2.7333 ÷ 0.4444 = 2.2889
(d) The standard deviation is then the class interval times the square root of the corrected
variance:
SD = 2889 . 2 = 1.513 class intervals × 10
= 15.13
This may seem a little complicated, but if you work through the example a few times, it will all
fall into place. Remember the following points:
 Work from an assumed mean at the midpoint of any convenient class.
 The correction is always subtracted from the approximate variance.
 As you are working in class intervals, it is necessary to multiply by the class interval as
the last step.
262 Measures of Dispersion
 The correction factor is the same as that used for the short-cut calculation of the mean,
but for the SD it has to be squared.
 The column for d
2
may be omitted since fd
2
= fd multiplied by d. But do not omit it until
you have really grasped the principles involved.
 The formula for calculating the standard deviation in this way can be written as:
2
n
Σfd
n
Σfd
c SD |
.
|

\
|
÷ =
2
 The assumed mean should be chosen from a group with the most common interval and
c will be that interval. If the intervals vary too much, we revert to the basic formula.
Characteristics of the Standard Deviation
In spite of the apparently complicated method of calculation, the standard deviation is the
measure of dispersion used in all but the very simplest of statistical studies. It is based on all
of the individual items. It gives slightly more emphasis to the larger deviations but does not
ignore the smaller ones and, most important, it can be treated mathematically in more
D. THE COEFFICIENT OF VARIATION
Suppose that we are comparing the profits earned by two businesses. One of them may be a
fairly large business with average monthly profits of £50,000, while the other may be a small
firm with average monthly profits of only £2,000. Clearly, the general level of profits is very
different in the two cases, but what about the month-by-month variability? We can compare
the variability of the two firms by calculating the two standard deviations.
Let us suppose that both standard deviations come to £500. Now, £500 is a much more
significant amount in relation to the small firm than it is in relation to the large firm. So
although they have the same standard deviations, it would be unrealistic to say that the two
businesses are equally consistent in their monthly profits.
To overcome this difficulty, we can express the standard deviation as a percentage of the
mean in each case. We call the result the "coefficient of variation".
Applying the idea to the figures which we have just quoted, we get coefficients of variation
(usually indicated in formulae by V or CV) as follows:
for the large firm: V = 100
000 , 50
500
× = 1.0%
for the small firm: V = 100
000 , 2
500
× = 25.0%
This shows that, relatively speaking, the small firm is more erratic in its earnings than the
large firm.
Note that although a standard deviation has the same units as the variate, the coefficient of
variation is a ratio and thus has no units.
Another application of the coefficient of variation comes when we try to compare distributions
the data of which are in different units – for example, when we try to compare a French
business with an American business. To avoid the trouble of converting the dollars to euros
(or vice versa) we can calculate the coefficients of variation in each case and thus obtain
comparable measures of dispersion.
Measures of Dispersion 263
E. SKEWNESS
When the items in a distribution are dispersed equally on each side of the mean, we say that
the distribution is symmetrical. Figure 9.2. shows two symmetrical distributions.
Figure 9.2: Symmetrical distributions
When the items are not symmetrically dispersed on each side of the mean, we say that the
distribution is skewed or asymmetric.
A distribution which has a tail drawn out to the right is said to be positively skewed, while one
with a tail to the left is negatively skewed.
Two distributions may have the same mean and the same standard deviation but they may
be differently skewed. This will be obvious if you look at one of the skew distributions in
Figure 9.3 and then look at the same one through from the other side of the paper!
Figure 9.3: Skewed distributions
What does skewness tell us? It tells us that we are to expect a few unusually high values in a
positively skewed distribution or a few unusually low values in a negatively skewed
distribution.
If a distribution is symmetrical, the mean, mode and median all occur at the same point, i.e.
right in the middle. But in a skew distribution, the mean and the median lie somewhere along
the side of the "tail", although the mode is still at the point where the curve is highest. The
more skew the distribution, the greater the distance from the mode to the mean and the
median, but these two are always in the same order: working outwards from the mode, the
median comes first and then the mean – see Figure 9.4.
Variate
Frequency
Variate
Frequency
(a)
Positively
skewed
(b)
Negatively
skewed
264 Measures of Dispersion
Figure 9.4: Positions of the mean, mode and median in skew distributions
For most distributions, except for those with very long tails, the following approximate
relationship holds:
mean ÷ mode = 3(mean ÷ median)
The more skewed the distribution, the more spread out are these three measures of location,
and so we can use the amount of this spread to measure the amount of skewness. The most
usual way of doing this is to calculate Pearson's first coefficient of skewness:
Pearson's first coefficient of skewness =
SD
mode mean÷
However as we have seen the mode is not always easy to find, and so we can use an
equivalent formula for Pearson's second coefficient of skewness:
Pearson's second coefficient of skewness =
SD
median) ean m ( 3 ÷
You need to be familiar with these formulae as the basis of calculating the skewness of a
distribution (or its "coefficient of skewness", as it is also called). When you do the calculation,
remember to get the correct sign (+ or ÷) when subtracting the mode or median from the
mean. You will get negative answers from negatively skewed distributions, and positive
The value of the coefficient of skewness is between ÷3 and +3, although values below ÷1
and above +1 are rare and indicate very skewed distributions.
Examples of variates with positive skew distributions include size of incomes of a large group
of workers, size of households, length of service in an organisation and age of a workforce.
Negative skew distributions occur less frequently. One such example is the age at death for
the adult population in the UK.
Variate
Frequency
M
e
a
n
M
e
d
i
a
n
M
o
d
e
M
e
a
n
M
e
d
i
a
n
M
o
d
e
Positively skewed Negatively skewed
Measures of Dispersion 265
Questions for Practice
1. (a) Explain the advantages of using the coefficient of variation as a measure of
dispersion.
(b) A small finance company has loaned money to 10,000 customers. Each loan is a
multiple of £1,000 and the details are given below:
Loan Amount (£000) Number of customers
1 535
2 211
3 427
4 1,360
5 1,650
6 1,504
7 1,280
8 1,615
9 1,012
10 406
11 or more 0
10,000
Calculate:
(i) The arithmetic mean of the loan amounts (£).
(ii) The standard deviation of the loan amounts (£).
(iii) The coefficient of variation of the loan amounts.
2. Calculate the arithmetic mean and standard deviation for the following data using the
short-cut method. Explain carefully what is measured by the standard deviation.
Income of borrower
£
No. of borrowers
Under 1,800 1
1,800 – 2,400 6
2,400 – 3,000 9
3,000 – 3,600 14
3,600 – 4,200 17
4,200 – 4,800 23
4,800 – 5,400 28
5,400 – 6,000 35
6,000 – 6,600 39
6,600 – 7,200 48
266 Measures of Dispersion
3. Consider the following table:
Percentages of owners with assets covered by estate duty statistics
Percentages
Assets over
(£)
Assets not
over
(£)
Year 1 Year 11
– 1,000 48.3 18.8
1,000 3,000 29.5 25.1
3,000 5,000 10.9 11.9
5,000 10,000 6.2 21.3
10,000 25,000 3.4 17.9
25,000 100,000 1.5 4.4
100,000 – 0.2 0.6
Total number of owners
(000s)
18,448 19,140
You are required to:
(a) Derive by graphical methods the median, lower and upper quartile values for Year
1 and Year 11.
(b) Calculate for the two distributions the median values and explain any difference
between these results and those obtained in (a).
(c) Estimate for each year the number of people whose assets are less than (i)
£2,500, and (ii) £25,000.
(d) Write a brief report on the data, using your results where relevant.
4. Items are produced to a target dimension of 3.25 cm on a single machine. Production
is carried out on each of three shifts, each shift having a different operator – A, B and
C. It is decided to investigate the accuracy of the operators. The results of a sample of
100 items produced by each operator are as follows:
Accuracy of operators
Dimension of item (cm)
Operator
3.21 3.22 3.23 3.24 3.25 3.26 3.27 3.28 3.29
A 1 6 10 21 36 17 6 2 1
B 0 0 7 25 34 27 6 1 0
C 3 22 49 22 4 0 0 0 0
Measures of Dispersion 267
Required:
(a) State, with brief explanation, which operator is:
(i) The most accurate in keeping to the target.
(ii) The most consistent in his or her results.
(b) Illustrate your answers to part (a) by comparing the results obtained by each
operator as frequency polygons on the same graph.
(c) Why is it that a measure of the mean level of the dimension obtained by each
operator is an inadequate guide to his or her performance?
(d) Calculate the semi-interquartile range of the results obtained by operator A, to 3
decimal places.
5. Calculate the coefficient of skewness for a distribution with a mean of 10, median of 11,
distribution?
Now check your answers with those given at the end of the unit.
268 Measures of Dispersion
1. (a) The coefficient of variation is the standard deviation expressed as a percentage
of the arithmetic mean. It enables the relative variation of two sets of data to be
compared. This is of use when comparing sets of data which have widely differing
means, a situation which makes absolute comparisons difficult. The coefficient of
variation has no units and thus the relative variation of sets of data expressed in
different units can be compared. This is particularly useful when dealing, say, with
profits expressed in different currencies.
(b) We need to draw up the following table:
Loan amount
(£000)
(x)
Number of
customers
(f)
fx (x – x ) (x – x )
2
f(x – x )
2
1 535 535 –5 25 13,375
2 211 422 –4 16 3,376
3 427 1,281 –3 9 3,843
4 1,360 5,440 –2 4 5,440
5 1,650 8,250 –1 1 1,650
6 1,504 9,024 0 0 0
7 1,280 8,960 1 1 1,280
8 1,615 12,920 2 4 6,460
9 1,012 9,108 3 9 9,108
10 406 4,060 4 16 6,496
10,000 60,000 51,028
(i)
10,000
60,000
x = = £6,000
(ii) SD = 1028 . 5
Σf
) x x Σf
2
=
÷ (
= 2.259 thousand pounds
= £2,259 to nearest £.
(iii) coefficient of variation = % 100
000 , 6
259 , 2
× = 37.65%
Measures of Dispersion 269
2. We need to draw up the following table, with x
o
= £5,100 and c = £600
Income of
borrower
£
Mid-value
(x)
No. of
borrowers
(f)
d =
c
x x
o
÷
fd fd
2
Under 1,800 1,500 * 1 ÷6 ÷6 36
1,800 to < 2,400 2,100 6 ÷5 ÷30 150
2,400 to < 3,000 2,700 9 ÷4 ÷36 144
3,000 to < 3,600 3,300 14 ÷3 ÷42 126
3,600 to < 4,200 3,900 17 ÷2 ÷34 68
4,200 to < 4,800 4,500 23 ÷1 ÷23 23
4,800 to < 5,400 5,100 28 0 0 0
5,400 to < 6,000 5,700 35 1 35 35
6,000 to < 6,600 6,300 39 2 78 156
6,600 to 7,200 6,900 48 3 144 432
220 86 1,170
* Assume the first class covers the range £1,200 – £1,800 so that its mid-value is
£1,500.
x = £5,100 + 600
220
86
×
= £5,100 + 234.55
= £5,334.55
= £5,335 to the nearest £
variance =
2
220
86
220
170 , 1
|
.
|

\
|
÷ in class interval units squared
= 5.3182 – 0.1528
= 5.1654
SD = 1654 . 5 = 2.2727 class interval units
= £1,363.6
= £1,364 to the nearest £.
The standard deviation measures the dispersion or spread of a set of data. It does this
by taking every reading into account. We take the deviation of each reading from the
mean and square it to remove any negative signs in a mathematically acceptable
fashion. We then add all the squared deviations, find their mean and finally take the
square root.
270 Measures of Dispersion
3 (a) We need first to draw up the following table
Cumulative % frequency (less than)
Assets over
(£)
Year 1 Year 11
1,000 48.3 18.8
3,000 77.8 43.9
5,000 88.7 55.8
10,000 94.9 77.1
25,000 98.3 95.0
100,000 99.8 99.4
From this, we can draw ogives for both years:
Ogive to show cumulative percentage of owners
with assets covered by estate duty statistics
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25
Cumulative
percentage
frequency
Assets (£000s)
Year 1
Year 11
Measures of Dispersion 271
Reading off the graph at a cumulative percentage frequency of 50 we obtain the
median:
Year 1: median = £1,100
Year 11: median = £4,000
Reading off the graph at cumulative percentage frequencies of 25 and 75 we
obtain the lower and upper quartile figures:
Year 1: lower quartile = £500, upper quartile = £2,800
Year 11: lower quartile = £1,500, upper quartile = £9,500.
(b) Year 1:
The 50% mark is in the £1,000 – £3,000 group:
median = £1,000 + 2,000 £
5 . 29
3 . 48 50
×
÷
= £1,000 + 2,000 £
5 . 29
7 . 1
×
= £1,000 + 115.25
= £1,115.25
Year 11:
The 50% mark is in the £3,000 – £5,000 group:
median = £3,000 + 2,000 £
9 . 11
9 . 43 50
×
÷
= £3,000 + 2,000 £
9 . 11
1 . 6
×
= £3,000 + 1,025.21
= £4,025.21
These figures are more precise than the results in (a). This is to be expected as
graphical methods are approximate and are limited by the size of the graph
paper.
(c) Reading from the ogive at a value on the horizontal scale corresponding to
£2,500, we find that in Year 1, 70% of owners had assets less than £2,500. In
Year 11 the corresponding figure was 37½%.
Thus, the number with assets less than £2,500 was:
Year 1: 18,448,000
100
70
× = 12,913,600
Year 11: 19,140,000
100
5 . 37
× = 7,177,500.
From the table of values in (a), we note that in Year 1, 98.3% of owners had
assets less than £25,000. In Year 11 the corresponding percentage was 95%.
Therefore the number of those with assets less than £25,000 was 18,134,384 in
Year 1 and 18,183,000 in Year 11.
(d) The total number of owners with assets covered by estate duty statistics rose by
692,000 from Year 1 to Year 11.
272 Measures of Dispersion
As well as an increase in total wealth, there appears to have been some
redistribution of wealth in this period. Whereas the modal group in Year 1 was
under £1,000, in Year 11 it was £1,000 – £3,000 closely followed by the £5,000 –
£10,000 group.
In Year 1, 1.7% had assets over £25,000 whereas 5% had assets over that
amount in Year 11. Although inflation might be responsible for lifting some owners
to higher groups, it is unlikely to account for a change in percentage as large as
this.
The actual number of people with assets less than £25,000 was greater in Year
11 than in Year 1, presumably because more owners are covered by the statistics
in Year 11. As a typical value for assets, we note that in Year 1, 50% of the
owners had assets less than £1,115 but in Year 11, 50% had assets less than
£4,025. The interquartile range for Year 1 was £500 – £2,800 but in Year 11 it
was much larger, £1,500 – £9,500, indicating a greater spread of assets in the
middle range in the later year.
4. (a) As this question says "with brief explanation", it implies that no calculations are to
be made. So we could make the following observations:
(i) B is most accurate in keeping to the target of 3.25 cm as his/her items are
(ii) C is the most consistent in his/her results as almost 50% of his/her items
are 3.23 cm and his/her maximum spread either way is 0.02 cm.
You could back up your assertions with calculations of the mean and standard
deviation in each case if time permits. It is good practice to work them out, so
details of the calculations are now given.
You need to draw up the following table. Note that f
A
and d
A
refer to operator A.
Similarly for operators B and C.
Dimension
(cm)
(x)
f
A
f
B
f
C
d
A
= d
B
d
C
f
A
d
A
f
B
d
B
f
C
d
C
f
A
d
A
2
f
B
d
B
2
f
C
d
C
2
3.21 1 - 3 ÷4 ÷2 ÷4 - ÷6 16 - 12
3.22 6 - 22 ÷3 ÷1 ÷18 - ÷22 54 - 22
3.23 10 7 49 ÷2 0 ÷20 ÷14 0 40 28 0
3.24 21 25 22 ÷1 1 ÷21 ÷25 22 21 25 22
3.25 36 34 4 0 2 0 0 8 0 0 16
3.26 17 27 - 1 - 17 27 - 17 27 -
3.27 6 6 - 2 - 12 12 - 24 24 -
3.28 2 1 - 3 - 6 3 - 18 9 -
3.29 1 - - 4 - 4 - - 16 - -
Total 100 100 100 ÷24 3 2 206 113 72
d
A
= d
B
=
01 . 0
25 . 3 x ÷
d
C
=
01 . 0
23 . 3 x ÷
Measures of Dispersion 273
x
A
= 3.25 +
100
) 24 (÷
× 0.01 = 3.25 ÷ 0.0024 = 3.2476 cm
x
B
= 3.25 +
100
) 3 (+
× 0.01 = 3.25 + 0.0003 = 3.2503 cm
x
C
= 3.23 +
100
2
× 0.01 = 3.23 + 0.0002 = 3.2302 cm
SD
A
= 0.01
2
100
24
100
206
|
.
|

\
| ÷
÷ = 0.01 0576 . 0 06 . 2 ÷
= 0.01 0024 . 2 = 0.0142 cm
SD
B
= 0.01
2
100
3
100
113
|
.
|

\
|
÷ = 0.01 009 . 0 13 . 1 ÷
= 0.01 1291 . 1 = 0.0106 cm
SD
C
= 0.01
2
100
2
100
72
|
.
|

\
|
÷ = 0.01 0004 . 0 72 . 0 ÷
= 0.01 7196 . 0 = 0.00848 cm
This confirms our earlier assertion as x
B
is the mean closest to 3.25 cm and C
has the smallest standard deviation.
(b) Frequency polygons to show output of three operators
(c) The mean on its own gives no indication of the spread of the readings about the
mean. The mean dimension produced by A is close to 3.25 cm, but his/her output
varies from 3.21 cm to 3.29 cm.
0
10
20
30
40
50
60
3.2 3.22 3.24 3.26 3.28 3.3
Operator A
Operator C
Operator B
Dimension of item (cm)
274 Measures of Dispersion
(d) Assume that each dimension has been measured to the nearest 0.01 cm – i.e.
we assume that the one item produced by A in the 3.21 cm category is, in fact,
less than 3.215 cm.
Cumulative frequency Less than (cm)
1 3.215
7 3.225
17 3.235
38 3.245
74 3.255
91 3.265
97 3.275
99 3.285
100 3.295
The lower quartile Q
1
is the value below which 25% of the data lies, i.e. it is the
value of the 25th item.
Q
1
= 3.235 + 01 . 0
21
) 17 25 (
×
÷
= 3.235 +
21
01 . 0 8 ×
= 3.2388 cm.
Similarly
Q
3
= 3.255 + 01 . 0
17
) 74 75 (
×
÷
= 3.255 + 01 . 0
17
1
× = 3.2556 cm.
The semi-interquartile range =
2
Q Q
1 3
÷
=
2
0168 . 0
= 0.008 cm to three decimal
places.
This tells us something about the spread of the middle 50% of the data. It could
be of use to the manufacturer for predicting the variation in A's production, so that
possible wastage when items are made too far from the target dimension could
be estimated.
5. Coefficient of skewness =
deviation standard
median) 3(mean÷
=
5
) 11 10 ( 3 ÷
= ÷0.6
Thus we can say that the distribution is slightly negatively skewed – i.e. asymmetric
with the peak of the curve being to the right.
275
Study Unit 10
Probability
Contents Page
Introduction 276
A. Estimating Probabilities 277
Theoretical Probabilities 277
Empirical Probabilities 278
Subjective Probabilities 279
B. Types of Event 279
C. The Two Laws of Probability 280
Addition Law for Mutually Exclusive Events 280
Addition Law for Non-Mutually Exclusive Events 282
Multiplication Law for Independent Events 283
Multiplication Law for Dependent Events 285
Listing Equally Likely Outcomes 286
D. Conditional Probability 286
E. Tree Diagrams 288
F. Sample Space 294
G. Expected Value 295
H. The Normal Distribution 298
Properties of the Normal Distribution 300
Using Tables of the Normal Distribution 301
Further Probability Calculations 302
Answers to Questions for Practice 305
Appendix 311
276 Probability
INTRODUCTION
Probability is one of those ideas about which we all have some notion, but it may not be a
very definite one. Initially, we will not spend our time trying to get an exact definition, but will
confine ourselves to the task of grasping the idea generally and seeing how it can be used.
Other words which convey much the same idea as probability are "chance" and "likelihood".
Just as there is a scale of temperature, because some things are hotter than others, so there
is a scale of probability. Some things are more probable than others – for example, snow is
more likely to fall in winter than in summer, or a healthy person has more chance of surviving
an attack of influenza than an unhealthy person. There is, you will note, some uncertainty in
these matters. Most things in real life are uncertain to some degree or other, and it is for this
reason that the theory of probability is of great practical value. It is the branch of mathematics
which deals specifically with matters of uncertainty.
For the purpose of learning the theory, it is necessary to start with simple things like coin-
tossing and die-throwing. This may seem a bit remote from business and industrial life, but it
will help you understand the more practical applications. We shall then move on to consider
the application of probability to real situations.
This study unit also introduces the concept of the normal distribution. This is a particular
aspect of frequency distributions which we considered previously. The normal distribution has
a number of properties which can be used to work out the probability of particular event
happening.
Objectives
When you have completed this study unit you will be able to:
 explain the meaning of a priori, empirical and subjective probability
 explain the probability terms mutually exclusive, non-mutually exclusive, independent,
dependent, non-independent, and complementary
 explain what is meant by conditional probability
 solve probability problems using the addition law for mutually exclusive and non-
mutually exclusive events, and the multiplication law for dependent and independent
events
 construct and use tree diagrams, contingency tables and sample spaces to solve
probability problems
 calculate, from appropriate data, the expected value of an outcome
 describe, in simple terms, the significance of the normal frequency distribution
 explain the properties of the normal distribution in terms of the location of observations,
the mean and the standard deviation
 use normal distribution tables to find probabilities that observations within specified
limits will occur.
Probability 277
A. ESTIMATING PROBABILITIES
Suppose someone tells you: "there is a 50-50 chance that we will be able to deliver your
order on Friday". This statement means something intuitively, even though when Friday
arrives there are only two outcomes. Either the order will be delivered or it will not.
Statements like this are trying to put probabilities or chances on uncertain events.
Probability is measured on a scale between 0 and 1: the probability limits are thus:
0 ≤ P(A) ≤ 1
where P(A) is the probability of an event A occurring.
Any event which is impossible has a probability of 0, and any event which is certain to occur
has a probability of 1. Accordingly, the probability of all outcomes for a particular event does
not exceed 1: the total probability rule is thus:
ΣP = 1 (for all outcomes).
For example, the probability that the sun will not rise tomorrow is 0; the probability that a light
bulb will fail sooner or later is 1. For uncertain events, the probability of occurrence is
somewhere between 0 and 1. The 50-50 chance mentioned previously is equivalent to a
probability of 0.5.
Try to estimate probabilities for the following events. Remember that events which are more
likely to occur than not have probabilities which are greater than 0.5, and the more certain
they are the closer the probabilities are to 1. Similarly, events which are more likely not to
occur have probabilities which are less than 0.5. The probabilities get closer to 0 as the
events get more unlikely. The events are:
(a) The probability that a coin will fall heads when tossed.
(b) The probability that it will snow next January.
(c) The probability that sales for your company will reach record levels next year.
(d) The probability that your car will not break down on your next journey.
(e) The probability that the throw of a die will show a six.
The probabilities are as follows:
(a) The probability of heads is one in two or 0.5.
(b) This probability is quite low. It is somewhere between 0 and 0.1.
(c) You can answer this one yourself.
(d) This depends on how frequently your car is serviced. For a reliable car it should be
considerably greater than 0.5.
(e) The probability of a six is one in six or
6
1
or 0.167.
Theoretical Probabilities
Sometimes probabilities can be specified by considering the physical aspects of the situation.
For example, consider the tossing of a coin. What is the probability that it will fall heads?
There are two sides to a coin. There is no reason to favour either side as a coin is
symmetrical. Therefore, the probability of heads, which we call P(H), is:
P(H) = 0.5
278 Probability
Another example is throwing a die. A die has six sides. Again, assuming it is not weighted in
favour of any of the sides, there is no reason to favour one side rather than another.
Therefore, the probability of a six showing uppermost, P(6), is:
P(6) =
6
1
= 0.167
As a third and final example, imagine a box containing 100 beads of which 23 are black and
77 white. If we pick one bead out of the box at random (blindfold and with the box well
shaken up), what is the probability that we will draw a black bead? We have 23 chances out
of 100, so the probability is:
P(B) =
100
23
= 0.23
Probabilities of this kind, which can be assessed from our prior knowledge of a situation, are
also called a priori probabilities.
In general terms, we can say that if an event E can happen in h ways out of a total of n
possible ways, then the probability of that event occurring (called a success) is given by:
P(E) =
outcomes possible of number total
occurring E of ways possible of number
=
n
h
In the preceding examples, we introduced the concept that some outcomes are equally likely,
while others are not equally likely. In probability, when there are the same chances for more
than one outcome, we say that the outcomes are equally likely to occur. For example:
 If someone tosses a coin, the chances of getting a heads or tails are the same, i.e. all
outcomes are equally likely. The probability of getting a heads is P(H) = 0.5 and the
probability of getting a tails is P(T) = 0.5.
 If someone throws a die, the chances of getting a six are no different to getting any
other number, i.e. all outcomes are equally likely. The probability of getting a one is
P(1) = 0.167, the probability of getting a two is P(2) = 0.167, etc.
But not all outcomes are equally likely to occur. In the example about the box containing 100
beads, of which 23 are black and 77 white, the probability of getting a black bead is P(B) =
0.23, whereas the probability of getting a white bead is P(R) = 0.77. Thus, the chances of
getting a red bead or a black bead are not the same, i.e. all outcomes for this particular event
are not equally likely.
Empirical Probabilities
Often it is not possible to give a theoretical probability of an event.
For example, what is the probability that an item on a production line will fail a quality control
test? This question can be answered either by measuring the probability in a test situation or
by relying on previous results (i.e. empirically). If 100 items are taken from the production line
and tested, then:
probability of failure, P(F) =
tested items of no. total
fail which items of no.
So, if 5 items actually fail the test
P(F) =
100
5
= 0.05
Sometimes it is not possible to set up an experiment to calculate an empirical probability. For
example, what are your chances of passing a particular examination? You cannot sit a series
Probability 279
of examinations to answer this. Previous results must be used. If you have taken 12
examinations in the past, and failed only one, you might estimate:
probability of passing, P(pass) =
12
11
= 0.92
Subjective Probabilities
A subjective probability describes an individual's personal judgement or instinct about how
likely a particular event is to occur. It is not based on any precise theoretical or empirical
foundation, but is often a reasonable assessment by a knowledgeable person. Thus, a
person's subjective probability of an event describes his or her degree of belief in the
outcome of that event. Like all probabilities, a subjective probability is conventionally
expressed on a scale from 0 to 1; a rare event has a subjective probability close to 0, a very
common event has a subjective probability close to 1.
In business, managers will often make strategic decisions based on the subjective probability
of the outcome of an event – i.e. on their instinct, given their knowledge of the business
environment in which they operate. For example, investors on the stock market may base
many of their decisions to buy and sell shares on their instinct. Successful investors are
those who know which stocks to buy at a relatively low price, and similarly they know at what
price to sell them prior to a downturn in the market. If these investment decisions were based
on a precise theoretical or empirical foundation, then we would all be doing it! But they are
not, they are based on instinct: investors attach (subjective) probabilities to the likelihood that
share prices will move in a particular direction and they make their decisions accordingly.
B. TYPES OF EVENT
In probability, we can identify five types of event:
 mutually exclusive
 non-mutually exclusive
 independent
 dependent or non-independent
 complementary.
(a) Mutually exclusive events
If two events A and B are mutually exclusive then the occurrence of one event
precludes the possibility of the other occurring. Thus:
P(A and B) = 0.
For example, the two sides of a coin are mutually exclusive since, on the throw of the
coin, heads automatically rules out the possibility of tails. On the throw of a die, a six
excludes all other possibilities. In fact, all the sides of a die are mutually exclusive; the
occurrence of any one of them as the top face automatically excludes any of the
others.
(b) Non-mutually-exclusive events
These are events which can occur together. For example, in a pack of playing cards
hearts and queens are non-mutually exclusive, since there is one card, the queen of
hearts, which is both a heart and a queen and so satisfies both criteria for success.
280 Probability
(c) Independent events
These are events which are not mutually exclusive and where the occurrence of one
event does not affect the occurrence of the other. We can express this as:
P(A) = P(A|B) and/or P(B) = P(B|A)
where the vertical line is a shorthand way of writing "given that".
For example, the tossing of a coin in no way affects the result of the next toss of the
coin; each toss has an independent outcome.
(d) Dependent or non-independent events
These are situations where the outcome of one event is dependent on another event.
For example, the probability of a car owner being able to drive to work in her car is
dependent on her being able to start the car. The probability of her being able to drive
to work given that the car starts is a conditional probability. We can express this as:
P(drive to work|car starts)
again where the vertical line is a shorthand way of writing "given that".
(e) Complementary events
An event either occurs or it does not occur, i.e. we are certain that one or other of these
situations holds.
For example, if we throw a die and denote the event where a six is uppermost by A,
and the event where either a one, two, three, four or five is uppermost by A (or not A),
then A and A are complementary, i.e. they are mutually exclusive with a total
probability of 1. Thus:
P(A) + P( A ) = 1
This relationship between complementary events is useful, as it is often easier to find
the probability of an event not occurring than to find the probability that it does occur.
Using this formula, we can always find P(A) by subtracting P( A ) from 1.
C. THE TWO LAWS OF PROBABILITY
In this section we will define the two laws of probability, namely the addition law and the
multiplication law, and discuss the situations when each can be used.
Addition Law for Mutually Exclusive Events
Consider again the example of throwing a die. You will remember that:
P(6) =
6
1
Similarly:
P(1) =
6
1
P(2) =
6
1
P(3) =
6
1
P(4) =
6
1
P(5) =
6
1
Probability 281
What is the chance of getting 1, 2 or 3?
From the symmetry of the die you can see that P(1 or 2 or 3) = 0.5. But also, from the
equations shown here you can see that:
P(1) + P(2) + P(3) =
6
1
+
6
1
+
6
1
= 0.5
This illustrates that
P(1 or 2 or 3) = P(1) + P(2) + P(3)
This result is a general one and it is called the addition law of probabilities for mutually
exclusive events. It is used to calculate the probability of one of any group of mutually
exclusive events.
It is stated more generally as:
P(A or B or ... or N) = P(A) + P(B) + ... + P(N)
where A, B ... N are mutually exclusive events.
If all possible mutually exclusive events are listed, then it is certain that one of these
outcomes will occur. For example, when the die is tossed there must be one number showing
afterwards.
P(1 or 2 or 3 or 4 or 5 or 6) = 1
Using the addition law for mutually exclusive events, this can also be stated as
P(1) + P(2) + P(3) + P(4) + P(5) + P(6) = 1
Again this is a general rule. The sum of the probabilities of a complete list of mutually
exclusive events will always be 1.
Example:
An urn contains 100 coloured balls. Five of these are red, seven are blue and the rest are
white. One ball is to be drawn at random from the urn. What is the probability that it will be
red?
Probability that ball is red P(R) =
100
5
= 0.05
What is the probability that it will be blue?
Probability that ball is blue P(B) =
100
7
= 0.07
What is the probability that it will be red or blue?
P(R or B) = P(R) + P(B) = 0.05 + 0.07 = 0.12
This result uses the addition law for mutually exclusive events since a ball cannot be both
blue and red.
What is the probability that it will be white?
The ball must be either red or blue or white. This is a complete list of mutually exclusive
possibilities. Therefore:
P(R) + P(B) + P(W) = 1
Then the probability of one ball being white is given as:
P(W) = 1 ÷ P(R) ÷ P(B)
= 1 ÷ 0.05 ÷ 0.07
= 0.88
282 Probability
Addition Law for Non-Mutually Exclusive Events
Events which are non-mutually exclusive are, by definition, capable of occurring together.
To determine the probability of one of the events occurring, we can still use the addition law,
but the probability of the events occurring together must be deducted:
P(A or B or both) = P(A) + P(B) ÷ P(A and B)
Example 1:
If one card is drawn from a pack of 52 playing cards, what is the probability (a) that it is either
a spade or an ace, or (b) that it is either a spade or the ace of diamonds?
(a) Let event B be "the card is a spade" and event A be "the card is an ace".
We need to determine the probability of the event being either a spade or an ace. We
also need to take account of the probability of the card being the ace of spades (i.e. the
event being both A and B). However, this probability will be included in the probability of
it being a spade and in the probability of it being an ace. Thus, we need to deduct that
probability from one of the other probabilities. This is given by:
P(spade or ace [or both]) = P(A or B)
= P(A) + P(B) ÷ P(A and B)
P(A) =
52
4
pack in no.
aces of no.
=
P(B) =
52
13
pack in no.
=
P(A and B) =
52
1
pack in no.
=
13
4
52
16
52
1
52
13
52
4
= = ÷ +
(b) Let event B be "the card is a spade" and event A be "the card is the ace of diamonds".
Here, the conditions are not met by both events occurring, so the probability is given
by:
P(spade or ace of diamonds) = P(A) + P(B)
P(A) =
52
1
pack in no.
diamonds of aces of no.
=
P(B) =
52
13
pack in no.
=
P(spade or ace of diamonds) =
26
7
52
14
52
13
52
1
= = +
Probability 283
Example 2:
unwrapped.
Then:
= P(T or W)
= P(T) + P(W) ÷ P(T and W)
= 0.5 + 0.6 ÷ 0.2
= 0.9
So,
10
9
Multiplication Law for Independent Events
Consider an item on a production line. This item could be defective or acceptable. These two
possibilities are mutually exclusive and represent a complete list of alternatives.
Assume that:
probability that it is defective, P(D) = 0.2
probability that it is acceptable, P(A) = 0.8
Now consider another facet of these items. There is a system for checking them, but only
every tenth item is checked. This is shown as:
probability that it is checked, P(C) = 0.1
probability that it is not checked, P(N) = 0.9
Again these two possibilities are mutually exclusive and they represent a complete list of
alternatives. An item is either checked or it is not.
Consider the possibility that an individual item is both defective and not checked. These two
events can obviously both occur together so they are not mutually exclusive. They are,
however, independent. That is to say, whether an item is defective or acceptable does not
affect the probability of it being tested.
There are also other kinds of independent events. If you toss a coin once and then again a
second time, the outcome of the second test is independent of the results of the first one.
The results of any third or subsequent test are also independent of any previous results. The
probability of heads on any test is 0.5 even if all the previous tests have resulted in heads.
To work out the probability of two independent events both happening, you use the
multiplication law for independent events. This can be stated as:
P(A and B) = P(A) × P(B) if A and B are independent events.
Again this result is true for any number of independent events. So:
P(A and B ... and N) = P(A) × P(B) ... × P(N)
284 Probability
Consider the last example: for any item
probability that it is defective: P(D) = 0.2
probability that it is acceptable: P(A) = 0.8
probability that it is checked: P(C) = 0.1
probability that it is not checked: P(N) = 0.9
Using the multiplication law to calculate the probability that an item is both defective and not
checked, we get:
P(D and N) = 0.2 × 0.9 = 0.18
The probabilities of the other combinations of independent events can also be calculated.
P(D and C) = 0.2 × 0.1 = 0.02
P(A and N) = 0.8 × 0.9 = 0.72
P(A and C) = 0.8 × 0.1 = 0.08
Example 1:
A machine produces two batches of items. The first batch contains 1,000 items of which 20
are damaged. The second batch contains 10,000 items of which 50 are damaged. If one item
is taken from each batch, what is the probability that both items are defective?
For the item from the first batch:
probability that it is defective: P(D
1
) = 02 . 0
1,000
20
=
For the item taken from the second batch:
probability that it is defective: P(D
2
) = 005 . 0
10,000
50
=
Since these two probabilities are independent:
P(D
1
and D
2
) = P(D
1
) × P(D
2
) = 0.02 × 0.005 = 0.0001
Example 2:
A card is drawn at random from a well shuffled pack of playing cards. What is the probability
that the card is a heart? What is the probability that the card is a three? What is the
probability that the card is the three of hearts?
Probability of a heart: P(H) =
52
13
=
4
1
Probability of a three: P(3) =
52
4
=
13
1
Probability of the three of hearts, since the suit and the number of a card are independent:
P(H and 3) = P(H) × P(3) =
52
1
13
1
4
1
= ×
Probability 285
Example 3:
A die is thrown three times. What is the probability of one or more sixes in these three
throws?
Probability of no six in first throw =
6
5
Similarly, probability of no six in second or third throw =
6
5
The result of each throw is independent, so:
probability of no six in all three throws =
216
125
6
5
6
5
6
5
= × ×
Since no sixes and one or more sixes are mutually exclusive and cover all possibilities:
probability of one or more sixes =
216
91
216
125
1 = ÷
Multiplication Law for Dependent Events
To work out the probability of two dependent events both happening, you use the
multiplication law for dependent events. This can be stated as:
P(A and B) = P(A) × P(B|A)
where: A and B are dependent events and the vertical line is a shorthand way of writing
"given that".
Example 1:
For example, suppose there are 10 marbles in a bag, and 3 are defective. Two marbles are
to be selected one after the other without replacement. What is the probability of selecting a
defective marble followed by another defective marble?
Probability that the first marble selected is defective: P(A)
10
3
=
Probability that the second marble selected is defective: P(B)
9
2
=
Thus, P(A and B) 7%
9
2
10
3
= × =
This means that if this experiment were repeated 100 times, in the long run 7 experiments
would result in defective marbles on both the first and second selections.
Example 2:
Select one card at random from a deck of cards and find the probability that the card is an 8
and a diamond.
We know that there are four '8' cards (i.e. P(8) )
4
1
= in the pack and that a quarter of the pack
are diamonds (i.e. P(D) )
52
4
= , thus:
P(D and 8)
52
1
4
1
52
4
= × =
286 Probability
Listing Equally Likely Outcomes
Another matter about which you must be careful is the listing of equally likely outcomes. Be
sure that you list all of them. For example, we can list the possible results of tossing two
coins:
First Coin Second Coin
Tails Tails
There are four equally likely outcomes. Do not make the mistake of saying, for example, that
there are only two outcomes (both heads or not both heads) – you must list all the possible
outcomes. (In this case "not both heads" can result in three different ways, so the probability
of this result will be higher than "both heads".)
In this example, the probability that there will be one heads and one tails (heads-tails, or tails-
heads) is 0.5. This is a case of the addition law at work:
Putting it another way, the probability of different faces is equal to the probability of the same
faces – in both cases 0.5.
D. CONDITIONAL PROBABILITY
The probability of an event A occurring where that event is conditional upon event B having
also occurred is expressed as:
P(A|B)
The events A and B must be causally linked in some way for A to be conditional upon B. For
example, we noted earlier the case where a car owner's ability to drive to work in her car
(event A) is dependent on her being able to start the car (event B). There is self evidently a
causal link between the two events. However, if we consider the case of her being able to
drive to work given that she is married, there is no necessary link between the two events –
and there is no conditional probability.
In a case involving equally likely outcomes for both events (i.e. both events A and B may or
may not occur), the conditional probability of event A occurring, given that B also occurs, is
calculated by the following formula:
P(A|B) =
) B ( n
B) and n(A
where n = the number of possible outcomes.
Similarly, the conditional probability of event B occurring, given that A also occurs, is
calculated by the following formula:
P(B|A) =
n(A)
B) and n(A
where n = the number of possible outcomes.
One approach to determining probability in such circumstances is to construct a table of the
possible outcomes – sometimes called a contingency table.
Consider the case of 100 candidates attempting their driving test for the first time. Suppose
80 of the candidates pass and 50 of them have had 10 or more lessons. Of those who have
had 10 or more lessons, 45 of them passed. What is the probability of passing at the first
attempt if you have had only 6 lessons?
Probability 287
We can construct a table which identifies the possible outcomes as follows in Table 10.1.
Table 10.1: A partially completed contingency table
Passed Failed Total
10 plus lessons 45 50
Less than 10 lessons
Total 80 100
We will call event A "passed" and event B "having had less than 10 lessons".
Make sure you understand the construction of this table:
 We have listed the possible outcomes of event A along the top – these outcomes are
either to have passed or failed the test. Thus, the columns will show the number of
outcomes for either possibility.
 We have listed the possible outcomes of event B along the side – these are either to
have had 10 or more lessons, or less than 10 lessons. Thus, the rows will show the
number of outcomes for either possibility.
 We have entered in the information that we know: that there are a total of 100 possible
outcomes; that the total number of outcomes in which event A was "passed" is 80; that
the total number of outcomes in which event B is "had 10 or more lessons" is 50; and
that the total number of outcomes in which event A was "passed" and event B was "had
10 or more lessons" is 45.
We can complete the rest of the table by subtraction as in Table 10.2.
Table 10.2: Contingency table
Passed Failed Total
10 plus lessons 45 5 50
Less than 10 lessons 35 15 50
Total 80 20 100
We can now work out the total outcomes applicable to the question "what is the probability of
passing at the first attempt if you have had only 6 lessons?"
Remember that event A is "passed" and event B is "having had less than 10 lessons".
Therefore:
 the total number of outcomes of B = 50
 the total number of outcomes of both A and B = 35.
Thus the probability is:
P(A|B) =
n(B)
B) and n(A
=
50
35
= 0.7
288 Probability
We could consider the question: "what is the probability of failing at the first attempt if you
have had more than 10 lessons?". Here event A is "failed" and event B is "having had more
than 10 lessons". The calculation is:
P(A|B) =
n(B)
B) and n(A
=
50
5
= 0.1
Questions for Practice 1
See if you can work out the answer to the following questions using the preceding table:
"What is the probability of failing at the first attempt if you have only had 8 lessons?"
Remember to be clear about which is event A and event B, then read off the total number of
outcomes for event B and for event A and B together, and then apply the formula.
Now check your answers with those given at the end of the unit.
E. TREE DIAGRAMS
A compound experiment, i.e. one with more than one component part, may be regarded as a
sequence of similar simple experiments. For example, the rolling of two dice can be
considered as the rolling of one followed by the rolling of the other; and the tossing of four
coins can be thought of as tossing one after the other. A tree diagram enables us to construct
an exhaustive list of mutually exclusive outcomes of a compound experiment.
Furthermore, a tree diagram gives us a pictorial representation of probability.
 By exhaustive, we mean that every possible outcome is considered.
 By mutually exclusive we mean, as before, that if one of the outcomes of the
compound experiment occurs then the others cannot.
In the following examples, we shall consider the implications of conditional probability – the
probability of one event occurring if another event has also occurred.
Example 1:
The concept can be illustrated using the example of a bag containing five red and three white
billiard balls. If two are selected at random without replacement, what is the probability that
one of each colour is drawn?
Let:
R indicate a red ball being selected
W indicate a white ball being selected
When the first ball is selected, the probabilities are:
P(R) =
8
5
P(W) =
8
3
Probability 289
When the second ball is selected, there are only seven balls left and the mix of red and white
balls will be determined by the selection of the first ball – i.e. the probability of a red or a
white ball being selected is conditional upon the previous event. Thus:
If the first ball selected was red, there will be four red and three white balls left. The
probabilities for the selection of the second ball are then:
P(R) =
7
4
, and P(W) =
7
3
If the first ball selected was white, there will be five red and two white balls left. The
probabilities for the selection of the second ball are then:
P(R) =
7
5
, and P(W) =
7
2
This process of working out the different conditional probabilities can be very complicated
and it is easier to represent the situation by means of a tree diagram as in Figure 10.1. The
probabilities at each stage are shown alongside the branches of the tree.
Figure 10.1: Example 1 – tree diagram
The diagram is constructed as follows, working from left to right.
 At the start we take a ball from the bag. This ball is either red or white so we draw two
branches labelled R and W, corresponding to the two possibilities. We then also write
on the branch the probability of the outcome of this simple experiment being along that
branch.
 We then consider drawing a second ball from the bag. Whether we draw a red or a
white ball the first time, we can still draw a red or a white ball the second time. So we
mark in the two possibilities at the end of each of the two branches of our existing tree
diagram. We enter on these second branches the conditional probabilities associated
with them. Thus, on the uppermost branch in the diagram we must insert the probability
of obtaining a second red ball given that the first was red.
Each complete branch from start to tip represents one possible outcome of the compound
experiment and each of the branches is mutually exclusive. We can see that there are four
different mutually exclusive outcomes possible, namely RR, RW, WR and WW.
2nd Ball 1st Ball
Start
R
R
W
W
W
R
8
5
8
3
7
3
7
4
7
5
7
2
290 Probability
To obtain the probability of a particular outcome of the compound experiment occurring, we
multiply the probabilities along the different sections of the branch, using the general
multiplication law for probabilities. We thus obtain the probabilities shown in Table 10.3. The
sum of the probabilities should add up to 1, as we know one or other of these mutually
exclusive outcomes is certain to happen.
Table 10.3: Example 1 – outcomes from compound (conditional) events
Outcome Probability
RR
56
20
7
4
8
5
= ×
RW
56
15
7
3
8
5
= ×
WR
56
15
7
5
8
3
= ×
WW
56
6
7
2
8
3
= ×
Total 1
Example 2:
A bag contains three red balls, two white balls and one blue ball. Two balls are drawn at
random (without replacement). Find the probability that:
(a) Both white balls are drawn.
(b) The blue ball is not drawn.
(c) A red then a white are drawn.
(d) A red and a white are drawn.
To solve this problem, let us build up a tree diagram.
Stage 1 will be to consider the situation when the first ball is drawn:
Figure 10.2: Example 2 – tree diagram of first event
Start
R1
W1
B1
6
3
6
2
6
1
Probability 291
We have given the first ball drawn a subscript of 1 – for example, red first = R
1
. We shall give
the second ball drawn a subscript of 2.
The probabilities in the first stage will be:
P(R
1
) =
6
3
balls of no. total
balls red of no.
=
P(W
1
) =
6
2
(P(B
1
) =
6
1
The possibilities in stage 2 may now be mapped on the tree diagram. When the second ball
is drawn, the probability of it being red, white or blue is conditional on the outcome of the first
event. Thus, note there was only one blue ball in the bag, so if we picked a blue ball first then
we can have only a red or a white second ball. Also, whatever colour is chosen first, there
are only five balls left as there is no replacement.
Figure 10.3: Example 2 – tree diagram of first and second events
We can now list all the possible outcomes, with their associated probabilities.
2nd Ball 1st Ball
B2
W2
R2
5
2
5
2
5
1
B2
W2
R2
5
1
5
3
5
1
W2
R2
5
3
5
2
Start
R1
W1
B1
6
3
6
2
6
1
292 Probability
Table 10.4: Example 2 – outcomes from compound (conditional) events
Outcome Probability
R
1
R
2
5
1
5
2
6
3
= ×
R
1
W
2
5
1
5
2
6
3
= ×
R
1
B
2
10
1
5
1
6
3
= ×
W
1
R
2
5
1
5
3
6
2
= ×
W
1
W
2
15
1
5
1
6
2
= ×
W
1
B
2
15
1
5
1
6
2
= ×
B
1
R
2
10
1
5
3
6
1
= ×
B
1
W
2
15
1
5
2
6
1
= ×
Total 1
It is possible to read off the probabilities we require from Table 10.4.
(a) Probability that both white balls are drawn:
P(W
1
W
2
) =
15
1
(b) Probability the blue ball is not drawn:
P(R
1
R
2
) + P(R
1
W
2
) + P(W
1
R
2
) + P(W
1
W
2
) =
3
2
15
1
5
1
5
1
5
1
= + + +
(c) Probability that a red then a white are drawn:
P(R
1
W
2
) =
5
1
(d) Probability that a red and a white are drawn:
P(R
1
W
2
) + P(W
1
R
2
) =
5
2
5
1
5
1
= +
Example 3:
A couple decide to continue having children, to a maximum of four, until they have a son.
Draw a tree diagram to find the possible family size and calculate the probability that they
have a son.
We can assume that any one child is equally likely to be a boy or a girl, i.e. P(B) = P(G) = 0.5.
Note that once they have produced a son, they do not have any more children. The tree
diagram will be as in Figure 10.4.
Probability 293
Figure 10.4: Example 3 – Tree diagram
The table of probabilities is as follows.
Table 10.5: Example 3 – probabilities of outcomes
Possible Families Probability
1 Boy
2
1
1 Girl, 1 Boy
4
1
2
1
2
= |
.
|

\
|
2 Girls, 1 Boy
8
1
2
1
3
= |
.
|

\
|
3 Girls, 1 Boy
16
1
2
1
4
= |
.
|

\
|
4 Girls
16
1
2
1
4
= |
.
|

\
|
Total 1
The probability that the family will have a son before their fourth child is therefore:
16
15
16
1
8
1
4
1
2
1
= + + +
Start
½
B
½
B
B
B G
G
G
½
½
½
½
½
½
G
2nd Child
1st Child
4th Child
3rd Child
294 Probability
F. SAMPLE SPACE
You need a clear head to perform probability calculations successfully. It helps, then, to have
some diagrammatic form of representation. We have seen the value of tree diagrams in the
last section, and now we consider another method of drawing the possible outcomes of an
event.
When we toss a coin three times and note the outcome, we are performing a statistical
experiment. If we make a list of all the possible outcomes of our experiment, we call this a
sample space. The sample space in the coin-tossing experiment is:
H H H H T T
T H H T H T
H T H T T H
H H T T T T
where, for example, T H H means that on the first toss we obtained a tail, on the second a
Let us consider another example.
Suppose we have five people – A, B, C, D, E – and we wish to select a random sample of
two for interview, i.e. each of A, B, C, D, E must have the same chance of being chosen.
What is the sample space, i.e. the list of all possible different samples?
The sample space is:
A B B C C D D E
A C B D C E
A D B E
A E
In this example the order of the sample, i.e. whether we choose A followed by B or B followed
by A, is not relevant as we would still interview both A and B.
Having written down our sample space, we might be interested in a particular section of it.
For instance, in our first example we might want to see in how many of the outcomes we
obtained only one head: i.e. we are interested in the event of obtaining exactly one head.
Looking back at the sample space, we see that there are three outcomes making up this
event. If we have a fair coin, the probability of obtaining a head or a tail at any toss is ½ and
all the outcomes in the sample space are equally likely. There are eight outcomes in all, so
we can now deduce that:
Probability of obtaining exactly one head in 3 tosses of a fair coin
=
space sample in outcomes of no. total
head one with space sample in outcomes of no.
=
8
3
Alternatively, we could write:
P(event A) =
outcomes of no. total
A in outcomes of no.
=
8
3
Probability 295
Now consider the example of an experiment involving the rolling of one fair die. What is the
probability of obtaining:
(a) a score greater than 3
(b) an odd number
(c) both a score greater than 3 and an odd number
The sample space is a listing of all the possible outcomes as follows:
1, 2, 3, 4, 5, 6
The total number of possible outcomes is 6, and it is now easy to read off the probabilities
from the sample space:
(a) a score greater than 3 – three possibilities: P =
2
1
6
3
=
(b) an odd number – three possibilities: P =
2
1
6
3
=
(c) both a score greater than 3 and an odd number – one possibility: P =
6
1
G. EXPECTED VALUE
Probabilities may also be used to assist the decision-making process in business situations.
Before a new undertaking is started, for example, it may be possible to list or categorise all
the different possible outcomes. On the basis of past experience it may also be possible to
attribute probabilities to each outcome. These figures can then be used to calculate an
expected value for the income from the new undertaking.
Example:
Suppose a company is considering opening a new retail outlet in a town where it currently
has no branches. On the basis of information from its other branches in similar towns, it can
categorise the possible revenue from the new outlet into categories of high, medium or low.
Table 10.6 shows the probabilities that have been estimated for each category. You will note
that the probabilities sum to 1.
Table 10.6: Probabilities of likely revenue returns
Likely Revenue Probability
High 0.4
Medium 0.5
Low 0.1
Using these figures it is possible to calculate the expected value for the revenue from the
new branch.
The expected value is an average for the revenue, weighted by the probabilities. It is the sum
of each value for the outcome multiplied by its probability. Thus, the expected value E of
variable x with associated probabilities P(x) is calculated by the following formula:
E(x) = ΣxP(x)
where x is the value of the outcome.
296 Probability
In our example, if we give values to the outcomes of each category (i.e. figures for the high,
medium and low revenues) and then multiply each by its probability and sum them, we get an
average revenue return for new branches. This expected revenue can then be compared
with the outlay on the new branch in order to assess the level of profitability.
Table 10.7: Expected value of revenue returns
Category Revenue
(£000)
Probability Prob. × Rev.
(£000)
High 5,000 0.4 2,000
Medium 3,000 0.5 1,500
Low 1,000 0.1 100
Total: £3,600
The concept of expected value is used to incorporate the consequences of uncertainty into
the decision-making process.
Questions for Practice 2
1. If three coins are thrown, what is the probability that all three will show tails?
2. (a) (i) What is the probability of getting a total of six points in a simultaneous
throw of two dice?
(ii) What is the probability of getting at least one six from a simultaneous throw
of two dice?
(b) Out of a production batch of 10,000 items, 600 are found to be of the wrong
dimensions. Of the 600 which do not meet the specifications for measurement,
200 are too small. What is the probability that, if one item is taken at random from
the 10,000:
(i) It will be too small?
(ii) It will be too large?
If two items are chosen at random from the 10,000, what is the probability that:
(iii) Both are too large?
(iv) One is too small and one too large?
3. (a) Calculate, when three dice are thrown together, the probability of obtaining:
(i) three sixes;
(ii) one six;
(iii) at least one six.
(b) Calculate the probability of drawing in four successive draws the ace and king of
hearts and diamonds suits from a pack of 52 playing cards.
Probability 297
(c) Calculate the probability that three marbles drawn at random from a bag
containing 24 marbles, six of each of four colours (red, blue, green and white),
will be:
(i) all red;
(ii) all of any one of the four colours.
4. An analysis of 72 firms which had introduced flexitime working showed that 54 had
used some form of clocking in system whilst the rest relied on employees completing
their own time sheets. It was found that for 60 of the firms output per employee rose
and in all other firms it fell. Output fell in only 3 of the firms using employee-completed
forms.
Construct a contingency table and then:
(a) Calculate the probability of output per employee falling if an automatic clocking in
system is introduced.
(b) Determine under which system it is more probable that output per employee will
rise.
5. The probability that machine A will be performing a useful function in five years' time is
0.25, while the probability that machine B will still be operating usefully at the end of the
same period is 0.33.
Find the probability that, in five years' time:
(a) Both machines will be performing a useful function.
(b) Neither will be operating.
(c) Only machine B will be operating.
(d) At least one of the machines will be operating.
6. Three machines, A, B and C, produce respectively 60%, 30% and 10% of the total
production of a factory. The percentages of defective production for each machine are
respectively 2%, 4% and 6%.
(a) If an item is selected at random, find the probability that the item is defective.
(b) If an item is selected at random and found to be defective, what is the probability
that the item was produced by machine A?
7. Using a sample space diagram, calculate the probability of getting, with two throws of a
die:
(a) A total score of less than 9.
(b) A total score of 9.
Now check your answers with those given at the end of the unit.
298 Probability
H. THE NORMAL DISTRIBUTION
Earlier in the course, we considered various graphical ways of representing a frequency
distribution. We considered a frequency dot diagram, a bar chart, a polygon and a frequency
histogram. For example, a typical histogram will take the form shown in Figure 10.5. You can
immediately see from this diagram that the values in the centre are much more likely to occur
than those at either extreme.
Figure 10.5: Typical frequency histogram
Consider now a continuous variable in which you have been able to make a very large
number of observations. You could compile a frequency distribution and then draw a
frequency bar chart with a very large number of bars, or a histogram with a very large
number of narrow groups. Your diagrams might look something like those in Figure 10.6.
Figure 10.6: Frequency bar chart and histogram
Frequency
Value
80 90 100 110 120 130 140
0
10
20
30
40
150
Probability 299
If you now imagine that these diagrams relate to a relative frequency distribution and that a
smooth curve is drawn through the tops of the bars or rectangles, you will arrive at the idea of
a frequency curve.
We can then abstract the concept of a frequency distribution by illustrating it without any
actual figures as a frequency curve which is just a graph of frequency against the variate
(variable):
Figure 10.7: Frequency curve
The "normal" or "Gaussian" distribution was discovered in the early eighteenth century and is
probably the most important distribution in the whole of statistical theory. This distribution
seems to represent accurately the random variation shown by natural phenomena – for
example:
 heights of adult men from one race
 weights of a species of animal
 distribution of IQ levels in children of a certain age
 weights of items packaged by a particular packing machine
 life expectancy of light bulbs.
The typical shape of the normal distribution is shown in Figure 10.8. You will see that it has a
central peak (i.e. it is unimodal) and that it is symmetrical about this centre. The mean of this
distribution is shown as "m" on the diagram, and is located at the centre. The standard
deviation, which is usually denoted by "σ", is also shown.
Figure 10.8: The normal distribution
Variate
Frequency
m
Variate
Frequency
o
300 Probability
Properties of the Normal Distribution
The normal distribution has a number of interesting properties. These allow us to carry out
certain calculations on it.
For normal distributions, we find that:
 68% of the observations are within ± 1 standard deviation of the mean and
 95% are within ± 2 standard deviations of the mean.
We can illustrate these properties on the curve itself:
Figure 10.9: Properties of the normal distribution
These figures can be expressed as probabilities. For example, if an observation x comes
from a normal distribution with mean m and standard deviation σ, the probability that x is
between (m ÷ σ) and (m + σ) is:
P(m ÷ σ < x < m + σ) = 0.68
Also
P(m ÷ 2σ < x < m + 2σ) = 0.95
m
s.d. s.d.
m
s.d. s.d. s.d. s.d.
Probability 301
Using Tables of the Normal Distribution
Because the normal distribution is perfectly symmetrical, it is possible to calculate the
probability of an observation being within any range, not just (m ÷ σ) to (m + σ) and (m ÷ 2σ)
to (m + 2σ). These calculations are complex, but fortunately, there is a short cut available.
There are tables which set out these probabilities.
We provide one such table in the appendix to this study unit. It relates to the area in the tail of
the normal distribution.
Note that the mean, here, is denoted by the symbol "μ" (the Greek letter mu). Thus, the first
column in the table represents the position of an observation, x, in relation to the mean (x ÷
μ), expressed in terms of the standard deviation by dividing it by σ.
Figure 10.10: Proportion of distribution in an area under the normal distribution curve
The area under a section of the curve represents the proportion of observations of that size
in relation to the total number of observations. For example, the shaded area shown in Figure
10.10 represents the chance of an observation being greater than m + 2σ. The vertical line
which defines this area is at m + 2σ.
Looking up the value 2 in the table gives the proportion of the normal distribution which lies
outside of that value:
P(x > m + 2σ) = 0.02275
which is just over 2%.
Similarly, P(x > m + 1σ) is found by looking up the value 1 in the table. This gives:
P(x > m + 1σ) = 0.1587
which is nearly 16%.
You can extract any value from P(x = m) to P(x > m + 3σ) from this table. This means that
you can find the area in the tail of the normal distribution wherever the vertical line is drawn
on the diagram.
Negative distances from the mean are not shown in the tables. However, since the
distribution is symmetrical, it is easy to calculate these.
P(x < m ÷ 5σ) = P(x > m + 5σ)
So P(x > m ÷ 5σ) = 1 ÷ P(x < m ÷ 5σ)
This is illustrated in Figure 10.11.
m m + 2σ m ÷ 2σ m ÷ σ m + σ
302 Probability
Figure 10.11: Using the symmetry to determine the area under the normal curve
Further Probability Calculations
It is possible to calculate the probability of an observation being in the shaded area shown in
Figure 10.12, using values from the tables. This represents the probability that x is between
m ÷ 0.7σ and m + 1.5σ: i.e.
P(m ÷ 0.7σ < x < m + 1.5σ)
Figure 10.12: Probability of an observation being in a particular area
m ÷ 0.7σ m + 1.5σ
m m + 2σ m ÷ 2σ m ÷ σ m + σ
B A
Area A is read from tables.
Area B = Area A
m m + 2σ m ÷ 2σ m ÷ σ m + σ
C
Area C = 1 ÷ Area B.
This is because Area C and
Area B are mutually exclusive
and cover the whole range.
Probability 303
First find P(x > m + 1.5σ) = 0.0668 from the table.
Then find P(x < m ÷ 0.7σ) or P(x > m + 0.7σ)
= 0.2420 since the distribution is symmetrical.
The proportion of the area under the curve which is shaded in Figure 10.12 is then:
1 ÷ 0.0668 ÷ 0.2420 = 0.6912
Hence P(m ÷ 0.7σ < x < m + 1.5σ) = 0.6912
Example:
A production line produces items with a mean weight of 70 grams and a standard deviation of
2 grams. Assuming that the items come from a normal distribution, find the probability that an
item will weigh 65 grams or less.
65 grams is 5 grams below the mean. Since the standard deviation is 2 grams, this is 2.5
standard deviations below the mean.
Let x be the weight of an individual item. Therefore:
P(x < 65) = P(x < m ÷ 2.5σ)
= P(x > m = 2.5σ)
= 0.00621 from the tables.
Now find the probability that an item will weigh between 69 and 72 grams.
69 grams is 0.5 standard deviations below the mean, and 72 grams is 1 standard deviation
above the mean. Therefore we need to find P(m ÷ 0.5σ < x < m + σ):
P(x > m + σ) = 0.1587
P(x < m ÷ 0.5σ) = P(x > m + 0.5σ) = 0.3085
So
P(m ÷ 0.5σ< x < m + σ) = 1 ÷ 0.1587 ÷ 0.3085
or
P(69 < x < 72) = 0.5328
304 Probability
Questions for Practice 3
1. A factory which makes batteries tests some of them by running them continuously.
They are found to have a mean life of 320 hours and a standard deviation of 20 hours.
By assuming that this distribution is normal, estimate:
(a) The probability that a battery will last for more than 380 hours.
(b) The probability that it will last for less than 290 hours.
(c) The probability that it will last for between 310 and 330 hours.
(d) The value below which only 1 in 500 batteries will fall.
2. A machine produces 1,800 items a day on average. The shape of the distribution of
items produced in one day is very similar to a normal distribution with a standard
deviation of 80 items. On any one day:
(a) If more than 2,000 items are produced, the employees get a bonus. What is the
probability that a bonus will be earned?
(b) When less than 1,700 items are produced, overtime is available. What is the
probability that overtime will be worked?
(c) What is the probability that the workforce neither gets a bonus nor works
overtime?
Now check your answers with those given at the end of the unit.
Probability 305
Questions for Practice 1
Event A is "failed" and event B is "having less than 10 lessons". The calculation then is:
P(A|B) =
) B ( n
) B and A ( n
=
50
15
= 0.3
Questions for Practice 2
1. The probability of tails is 0.5 for each coin. The events are independent. Therefore:
P(3T) =
8
1
2
1
2
1
2
1
= × ×
2. (a) (i) Six points can be obtained from the following combinations:
1st die 2nd die
1 5
2 4
3 3
4 2
5 1
Since each die can fall with any of its faces uppermost, there are a total of
36 possible combinations, i.e. 6 × 6 = 36. Therefore:
P(six points) =
36
5
(ii) P(no six) =
36
25
6
5
6
5
= ×
P(at least one six) = 1 ÷ P(no six) =
36
11
(b) (i) P(too small) 02 . 0
000 , 10
200
= =
(ii) P(too large) 04 . 0
000 , 10
400
= =
(iii) Note that, if the first item is too large, it will be removed from the calculation
of the second item. Therefore:
P(both too large) = P(1
st
too large) × P(2
nd
too large)
= 0.04 ×
9,999
399
= 0.001596
(iv) The first item can be either too large or too small. Therefore:
P(one too large, one too small) = 0.04 ×
9,999
399
02 . 0
9,999
199
× +
= 0.001594
306 Probability
3. (a) (i) Each die must show a six. Therefore:
P(three sixes) =
216
1
6
1
3
= |
.
|

\
|
(ii) One die must show a six and the others must not show a six. Any one of
the three dice could be the one that shows a six.
Number of ways the die showing six can be chosen = 3
probability of six on one die = |
.
|

\
|
6
1
probability of not obtaining six on the other two dice =
2
6
5
|
.
|

\
|
Therefore:
P(one six) = 3 × |
.
|

\
|
6
1
×
2
6
5
|
.
|

\
|
=
72
25
(iii) Probability of obtaining at least one six
= 1 ÷ probability of not obtaining any sixes
= 1 ÷ probability that all three dice show 1, 2, 3, 4 or 5
= 1 ÷
216
91
216
125
1
6
5
3
= ÷ = |
.
|

\
|
(b) Required probability=
49
1
50
2
51
3
52
4
× × ×
= 0.000003694 to 4 significant figures.
(c) (i) Probability that first is red =
24
6
probability that second is red =
23
5
probability that third is red =
22
4
probability that all three are red = = × ×
22
4
23
5
24
6
0.009881 to 4 sig. figs.
(ii) Probability that the three will be all of any one of the four colours
= probability that all three are red
+ probability that all three are blue
+ probability that all three are green
+ probability that all three are white
=
22
4
23
5
24
6
4 × × × = 0.03953 to 4 sig. figs.
Probability 307
4. The contingency table will be as follows:
Output
up
Output
down
Total
Clocking in system 45 9 54
Employee completion
system
15 3 18
Total 60 12 72
(a) The probability of output per employee falling (event A) given that an automatic
clocking in system is used (event B) is:
P(A|B) =
) B ( n
) B and A ( n
=
54
9
=
6
1
(b) The probability of output per employee rising (event A) given that an automatic
clocking in system is used (event B) is:
P(A|B) =
) B ( n
) B and A ( n
=
54
45
=
6
5
The probability of output per employee rising (event A) given that an employee
completion system is used (event B) is:
P(A|B) =
) B ( n
) B and A ( n
=
18
15
=
6
1
The probabilities are the same.
5. Let: A = Machine A performing usefully, and A = Machine A not performing usefully
B = Machine B performing usefully, and B = Machine B not performing usefully.
We can draw a tree diagram to map the situation as follows:
Joint probabilities
3
1
x
4
3
=
12
3
3
2
x
4
3
=
12
6
3
2
x
4
1
=
12
3
3
1
x
4
1
=
12
1
4
1
A
A
B
B
3
1
3
2
B
B
3
1
3
2
4
3
Total = 1
308 Probability
(a) Probability of both machines operating is
12
1
(b) Probability of neither operating is
12
6
or
2
1
(c) Probability of only B operating is
12
3
or
4
1
(d) Probability of at least one machine operating is
2
1
12
3
12
2
12
1
= + +
6. Let: A = produced on machine A
B = produced on machine B
C = produced on machine C
D = defective
D = not defective.
The tree diagram for the situation is as follows.
(a) P(D) = 0.012 + 0.012 + 0.006 = 0.03
(b) P(defective that came from machine A) =
defective of y probabilit
A from defective a of y probabilit
=
03 . 0
012 . 0
= 0.4
Joint probabilities
0.30 x 0.04 = 0.012
0.60 x 0.98 = 0.588
0.60 x 0.02 = 0.012
A
Total = 1.000
0.30 x 0.96 = 0.288
0.10 x 0.06 = 0.006
0.10 x 0.94 = 0.094
0.94
0.06
0.04
0.96
0.98
0.02
B
0.10
D
D
D
D
D
D
C
0.30
0.60
Probability 309
7. The sample space diagram sets out all the possible total scores obtainable when two
dice are thrown. There are 36 equally likely outcomes:
1st die: 1 2 3 4 5 6
2nd die
1 2 3 4 5 6 7
2 3 4 5 6 7 8
3 4 5 6 7 8 9
4 5 6 7 8 9 10
5 6 7 8 9 10 11
6 7 8 9 10 11 12
(a) 26 outcomes give scores less than 9. Therefore:
P(total score less than 9) = 722 . 0
18
13
36
26
= =
(b) 4 outcomes give a score of 9. Therefore:
P(total score of 9) = 111 . 0
9
1
36
4
= =
Questions for Practice 3
1. (a) 380 hours is 60 hours or 3 standard deviations above the mean. Therefore:
P(observation > m + 3σ) = 0.00135 from the tables.
(b) 290 hours is 30 hours or 1.5 standard deviations below the mean. Therefore:
P(observation < m ÷ 1.5σ) = P(observation > m + 1.5σ) = 0.0668
(c) 310 hours is 0.5 standard deviations below the mean and 330 hours is 0.5
standard deviations above the mean.
P(observation > m + 0.5σ) = 0.3085 from the tables
Similarly, P(observation < m ÷ 0.5σ) = 0.3085
Therefore:
P(m ÷ 0.5σ < observation < m + 0.5σ) = P(310 < observation < 330)
= 1 ÷ 0.3085 ÷ 0.3085 = 0.383
Alternatively, using the symmetry of the normal distribution:
P(m < observation < m + 0.5σ) = 0.5σ ÷ 0.3085 = 0.1915.
Hence, P(m ÷ 0.5σ < observation < m + 0.5σ) = 2 × 0.1915 = 0.383
(d) Expressing 1 in 500 as a proportion:
1 in 500 = 002 . 0
500
1
=
310 Probability
Using the tables in reverse, and looking up the value 0.002 in the body of the
tables, the nearest we can get is 0.00199. This is when:
88 . 2
σ
μ x
= |
.
|

\
| ÷
Therefore, 1 in 500 batteries will be more than 2.88 standard deviations below
the mean.
Answer: = 320 ÷ 2.88 × 20 = 320 ÷ 57.6 = 262.4
Therefore 1 in 500 batteries will have a life of less than 262.4 hours.
2. Mean, m = 1,800
Standard deviation, σ = 80
(a) In a normal distribution, 2,000 items is:
80
1,800) (2,000÷
= 2.5 standard deviations from the mean
Therefore, the probability of 2,000 items being produced is:
P(x > m + 2.5σ) = 0.00621
(b) In a normal distribution, 1,700 items is:
80
1,800) (1,700÷
= ÷1.5 standard deviations from the mean
Therefore, the probability of 1,700 items being produced is:
P(x < m ÷ 1.25σ) = P(x > m + 1.25σ) = 0.1056
(c) Probability of no bonus and no overtime:
= 1 ÷ 0.1056 ÷ 0.00621 = 0.88819
Probability 311
Appendix: Areas in the Tail of the Normal Distribution
The table of values of the standard normal distribution set out below provides a means of
determining the probability of an observation (x) lying within specified standard deviations (o)
of the mean of the distribution (μ).
σ
μ) - (x
.00 .01 .02 .03 .04 .05 .06 .07 .08 .09
0.0 .5000 .4960 .4920 .4880 .4840 .4801 .4761 .4721 .4681 .4641
0.1 .4602 .4562 .4522 .4483 .4443 .4404 .4364 .4325 .4286 .4247
0.2 .4207 .4168 .4129 .4090 .4052 .4013 .3874 .3936 .3897 .3859
0.3 .3821 .3783 .3745 .3707 .3669 .3632 .3594 .3557 .3520 .3483
0.4 .3446 .3409 .3372 .3336 .3300 .3264 .3228 .3192 .3156 .3121
0.5 .3085 .3050 .3015 .2981 .2946 .2912 .2877 .2843 .2810 .2776
0.6 .2743 .2709 .2676 .2643 .2611 .2578 .2546 .2514 .2483 .2451
0.7 .2420 .2389 .2358 .2327 .2296 .2266 .2236 .2206 .2177 .2148
0.8 .2119 .2090 .2061 .2033 .2005 .1977 .1949 .1922 .1894 .1867
0.9 .1841 .1814 .1788 .1762 .1736 .1711 .1685 .1660 .1635 .1611
1.0 .1587 .1562 .1539 .1515 .1492 .1469 .1446 .1423 .1401 .1379
1.1 .1357 .1335 .1314 .1292 .1271 .1251 .1230 .1210 .1190 .1170
1.2 .1151 .1131 .1112 .1093 .1075 .1056 .1038 .1020 .1003 .0985
1.3 .0968 .0951 .0934 .0918 .0901 .0885 .0869 .0853 .0838 .0823
1.4 .0808 .0793 .0778 .0764 .0749 .0735 .0721 .0708 .0694 .0681
1.5 .0668 .0655 .0643 .0630 .0618 .0606 .0594 .0582 .0571 .0559
1.6 .0548 .0537 .0526 .0516 .0505 .0495 .0485 .0475 .0465 .0455
1.7 .0446 .0436 .0427 .0418 .0409 .0401 .0392 .0384 .0375 .0367
1.8 .0359 .0351 .0344 .0336 .0329 .0322 .0314 .0307 .0301 .0294
1.9 .0287 .0281 .0274 .0268 .0262 .0256 .0250 .0244 .0239 .0233
2.0 .02275 .02222 .02169 .02118 .02068 .02018 .01970 .01923 .01876 .01831
2.1 .01786 .01743 .01700 .01659 .01618 .01578 .01539 .01500 .01463 .01426
2.2 .01390 .01355 .01321 .01287 .01255 .01222 .01191 .01160 .01130 .01101
2.3 .01072 .01044 .01017 .00990 .00964 .00939 .00914 .00889 .00866 .00842
2.4 .00820 .00798 .00776 .00755 .00734 .00714 .00695 .00676 .00657 .00639
2.5 .00621 .00604 .00587 .00570 .00554 .00539 .00523 .00508 .00494 .00480
2.6 .00466 .00453 .00440 .00427 .00415 .00402 .00391 .00379 .00368 .00357
2.7 .00347 .00336 .00326 .00317 .00307 .00298 .00289 .00280 .00272 .00264
2.8 .00256 .00248 .00240 .00233 .00226 .00219 .00212 .00205 .00199 .00193
2.9 .00187 .00181 .00175 .00169 .00164 .00159 .00154 .00149 .00144 .00139
3.0 .00135
312 Probability

i

INTRODUCTION TO QUANTITATIVE METHODS
Contents
Unit Title Page iii v vii 1 3 4 5 7 11 13 22 26 29 31 39 40 40 43 50 55 63 64 64 68 70 76 79 89 90 91 96 98 106

Introduction to the Manual Syllabus Formulae, rules and tables provided with examination paper 1 Basic Numerical Concepts Introduction Number Systems Numbers – Approximation and Integers Arithmetic Dealing with Negative Numbers Fractions Decimals Percentages Ratios Further Key Concepts Algebra, Equations and Formulae Introduction Some Introductory Definitions Algebraic Notation Solving Equations Formulae Basic Business Applications Introduction Currency and Rates of Exchange Discounts and Commission Simple and Compound Interest Depreciation Pay and Taxation Simultaneous, Quadratic, Exponential and Logarithmic Equations Introduction Simultaneous Equations Formulating Problems as Simultaneous Equations Quadratic Equations Logarithmic and Exponential Functions

2

3

4

ii

Unit 5

Title Introduction to Graphs Introduction Basic Principles of Graphs Graphing the Functions of a Variable Graphs of Equations Gradients Using Graphs in Business Problems Introduction to Statistics and Data Analysis Introduction Data for Statistics Methods of Collecting Data Classification and Tabulation of Data Frequency Distributions Graphical Representation of Information Introduction Graphical Representation of Frequency Distributions Pictograms Pie Charts Bar Charts Scatter Diagrams General Rules for Graphical Presentation Measures of Location Introduction The Arithmetic Mean The Mode The Median Measures of Dispersion Introduction Range The Quartile Deviation, Deciles and Percentiles Standard Deviation The Coefficient of Variation Skewness Probability Introduction Estimating Probabilities Types of Event The Two Laws of Probability Conditional Probability Tree Diagrams Sample Space Expected Value The Normal Distribution

Page 119 120 121 124 126 141 147 171 172 173 175 177 185 193 194 194 204 206 207 209 210 221 222 223 232 236 249 250 251 251 255 262 263 275 276 277 279 280 286 288 294 295 298

6

7

8

9

10

iii

Introduction to the Study Manual

iv

v

Unit Title: Introduction to Quantitative Methods Level: 3 Learning Outcomes and Indicative Content: Candidates will be able to: 1. Demonstrate the rules of numeracy 1.1 1.2 1.3 2.

Unit Code: IQM Learning Hours: 100

Apply the four rules to whole numbers, fractions and decimals Express numbers in standard form Multiply and divide negative numbers

Apply calculations 2.1 2.2 Compare numbers using ratios, proportions and percentages; Obtain values for simple financial transactions involving purchases, wages, taxation, discounts p g Determine values for simple discounts and compound interest, and for purchases, wages, taxation, depreciation of an asset using the straight line method Determine values for simple and compound interest and the reducing balance method depreciation Convert foreign currency Make calculations using a scientific calculator including roots and powers; logarithms and exponential values Evaluate terms involving a sequence of operations and use of brackets Interpret, transpose and evaluate formulae Approximate data using rounding, significant figures

2.3 2 2.3 2.4 2.5 2.6 2.7 2.8 3.

Use algebraic methods 3.1 3.2 3.3 3.4 3.5 Solve linear and simultaneous equations Solve quadratic equations using factorisation and formulae Solve equations using roots or logarithms Determine the equation of a straight line through two points and also when given one point and its gradient Determine the gradient and intercepts on the x or y axes for a straight line

4.

Construct and use: graphs, charts and diagrams 4.1 4.2 4.3 4.4 Draw charts and diagrams derived from tabular data: eg bar charts, pie charts, scatter diagrams Plot graphs, applying general rules and principles of graphical construction including axes, choice of scale and zero Plot and interpret mathematical graphs for simple linear, quadratic, exponential and logarithmic equations Identify points of importance on graphs eg maximum, minimum and where they cut co-ordinate axes

vi .

usually annual) n = Number of time periods.vii INTRODUCTION TO QUANTITATIVE METHODS FORMULAE FOR BUSINESS MATHEMATICS AND STATISTICS INTEREST The formula for calculating compound interest: ⎛ r ⎞ A = P ⎜1 + ⎝ 100 ⎟ ⎠ n where: A = Accrued amount P = Original principal r = Rate of interest (for a particular time period. takes the following form: y = mx + c Note that: c = the y intercept (the point where the line crosses the y axis) m = the gradient (or slope) of the line . The expression of a linear function. DEPRECIATION ● Straight-line method: Cost of asset Annual depreciation = –––––––––––– Useful life (Cost of asset) – (Value at end of useful life) or Annual depreciation = –––––––––––––––––––––––––––––––––––– Useful life ● Reducing balance method: D = B (1 – i) n where: D = Depreciated value at the end of the n th time period B = Original value at beginning of time period i = Depreciation rate (as a proportion) n = Number of time periods (normally years) STRAIGHT LINE A linear function is one for which. and hence the formula of a straight line. when the relationship is plotted on a graph. a straight line is obtained.

⎛ p⎞ log ⎜ ⎟ = log p – log q ⎝q⎠ 3. If y = axn then n = (log y – log a) ÷ log x PROBABILITY ● Probability rules: Probability limits: Total probability rule: For complementary events: For two mutually exclusive events: For independent events: 0 ≤ P(A) ≤ 1 – P(A) + P(A ) = 1 P(A and B) = 0 P(A) = P(A|B) and/or P(B) = P(B|A) ΣP = 1 (for all outcomes) ● Multiplication rules: For independent events: For dependent events: P(A and B) = P(A)P(B) P(A and B) = P(A)P(B|A) ● Additional rules: For mutually exclusive events: For non-mutually exclusive events: P(A or B) = P(A) + P(B) P(A or B) = P(A) + P(B) – P(A and B) ● Conditional rules: P(A and B) P(A|B) = –––––––––– P(B) P(A and B) and P(B|A) = ––––––––– P(A) Expected value of variables x with associated probabilities P(x) is E(x) = ΣxP(x) . log(p × q) = log p + log q 2.viii QUADRATIC EQUATION A quadratic equation of the form ax 2 + bx + c = 0 can be solved using the following formula: x= – b ± b 2 – 4ac 2a RULES FOR LOGARITHMS 1. log pn = n log p 4.

● Standard deviation for grouped data: σ= ● ( ∑ mf ) ∑m f – 2 2 N N and s = ( ∑ mf ) ∑m f – 2 2 n n –1 Pearson’s measure of skewness: Mean – Mode 3(Mean – Median) Psk = ––––––––––––––– or –––––––––––––––– Standard deviation Standard deviation . ● – Mean for grouped data: μ = Σmf/N and x = Σmf/n where m is the midpoint and f is the frequency of a class. ● ● Range = Largest value – Smallest value.ix STATISTICAL MEASURES ● Mean for ungrouped data: μ = ∑ x x and x = ∑ n N – where N and x are the population and sample means respectively. where the number of ⎝ 2 ⎟ ⎠ = observations is odd and where n is the number of observations. Standard deviation for ungrouped data: σ= ∑x 2 (∑ x ) – N N 2 and s = ∑x 2 (∑ x ) – n n –1 2 where σ and s are the population and sample standard deviations respectively. ● Median for ungrouped data: ⎛ n + 1⎞ = Value of the ⎜ th observation in a ranked data set.

3085 .00307 .00820 .4562 .3897 .00570 .3264 .00391 .01191 .0 1.00135 .00587 .3 1.2148 .01255 .3821 .4325 .1112 .4247 .01130 .0681 .3228 .0548 .0351 .2611 .3156 .1020 .0418 .2266 .0239 .4721 .0384 .2389 .0516 .00187 .00256 .3 0.01 .2546 .01970 .2 1.5 2.01578 .00508 .1492 .0314 .0793 .0250 .00326 .2177 .00199 .01321 .00164 .00181 .1867 .01355 .2483 .0951 .00554 .1170 .1685 .00866 .1977 .0606 .00317 .01160 .00604 .2296 .4129 .01463 .2643 .00193 .4168 .0287 .1922 .2810 .01659 .00159 .01700 .01072 .01222 .3707 .7 1.3936 .1587 .01426 .3745 .1 1.5 0.2005 .0721 .01923 .0 2.1515 .1230 .00939 .0571 .0749 .6 1.00676 .4840 .3 2.0655 .02118 .6 2.4880 .02222 .02068 .00964 .4761 .2061 .0367 .1379 .02 .02018 .1814 ..4801 .0582 .04 .9 2.0392 .5000 .00494 .9 3.00240 .4286 .0853 .00169 .3632 .0 0.4 1.3557 .2743 .00539 .00226 .2090 .01287 .00219 .4522 .00695 .2843 .1841 .7 0.1562 .00175 .1 0.0694 .2451 .8 1.4404 .00280 .4 2.0630 .2912 .0281 .0301 .2327 .1949 .00798 .00714 .1314 .4443 .05 .1788 .1401 .4681 .4013 .4364 .0537 .6 0.01101 .00 σ 0.08 .2578 .02169 .1038 .0485 .1423 .00289 .00415 .0322 .2033 .02275 .0985 .00523 .1762 .00639 .3520 .0708 .4483 .0778 .1711 .01539 .00990 .4207 .0475 .1093 .00298 .1 2.1446 .00149 .01618 .0307 .1190 .1251 .3015 .0934 .3446 .00212 .0262 .00402 .0244 .4602 .1056 .0838 .2236 .0559 .3594 .1539 .2358 .00248 .0336 .4960 .2119 .4641 .0901 .0409 .0455 .2420 .1292 .0401 .00466 .2709 .2 2.0359 .1660 .00272 .01831 .1003 .2206 .0643 .7 2.0526 .3783 .x STANDARD NORMAL DISTRIBUTION The table of values of the standard normal distribution set out below provides a means of determining the probability of an observation (x ) lying within specified standard deviations (σ) of the mean of the distribution (μ ).00144 .1151 .00453 .5 1.00621 .0446 .0618 .3669 .2676 .01500 .0594 .3192 .00776 .0233 .3336 .00734 .3874 .00427 .00154 .0505 .00336 .3121 .0294 .0375 .00889 .0274 .2514 .01786 .01743 .0735 .3050 .00480 .0918 .00264 .00347 .00357 .2946 .2 0.00914 .0344 .1357 .0968 .07 .0465 .09 .0436 .1131 .9 1.1469 .0268 .1075 .1611 .0427 .0885 .0495 .1635 .00368 .06 .4920 .3372 .3300 .01017 .3483 .8 0.00139 .01044 .1736 .4 0.0329 .0764 .1894 .1271 .0808 .00657 .4090 .1335 .0 .0668 .3409 .2877 .00842 .03 .0869 .0823 .8 2.01390 .4052 .3859 .01876 .2981 .2776 .00440 .1210 .0256 .00233 . μ x AREAS IN TAIL OF THE STANDARD NORMAL DISTRIBUTION (x – μ) –––––.00755 .00205 .00379 .

Number Systems Denary Number System Roman Number System Binary System Numbers – Approximation and Integers Approximation Integers Arithmetic Addition (symbol ) Subtraction (symbol ) Multiplication (symbol ) Division (symbol ) The Rule of Priority Brackets D. C. (Continued over) © ABE and RRC . F. E.1 Study Unit 1 Basic Numerical Concepts Contents Introduction A. Dealing with Negative Numbers Addition and Subtraction Multiplication and Division Fractions Basic Rules for Fractions Adding and Subtracting with Fractions Multiplying and Dividing Fractions Decimals Arithmetic with Decimals Limiting the Number of Decimal Places Page 3 4 4 4 4 5 5 6 7 7 8 8 9 9 10 11 11 13 13 14 17 20 22 23 25 B.

Percentages Arithmetic with Percentages Percentages and Fractions Percentages and Decimals Calculating Percentages Ratios Further Key Concepts Indices Expressing Numbers in Standard Form Factorials 26 27 27 27 28 29 31 31 32 33 35 H. I.2 Basic Numerical Concepts G. Answers to Questions for Practice © ABE and RRC .

as well as being essential to understanding and solving many business problems. and vice versa add. subtraction. root. there will be plenty of practice questions for you to work through.Basic Numerical Concepts 3 INTRODUCTION The skill of being able to handle numbers well is very important in many jobs. carry out calculations using percentages and divide a given quantity according to a ratio explain the terms index. You should therefore get to know how to use one quickly and accurately. It is perfectly acceptable to use calculators to perform virtually all arithmetic operations – indeed. multiplication and division manipulate negative numbers cancel fractions down to their simplest terms change improper fractions to mixed numbers or integers. We shall consider this later in the course. Firstly though. Sometimes we can express numerical information more clearly in pictures – for example. Understanding how the principles work is essential if you are to use the calculator correctly. © ABE and RRC . It is likely that you will be familiar with at least some of these. However. Objectives When you have completed this study unit you will be able to:           identify some different number systems round-up numbers and correct them to significant figures carry out calculations involving the processes of addition. You may often be called upon to handle numbers and express yourself clearly in numerical form – and this is part of effective communication. as this is a practical subject. To get the most out of the material. but it is essential to ensure that you fully understand the basics before moving on to some of the more advanced applications in later units. in charts and diagrams. reciprocal and factorial. it is accepted practice in examinations. One initial word about calculators. you should also ensure that you know exactly how to perform all the same operations manually. subtract. In this study unit. we shall start by reviewing a number of basic numerical concepts and operations. we need to examine the basic processes of manipulating numbers themselves. Throughout the course. power. we recommend strongly that you do attempt all these questions. multiply and divide fractions and mixed numbers carry out calculations using decimals calculate ratios and percentages.

tabulation. there are: two hundreds (289) eight tens (289) nine units (289). we call this number system the base ten system or denary system.1: The denary system Seventh column Millions Sixth column Hundreds of thousands Fifth column Tens of thousands Fourth column Thousands Third column Hundreds Second column Tens First column Units The number of columns indicates the actual amounts involved. © ABE and RRC . We could indicate a positive number by a plus sign ().e. Here are some Roman numbers and their denary equivalent: XIV IX 14 9 XXXVI LVlll 36 58 Binary System As we have said. counting from 0 to 1 uses up all the symbols. so a new column has to be started when counting to 2. This is illustrated in the following table. Roman Number System We have seen that position is very important in the number system that we normally use. NUMBER SYSTEMS Denary Number System We are going to begin by looking at the number system we use every day. the dates on film productions. 0 and 1. symbols 0 to 9) are used. but in practice this is not necessary and we adopt the convention that. Figure 1. In this system. in this number. 20°C below zero). or the binary system. Roman numerals are still in occasional use in. This is equally true of the Roman numerical system. This means that. Numbers which are more than zero are positive. and similarly the third column at 4. By contrast.4 Basic Numerical Concepts A. 20°C (i. computers use base two. our numbering system is a base ten system. for example. etc. Although we invariably use the normal denary system for calculations. 73 means 73. We indicate a negative number by a minus sign (). Consider the number 289. A common example of the use of negative numbers is in the measurement of temperature – for example. we can note that it is possible to have negative numbers – numbers which are less than zero. but this system uses letters instead of figures. say. Before we look at other number systems. house numbers.e. Because ten figures (i. This uses just two symbols (figures).

There are strict mathematical rules for deciding whether to round up or down to the nearest figure: Look at the digit after the one in the last required column. so we would round up to 3. the last required column is the tens column and the digit after that is 5. The number 27. the answer would be 28. the digit after the one in the last required column is 3 (the last required column being the hundreds column) and the answer would be to round down to 200. NUMBERS – APPROXIMATION AND INTEGERS Approximation Every figure used in a particular number has a meaning. However. on some occasions. round down for figures of five or more. you need only express the distance to the nearest mile. Thus.000. (a) Rounding off Suppose you are given the figure 27. precision is not needed and may even hinder communication. when rounding off 236 to the nearest hundred.836 is more than this and so. B.Basic Numerical Concepts 5 Figure 1. if you are calculating the cost of the journey. The halfway point between the two is 27.500. Then:   for figures of four or less. When rounding off 3. When a number is transferred to the computer. © ABE and RRC . corresponding to "components" of the computer being either switched off or on.460. It is important to understand the methods of approximation used in number language.000 and 28.836 and asked to quote it to the nearest thousand. Clearly. For example. the figure lies between 27. to the nearest thousand. it is translated into the binary system.455 to the nearest ten.2: The binary system Denary 0 1 2 3 4 5 6 7 8 9 Binary 0 1 10 11 100 101 110 111 1000 1001 This system is necessary because computers can only easily recognise two figures – zero or one.000. round up.

The following are not integers: ¾. so that the digits keep their correct place values. Thus. we consider the second significant figure.017 correct to three significant figures becomes 2. 1½.002 has six significant figures – the zeros here are significant as they occur in the middle of the number.480. giving 2. This is likely to be done to aid the process of communication – i.25 By adjusting any number to the nearest whole number. in VAT calculations.820 8. Do not forget to include the insignificant zeros. the Customs and Excise rule is always to round down to the nearest one pence.) (b) Significant figures By significant figures we mean all figures other than zeros at the beginning or end of a number. etc. Consider the figure 2.290 5. you should note that different rules may apply in certain situations – for example. Thus:    632. are integers. we need to drop the fourth figure – i. the zero. We can round off a number to a specified number of significant figures.8.017 correct to one significant figure. Integers Integers are simply whole numbers.632 also has only three significant figures 630. 0. To write 2.000 © ABE and RRC . we obtain an integer. 4. Questions for Practice 1 1. 1. To write it correct to three significant figures.e. Thus.000 has three significant figures as the zeros are not significant 000. 3.e. 20. 1. as are all the negative whole numbers.231 6. Round off the following the figures to the nearest thousand: (a) (b) (c) 2. 7¼.570. We do this by using the rounding off rule – round down from 4 or round up from 5 – so 2. and apply the rule for rounding off. the 7. to make the number language easier to understand – and is particularly important when we consider decimals later in the unit.746 8.45.000.020. a number which is not an integer can be approximated to become one by rounding off. 2. 0. 3.017.6 Basic Numerical Concepts (However.501 Round off the following figures to the nearest million: (a) (b) (c) 6.236. This has four significant figures.

in fact. we shall note the names given to the various elements involved in a multiplication operation: © ABE and RRC . one ten – so adding that to the first number makes 14. The one that you borrow is.8 Basic Numerical Concepts Columns A 246 211 29 300 47 833 B 210 203 107 99 102 721 C 270 109 13 7 199 598 D 23 400 16 3 91 533 Totals 749 923 165 409 439 2. it is likely that you would use a calculator for this type of "long multiplication". Thus. Now consider this calculation: 34 18 16 Total Where it is not possible to subtract one number from another (as in the units column where you cannot take 8 from 4). Subtraction (symbol ) Again. which leaves 1. 2  4  8. starting from the right – the units column. the second column now becomes 3  (1  1). column by column. and subtracting 8 from that allows you to put 6 into the total. to subtract 21 from 36. Before doing so though. thus providing a full cross-check. the procedure is to "borrow" one from the next column to the left and add it to the first number (the 4). Moving on to the next column. it is essential to put back the one borrowed and this is added to the number to be subtracted. When larger numbers are involved. Multiplication (symbol ) The multiplication of single numbers is relatively easy – for example. but we shall briefly review the principles of doing it manually. the first task is to set the sum out: 36 21 15 Total As with addition. you must work column by column.685 Rows Totals Note that each row is added up as well as each column. starting with the units column. or 3  2. it is again very helpful to set out the calculation in columns: 246  23 In practice. So. it is important to set out the figures underneath each other correctly in columns and to work through the calculation.

it is 246 the multiplier is the number by which the multiplicand is to be multiplied – here it is 23 and the answer to a multiplication operation is called the product – so to find the product of two or more numbers. They may involve a number of different operations – for example: 863 8 2 When faced with such a calculation. 16  2  8. Division involving a single number divisor is relatively easy – for example. or 360/24. the calculation may be set out in a number of ways: 360  24. you multiply them together. it is 360 the divisor is the number by which the dividend is to be divided – here it is 24 and the answer to a division operation is called the quotient – here the quotient for our example is 15. or most commonly 360 24 Just as there were three names for the various elements involved in a multiplication operation. the task becomes more complicated and you should use a calculator for all long division. © ABE and RRC . here we have to remember that we are multiplying by tens and so we need to add a zero into the units column of this subtotal before entering the product for this operation (492). which is 2). Thus. Finally. the first operation is to multiply the multiplicand by the last number of the multiplier (the number of units. which is 3) and enter the product as a subtotal – this shown here as 738. Returning to our example of long multiplication. the subtotals are added to give the product for the whole operation: 246  23 738 4. the rule of priority in arithmetical operations requires that: division and multiplication are carried out before addition and subtraction.Basic Numerical Concepts 9    the multiplicand is the number to be multiplied – in this example. the actual product of 246  20 is 4.920.920 5. For larger numbers. Then we multiply the multiplicand by the next number of the multiplier (the number of tens. However. so there are three names or terms used in division:    the dividend is the number to be divided – in this example. The Rule of Priority Calculations are often more complicated than a simple list of numbers to be added and/or subtracted.658 Subtotal (units) Subtotal (tens) Total product Division (symbol ) In division.

8 4 2 then. then the innermost brackets are evaluated first.e. putting the results of the first step into the calculation. Thus. by three).e.10 Basic Numerical Concepts Thus. the procedure would be: (a) (b) first the multiplication and division: 8  6  48. Example: 2  3  (6  8  2)  2  3  (6  4) 232 26 8 Thus we can see that brackets are important in signifying priority as well as showing how to treat particular operations. we could have written the above calculation as follows: (8  6)  3  8 2 This makes it perfectly clear what numbers are to be multiplied. in the above example. Brackets Brackets are used extensively in arithmetic. It must be strictly followed except when brackets are included. we can do the addition and subtraction: 48  3  4  49. and we shall meet them again when we explore algebra. (multiplication before addition) (within the brackets. In that case. or multiply the eight by "six minus three" (i. for example: (12  3)  (3  6)  1 12  (3  3)  (6  1) (12  3)  3  (6  1) (12  [3  3])  6  1 Note that we can put brackets within brackets to further separate elements within the calculation and clarify how it is to be performed. by seven). We use brackets to separate particular elements in the calculation so that we know exactly which operations to perform on which numbers. The rule of priority explained above now needs to be slightly modified. Consider another example: 12  3  3  6  1 Again there are a number of ways in which this calculation may be interpreted. the division first) © ABE and RRC . Look at the previous calculation again. It is not really clear whether we should: (a) (b) (c) multiply the eight by six. the contents of the brackets are evaluated first. If there are brackets within brackets. or even multiply the eight by "six minus three plus four" (i.

we change the subtraction sign to a plus and treat the number as positive.) The order of the figures of these calculations does not matter. Start at 3 on the scale and count four divisions to the right: 4 3 2 1 0 1 2 3 4 Therefore. the process is slightly different. For example: 4  (3)  4  3  7 To sum up:   a plus and minus make a minus two minuses make a plus. Consider the following examples. All we do is extend the scale to the left of zero: 6 5 4 3 2 1 0 1 2 3 4 5 6 We can now visualise the addition and subtraction of negative numbers in the same way as above. we have shown the figure of "minus 3" in brackets to avoid confusion with the addition and subtraction signs. Negative numbers can also be represented on the number scale. therefore. here. With subtraction.) 7  (3)  7  3  10 2  (6)  2  6  4 © ABE and RRC . (3)  4  1. To subtract a minus number. In effect. (Note that.12 Basic Numerical Concepts 0 1 2 3 4 5 6 7 8 This illustrates the sum 6  4  2. a plus and a minus together make a minus. (Draw your own number scales to help visualise the sums if you find it helpful. so we can also write: (3)  4  4  (3)  1 We can illustrate the second sum as follows: 3 2 1 0 1 2 3 4 5 This is the same as 4  3. Consider the sum (3)  4.

the result of the division is a whole number and a fraction of a whole number. 6. 7  (4) 6  (8) (5)  (4) 10  (10) 2. The fraction is expressed as the remainder divided by the original divisor and the way in which it is shown. with a remainder of 1 The remainder is not a whole number. 7. the division of one whole number by another does not always result in a whole number. 8. 4. then. 10. (4)  7 (9)  3 (3)  (5) 3. with a remainder of 2 10  3  3. When two or more whole numbers are multiplied together. In contrast. E. 5. but a part of a whole number – so.Basic Numerical Concepts 13 Multiplication and Division The rules for multiplication and division are as follows: Type of calculation Minus times plus Minus times minus Minus divide by plus Minus divide by minus Plus divide by minus Product/Quotient minus plus minus plus minus Example 3  7  21 7  (3)  21 21  3  7 21  (3)  7 21  (3)  7 Questions for Practice 3 Solve the following sums. is in the same form as a division term: 2 12  5  2. FRACTIONS A fraction is a part of a whole number. in each case. or 5 © ABE and RRC . the product is always another whole number. using a number scale if necessary: 1. with a remainder of 2 parts of 5. Consider the following two examples: 12  5  2. 9. or two divided by five. 8  (6) (8)  (2) (7)  (7) Now check your answers with those given at the end of the unit.

there are three possible outcomes. However. as in the two examples above. 1 3 Thus. Basic Rules for Fractions  Cancelling down Both the numerator and the denominator in a fraction can be any number. We shall start by looking at these rules before going on to examine their application in performing arithmetic operations on fractions. and the denominators are 5 and 3. . The result of dividing out an improper fraction is a whole number plus a part of one whole. When two numbers are divided. though. For example: 5 551 5 Fractions are often thought of as being quite difficult. The result of dividing out is one whole part and no remainder. The answer must be 1. Examples of proper fractions are: 1 2 4 4 . 5 3 7 5 (c) The numerator is the same as the denominator In this case.14 Basic Numerical Concepts 10  3  3. the numerators are 2 and 1. is always to reduce the fraction to the smallest possible whole numbers for both its numerator and denominator. . or Each part of the fraction has a specific terminology:   the top number is the numerator and the bottom number is the denominator. The result of dividing out a proper fraction is only a part of one whole and therefore a proper fraction has a value of less than one. . for the above fractions. or 4 9 9 30 6  1 whole part and a remainder of 6 parts of 24. © ABE and RRC . 5 50 93 9 The rule in expressing fractions. etc. the fraction is known as an improper fraction. Taking two more examples: 42 6  4 whole parts and a remainder of 6 parts of 9. the fraction is known as a proper fraction. or one divided by three. . the expression is not a true fraction. or 1 24 24 (b) The denominator is larger than the numerator In this case. . (a) The numerator is larger than the denominator In this case. The resulting fraction is then in its lowest possible terms. The following are all valid fractions: 2 20 47 6 . they are not really hard as long as you learn the basic rules about how they can be manipulated. with a remainder of 1 part of 3.

2 8 could be expressed as by multiplying both terms by 4. and 10. To do this. we follow another simple rule. to express ½ in eighths – the rule is to divide the required denominator by the existing denominator (8  2  4) and then multiply both terms of the original fraction by this result: 1 4 4  24 8 © ABE and RRC . large vulgar fractions can often be reduced in stages to their lowest possible terms. However. so we should use 10. Thus: 20 2  50 5 A fraction which has not been reduced to its lowest possible terms is called a vulgar fraction. we can do that by multiplying both the numerator and denominator by the same number. and we can now see that both terms can be further divided – this time by 11: 55  11 5  77  11 7 So. but you are very unlikely to see that straight away. 5. So. you would probably quickly see that both are divisible by 3. Consider the following fraction from those listed previously: 20 50 Which whole number will divide into both 20 and 50? There are several: 2. Dividing both parts of the fraction by 10 reduces it to: 2 5 Note that we have not changed the value of the fraction. 20 5 If we wanted to change the denominator to a specific number – for example. just the way in which the parts of the whole are expressed. Cancelling down can be a tricky process since it is not always clear what the largest whole number to use might be. so you could start the process of cancelling down by dividing by three: 165  3 55  77 231  3 This gives us a new form of the same fraction. Divide both the numerator and denominator by the largest whole number which will divide into them exactly. If we wanted to express a given fraction in a different way. However the rule states that we use the largest whole number. for example. consider the following fraction: 165 231 In fact.  Changing the denominator to required number The previous process works in reverse.Basic Numerical Concepts 15 The process of reducing a fraction to its lowest possible terms is called cancelling down. both terms can be divided by 33. For example.

then add the numerator of the fraction to this product and place the sum over the denominator. changing the three whole ones into nine thirds then add the numerator (2) to this  2  9  11. thus adding the two thirds to the nine thirds from the first operation finally place the sum over the denominator  11 3 Questions for Practice 4 1.16 Basic Numerical Concepts Note again that we have not changed the value of the fraction. The rule is to multiply the whole number by the denominator in the fraction. just the way in which the parts of the whole are expressed.  Changing a mixed number into an improper fraction This is the exact reverse of the above operation. Thus we find out how many whole parts there are and how many parts of the whole (in this case. to write 3 (a) (b) (c) 2 as an improper fraction: 3 first multiply the whole number (3) by the denominator (3)  3  3  9. For example. Reduce the following fractions to their lowest possible terms: (a) (c) (e) (g) (i) 9 36 21 70 14 49 36 81 24 192 8 64 18 54 25 150 45 135 33 88 (b) (d) (f) (h) (j) © ABE and RRC . eighths) are left over: 35 3  35  8  4 and a remainder of three  4 8 8 This result is known as a mixed number – one comprising a whole number and a fraction.  Changing an improper fraction into a whole or mixed number Suppose we are given the following improper fraction: 35 8 The rule for converting this is to divide the numerator by the denominator and place any remainder over the denominator.

In this case.Basic Numerical Concepts 17 2. thus making its denominator 10 – the same as the second fraction: 4 3 7   10 10 10 © ABE and RRC . For example: 3 2 5   7 7 7 5 1 4 1    8 8 8 2 Where the denominators are not the same. This process is called finding the common denominator. Where the denominators are the same. It is not possible to do these operations if the denominators are different. Change the following mixed numbers to improper fractions: (a) (c) 2 4 (b) (d) 3 9 Now check your answers with those given at the end of the unit. Change the following fractions as indicated: (a) (c) 1 to 10ths 5 4 to 81sts 9 12 5 25 6 2 3 1 8 (b) (d) 3 to 20ths 5 9 to 88ths 11 17 8 31 3 3 8 3 4 3. Consider the following calculation: 2 3  10 5 We cannot add these two fractions together as they stand – we have to find the common denominator. Adding and Subtracting with Fractions (a) Proper fractions Adding or subtracting fractions depends upon them having the same denominator. we can change the expression of the first fraction by multiplying both the numerator and denominator by 2. we need to change the form of expression so that they become the same. adding or subtracting is just a simple case of applying the operation to numerators. Change the following improper fractions to mixed numbers: (a) (c) (b) (d) 4.

1 1 3 5 2 4 2 4 Consider the following calculation: 34  1 1 3  (3  4  5  2)      4 2 4  1 2 3  14      4 4 4  14  6 4 1 2  14  1  15 1 2 In general.18 Basic Numerical Concepts Not all sums are as simple as this. we deal with the integers and the fractions separately. and then add together the sum of the integers and the fractions. the same principles are applied to subtracting mixed numbers. Mixed numbers When adding mixed numbers. We must then convert all three to the new denominator by applying the rule explained above for changing a fraction to a required denominator: 3 1 1 12 5 2 15 3        10 20 20 20 20 5 4 4 The rule is simple:    (b) find the lowest common denominator – and where this is not readily apparent. For example: 4 5 1 2 1  5 1 5 3  1  (4  1)      3      3  3 6 2 6 3  6 2 6 6 © ABE and RRC . The procedure is as follows:    first add all the integers together then add the fractions together as explained above. It is often the case that we have to change the form of expression of all the fractions in a sum in order to give them all the same denominator. Consider the following calculation: 3 1 1   10 5 4 The lowest common denominator for these three fractions is 20. a common denominator may be found by multiplying the denominators together change all the fractions in the calculation to this common denominator then add (or subtract) the numerators and place the result over the common denominator.

the procedure is slightly different where the second fraction is larger than the first. 3. 8. 11. 6. 7. Consider this calculation: 4 1 5 1 2 6 Here. 12. © ABE and RRC . it is not possible to subtract the fractions even after changing them to the common denominator: 3 5  6 6 To complete the calculation we need to take one whole number from the integer in the first mixed number.Basic Numerical Concepts 19 However. Thus: 4 1 1   3  1    3  2 2  9 6 3    3 6 6 6 We can now carry out the subtraction as above: 4 1 5 9 5 4 2 9 5  1  3  1  (3  1)      2  2 2 6 6 6 6 3 6 6 Questions for Practice 5 Carry out the following calculations: 1. 17. 1 1  3 4 1 2  8 3 2 3  5 4 5 1  6 3 5 1  7 3 1 1  10 6 1 1  5 6 2 1  3 4 7 1  8 5 3 3  4 8 2. 10. 15. 2 1 3 1 2 4 4 1 6 4 3 3 1 2 1 8 4 2 3 2  4 3 1 1 2 8 3 3 2 1 1 2 4 3 2 2 1 1 31 2 3 2 5 4 2 5 1 2 9 3 1 1 1 2 2 8 3 6 Now check your answers with those given at the end of the unit. 18. convert it to a fraction with the same common denominator and add it to the fraction in the mixed number. 4. 16. 9. 5. 14. 13.

we must complete one further step – convert the mixed number to an improper fraction. Before we can apply the same methods to mixed numbers.000. This is just another way of expressing multiplication. For example: 3 2 32 6 1     12 4 3 43 2 3 3 33 9    40 8 5 85 (d) Dividing by a fraction In this case. we simply multiply the numerator by the whole number and place the product (which is now the new numerator) over the old denominator. we turn the divisor upside down and multiply. (b) Dividing a fraction by a whole number In this case. For example: 5 4 1 4 11 4 44     2 11 2 11 22 2 3 1 19 2 38 1     9 4 4 2 1 4 2 © ABE and RRC . and then reduce the result to the lowest possible terms. we multiply the numerators by each other and the denominators by each other. For example: 2 25 10 1 5  3 3 3 3 3 3 37 21 1 7  5 4 4 4 4 We often see the word "of" in a calculation involving fractions – for example.20 Basic Numerical Concepts Multiplying and Dividing Fractions (a) Multiplying a fraction by a whole number In this case. ¼ of £12. we simply multiply the denominator by the whole number and place the old numerator over the product (which is now the new denominator). For example: 4 3 4 16 1 4  5 3 4 3 3 1 1 1 3 3 1      6 3 6 1 6 2 (e) Dealing with mixed numbers All that we have said so far in this section applies to the multiplication and division of proper and improper fractions. For example: 3 3 3 5  20 4 45 2 2 2 5  3 35 15 (c) Multiplying one fraction by another In this case.

you can use cancellation during a division calculation at the step when. having turned the divisor upside down. This involves dividing both the numerator and the denominator by the same number. for example: 6 3 cancels down to by dividing both terms by 2. you are multiplying the two fractions. In this case it can be applied across the multiplication sign – dividing the numerator of one fraction and the denominator of the other by the same number. even if you make a mistake in the arithmetic. we could show it as follows: 4 3 1 2 1 19 1 19     9 4 2 4 2 1 2 2 © ABE and RRC . they will not be able to follow the process and see that the error is purely mathematical.Basic Numerical Concepts 21 (f) Cancelling during multiplication We have seen how it is possible to reduce fractions to their lowest possible terms by cancelling down. It is also the case that. examiners will give "marks for workings". However. So. in an examination. Some calculations can be very long. too.) Cancellation can also be used in the multiplication of fractions. involving many steps. Thus. the following operation would be wrong: 3 3 1   4 1 12 3 3 Note. it is much easier to identify the error if all the workings are shown. 8 4 We often show the process of cancelling down during a calculation by crossing out the cancelled terms and inserting the reduced terms as follows: 2 8 2  12 3 3 (It is always helpful to set out all the workings throughout a calculation. rather than in the steps you have gone through. if you do not show all the workings. However. repeating the calculation involving mixed numbers from above. If you make a mistake somewhere. that you cannot cancel across addition and subtraction signs. For example: 3 1 1 1   12 4 16 4 Remember that every cancelling step must involve a numerator and a denominator.

the part of the whole is shown by a number following a decimal point:   the first number following the decimal point represents the number of tenths.382  5 1000 1000 1000 0. fractions with a different denominator are also relatively easy to convert by simple division. 7. again if there is one.67 is six-tenths and seven-hundredths (i.054  . 8 6 2 6 3 7 3 8 3 6 4 5 4 7 Now check your answers with those given at the end of the unit. decimals are written completely differently and this makes carrying out arithmetic operations on them much easier.134 is 134 thousandths of a whole  and so on up to as many figures as are necessary.005 is five-thousandths of a whole and 0. represents the number of thousandths. ten thousand. etc.02 is two-hundredths of a whole and 0. 0. so 0. Unlike fractions though. 0. so 0. represents the number of hundredths.81  3 100 100 100 1 54 382 . so 0. Decimals do not have a numerator or a denominator. F.6  1 10 10 10 1 67 81 . However. thousandths. 4. can be easily converted into a decimal.001  We can see that any fraction with a denominator of ten.3  . DECIMALS Decimals are an alternative way of expressing a particular part of a whole. 6. and decimals are effectively fractions expressed in tenths. 8. one thousand.1  1 3 6 . The term "decimal" means "in relation to ten". Rather. if there is one. 1.22 Basic Numerical Concepts Questions for Practice 6 Carry out the following calculations: 1. 3. 0. 3. etc.01  0.e.3 is three-tenths of a whole the next number to the right. one hundred.67  . 67 hundredths) of a whole the next number to the right. 7 5 3 4 3 6 2 3 3 7 2. hundredths. 5. 5.1 is one-tenth part of a whole and 0. We can see the relationship between decimals and fractions as follows: 0. © ABE and RRC .

3  5). we insert the decimal point two places to the right in the answer: 0. simply move the decimal point one place to the left – for example. (c) Division with decimals Again. to multiply by 100. so we abbreviate the decimal to a certain size – as we shall discuss in the next section.e. but in most instances it is easier to use a calculator. you simply move the decimal point one place to the right.2 Likewise.6 However. For example: 3. For example.3333333 or however many 3s for which there is room on the paper or the calculator screen. to change 3 into decimal form. Thus. In most business applications. To change a decimal into a fraction. it is far easier to use a calculator when dividing with decimals.32  100  0. the decimal becomes the numerator and the denominator is 10 or 100 etc. It is simply a process of dividing the numerator by the denominator and expressing the answer in decimal form. etc. not all fractions will convert so easily into decimals in this way.736  3 © ABE and RRC . enabling quick manual calculations to be done with ease. when dividing by ten or by one hundred (or even by a thousand or a million). 68.6832. is known as a recurring decimal. simply move the decimal point two places to the left – for example. can also be converted into decimals.62  100  362 This can be a very useful technique.2 One final point to note about multiplying decimals is that. and vice versa Fractions other than tenths. where one or more numbers repeat infinitely. we can use the technique noted above in respect of multiplication to find the answer quickly without using a calculator:   (d) to divide by ten.832 to divide by one hundred. 3 Dividing this out (or using a calculator) gives the answer 0. depending upon the size of the decimal. you divide 3 by 5 (i. thousandths. you move the decimal point two places to the right: 3. You can do this manually. However. but this time there are 2 figures after decimal points in the original sum. Consider the fraction 1 . So.36  36 9  25 100 736 92 3 1000 125 3. For example: 0. 5 Using a calculator the answer is easily found: 0. Converting fractions to decimals. hundredths.62  10  36.32  10  6.24 Basic Numerical Concepts The multiplication would be the same. 68. to multiply a decimal by ten. This type of decimal. we simply place the decimal over ten or one hundred etc. we do not need such a degree of precision.

This is usually set at two or three decimal places.182 12. "0. We then speak of a number or an answer as being "correct to two (or three) decimal places".286. we use the principles of rounding discussed earlier in the study unit to reduce them to the correct size. Following the rule for rounding. although in engineering.42 10. but you should note that the level of accuracy suffers. which convert into 0. Now consider the decimal forms of one-third and two-thirds. if we want to express the decimal form of two-sevenths (0. Such large decimals also present problems in carrying out arithmetic procedures – for example. in business applications. round up the figure in the last required place for figures of four or less.263 8. multiplying 0. there are also plenty of other fractions which do convert into exact decimals. we would normally just quote the decimal as. Working to only two or three decimal places greatly simplifies decimals.Basic Numerical Concepts 25 Limiting the Number of Decimal Places We noted that certain fractions do not convert into exact decimals.792 7. For example: 2  0. that level of precision may be critical.33 and 0. (When referring to these types of recurring decimals.33 by two gives us 0. we look at the digit in the decimal after the one in the last required place and then:   for figures of five or more.84  3. respectively.6470568 17 This gives us a problem: how much detail do we want to know? It is very rare that.81  3. For example.2857142) correct to three decimal places. In such cases.08  11. but do not..67 respectively.32 11. we would want to know an answer precise to the millionth part of a whole.67. These types of small error can increase when adding.3 recurring".36  2. we look at the digit in the fourth decimal place: 7. For decimals with a larger number of decimal places than the required number.43 © ABE and RRC .2 7. we would then round up the number in the third decimal place to give 0. If we were to express these correct to two decimal places.16  3. So. subtracting. we limit the number of decimal places to a manageable number. it is not easy to add the above two decimals together and multiplying them does not bear thinking about...14  8.79  6.07  8.. The most common examples are one-third and two-thirds. but only at the level of thousandths or greater.67  4. we need to draw attention to the problem by stating that "errors are due to rounding".3333333.6  3. Questions for Practice 7 1.36  4. these would be 0. multiplying or dividing several decimals which have been rounded – something which we shall see later in the course in respect of a set of figures which should add up to 100. Carry out the following calculations: (a) (c) (e) (g) 3.6666666.2857142. or 7 11  0.8  4. say. round down by leaving the figure in the last required place unchanged.66. but we have shown two-thirds as being 0. and 0. In most business circumstances then.) However.209 (b) (d) (f) (h) 8. so we are only interested in the level of detail down to hundredths or thousandths. Thus.

The per cent sign is %. © ABE and RRC .63  5.125 1.60 0.3 0. The expression "20 per cent" means 20 out of 100.750 Now check your answers with those given at the end of the unit.75 (b) (d) 0.07 (j) (l) (n) (p) 4.1 154  2. correct to the number of decimal places (dp) specified: (a) (c) (e) (g) 2 to 2 dp 5 3 to 2 dp 4 5 to 3 dp 8 4 to 3 dp 7 (b) (d) (f) (h) 3 to 3 dp 8 7 to 3 dp 8 5 to 3 dp 6 5 to 3 dp 16 3.052 Change the following fractions into decimal form.2 11.592  2.8  4.26 Basic Numerical Concepts (i) (k) (m) (o) 2. so when we say "percentage".7 2.8 5. PERCENTAGES A percentage is a means of expressing a fraction in parts of a hundred – the words "per cent" simply mean "per hundred".72  1.8  3. Some examples:   4% means four-hundredths of a whole. G. 3. we mean "out of a hundred".62  2. or a hundred parts of a hundred  100 1 100  Percentages are used extensively in many aspects of business since they are generally more convenient and straightforward to use than fractions and decimals. It is therefore important that you are fully conversant with working with them.6  0. so 20 per cent is written as 20%. or twenty-eight parts of a hundred  28 100 100% means one hundred-hundredths of a whole. or four parts of a hundred  4 100 28% means twenty-eight-hundredths of a whole. Change the following decimals into fractions and reduce them to their lowest possible terms: (a) (c) 0.2 2.

These processes are quite straightforward.5% 8 8 2 200  100  %  66. For example: 3 300  100  %  37½% or 37. In effect all this means is that you place the percentage over 100 and then reduce it to its lowest possible terms. Here are some of them: 12½%  33. move the decimal point two places to the right: 0. what percentage remains? Here the two percentages are not parts of the same whole – one refers to cake and the other to pie. To change a decimal to a percentage.Basic Numerical Concepts 27 Arithmetic with Percentages Multiplication and division have no meaning when applied to percentages. again you divide it by 100.67% 3 3 As you become familiar with percentages you will get to know certain equivalent fractions well. this is easy – you simply move the decimal point two places to the left. Therefore.36%  36% © ABE and RRC . As we saw in the last section. Take an example: if Peter spends 40% of his income on rent and 25% on household expenses. Thus: 11. For example: 60%  24%  60 2  100 5 24 6  25 100 Changing a fraction into a percentage is the reverse of this process – simply multiply by 100.5%  0.3%  1 8 1 3 20%  50%  1 5 1 2 25%  75%  1 4 3 4 Percentages and Decimals To change a percentage to a decimal. so we need only be concerned with their addition and subtraction. consider this example: if Alan eats 20% of his cake and 10% of his pie.115 Working the conversion the other way is equally easy. Percentages and Fractions Since a percentage is a part of one hundred. but you need to remember that we can only add or subtract percentages if they are parts of the same whole. what percentage of his income remains? Here both percentages are parts of the same whole – Peter's income – so we can calculate the percentage remaining as follows: 100%  (40%  25%)  35% However. to change a percentage to a fraction you simply divide it by 100. it is not possible to add the two together.

as given above. how much did she buy it for? The cost plus the profit  120% of cost  £1.5% 2 There are occasions when we deal with percentages greater than 100%.200. we need to express both quantities in the same units before working out the percentage. to find 15% of £260: £260  or 15 £3900   £39 100 100 £260  0. For example. change the percentage rate into a fraction or a decimal and then multiply the given number by that fraction/decimal.000  0.000 120 100 : 120 To prove that this is correct. if a woman sells a car for £1.000 and she sold it at 20% profit. so the selling price is: £1.200 and says that she made 20% profit. we work the calculation through the other way. what percentage of cars have been sold if 27 have gone from a total stock of 300? 27 2700  100   9% 300 300 Do not forget that percentages refer to parts of the same whole and therefore can only be calculated where the units of measurement of the quantities concerned are the same.15  £39 To express one number as a percentage of another.200.23  100  11. we need to multiply the selling price by Cost  1.000  20% of £1. © ABE and RRC . to show 23 pence as a percentage of £2. For example. The cost was £1.000)  £1.5% 200 or 0. For example. and then change the fraction into a percentage by multiplying by 100. The cost. is 100% To work out the cost. This could be both as pence or both as pounds: 23  100  11. For example.28 Basic Numerical Concepts Calculating Percentages To find a particular percentage of a given number.000  £(1.200  100  £1. we convert the numbers into a fraction with the first number as the numerator and the second the denominator. therefore.2  1.

– or the comparison will not be valid. Note that the ratio itself is not in any particular unit – it just shows the relationship between quantities of the same unit. Then re-express the ratio in the form of numerator : denominator. we would have to convert the two quantities to the same units before giving the ratio. To do this.790 4. RATIOS A ratio is a way of expressing the relationship between two quantities. if we wanted to give the ratio of the width of a table to its depth where the width is 2 m and the depth is 87 cm.5% of £138 (b) (d) 25% of 360° 27% of £1. It is essential that the two quantities are expressed in the same units of measurement – pence. H. Calculate the following: (a) (c) (e) 15% of £200 12½% of 280 bottles 17. so the correct form of showing the above ratio would be: 200: 87 (pronounced "200 to 87") There are two general rules in respect of ratios:  They should always be reduced to their lowest possible terms (as with vulgar fractions). 3 100 3 8 (b) Change the following percentages into fractions and reduce them to their lowest possible terms (a) (c) 2% 7½% (b) (d) 8% 18% 3.Basic Numerical Concepts 29 Questions for Practice 8 1. etc. number of people.000 (b) (d) 25 as a percentage of 75 30 as a percentage of 20 Now check your answers with those given at the end of the unit. Change the following fractions into percentages: (a) 2. express the ratio as a fraction by putting the first figure over the second and cancel the resulting fraction. Expressing both in centimetres would give the ratio of 200 to 87. Calculate the following: (a) (c) 60 as a percentage of 300 £25 as a percentage of £1. We use a special symbol (the colon symbol : ) in expressing ratios. © ABE and RRC . For example.

000. Y and Z share profits in the ratio of 3 : 4: 5. Now check your answers with those given at the end of the unit.000 A B C D 2.000 75. The number of times that the multiplication is repeated is indicated by a superscript number to the right of the number being multiplied. The number being multiplied (here.000 15.) The power of a number is the result of multiplying that number a specified number of times.200. FURTHER KEY CONCEPTS In this last section we shall briefly introduce a number of concepts which will be used extensively later in the course.000 63. Thus in the case of 26 we refer to the term as "two raised to the sixth power" or "two to the power of six". However. © ABE and RRC .Basic Numerical Concepts 31 Questions for Practice 9 1.000 10. Thus: 2  2 is written as 22 (the little "2" raised up is the superscript number) 2  2  2 is written as 23 2  2  2  2  2  2  2 is written as 27 etc. 22 should be referred to as "two to the power of two". Total profit is £7. similarly. we do include practice in relation to the expression of numbers in standard form. (Strictly.000 Profits £ 25. G and H share profits in the ratio of 2 : 3 : 3. Total profit is £60. but we normally call it "2 squared". 2) is termed the base and the superscript number is called the exponent or index of its power. 8 is the third power of 2 (23) and 81 is the fourth power of 3 (34). 23 is referred to as "2 cubed".000 30.000 7. I. Indices Indices are found when we multiply a number by itself one or more times. Express the following sales and profits figures as profit to sales ratios for each firm: Firm Sales £ 100. Here we concentrate mostly on what these concepts mean. and you will get the opportunity to practice their use in later study units. F. Thus. Calculate the distribution of profits for the following partnerships: (a) (b) X.

help is at hand.43605  1014. the fourth root of 16 is 2 (2  2  2  2  16). It is. and it is helpful if we can find a different way of expressing them which will make them easier to deal with.000.32 Basic Numerical Concepts It is quite possible to have a negative exponent. customary to omit the 2 from the root sign when referring to the square root. under this system are: 9. which is 6. The above two numbers. followed by the number of 10s you need to multiply it by in order to make it up to the correct size. The method of expressing numbers in standard form reduces the number to a value between 1 and 10. as in the case of 23.605. However. as in the case of 16¼.825. let alone multiplying it by 843.000. Written out in full. Thus 2 9  3 (3  3  9) The root index (the little 4 or little 2 in the above examples) can be any number. Thus. Expressing Numbers in Standard Form The exploration of indices leads us on to this very useful way of expressing numbers.000.000.3825  1013 and 8. is equal to the first number. though. It is more usual though to refer to this as the fourth root of the base and to write it as follows: 4 16 The symbol  indicates a root. It is a very big number and you would probably have great difficulty stating it. A square root of any number is the number which. The 13 and 14 after the 10 are index numbers. A negative exponent indicates a reciprocal. Consider the number 93. In fact. A root is the number that produces a given number when raised to a specified power. This indicates that the base is raised to the power of a quarter. when multiplied by itself. Working with very large (or very small) numbers can be difficult. can you even tell which is the bigger? Probably not. Thus 36 refers to the square root of 36. It is made a little easier by the two numbers being very close to each other on the page. but if one was on a different page. Standard form provides such a way. the first number would be: 9. The reciprocal of any number is one divided by that number.125 3 8 2 Indices also do not have to be whole numbers – they can be fractional. © ABE and RRC .3825  (10  10  10  10  10  10  10  10  10  10  10  10  10).000. it would be very difficult indeed. For example: the reciprocal 5 is 23 represents 1 and 5 1 1   0. as discussed above.000.

We can go on multiplying 10 by itself to get: 103  1. 4 factorial is written as 4! (and is sometimes read as "four bang") and is equal to 4  3  2  1. Thus. we can see immediately that 8.3825  1013.001. Thus: 101  1  0. etc.000.000 104  10. For example.605. Before we consider this in more detail.43605  1014 is larger than 9. one divided by the number as many times as the index specifies. © ABE and RRC . 2.01 and 103  0.000.0243. 4 3 6 0 5 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Converting numbers in this way makes arithmetic operations much easier – something which we shall return to later in your studies – and also allows us to compare numbers at a glance.000. 10  843. Factorials A factorial is the product of all the whole numbers from a given number down to one. Now you should be able to work out that 2. You may encounter factorials in certain types of statistical operations. it indicates a reciprocal – i. We also know from the above that. when there is a minus number in the index. written out in full would be: 8 .e. So. the second number from above. We know that 10  10  100. Again.Basic Numerical Concepts 33 This is not particularly useful.000 We could go on to say that 102  0. so an easier way of dealing with a number expressed in standard form is to remember that the index number after the 10 tells you how many places the decimal point has to be moved to the right in order to create the full number (remembering also to put in the 0s).1. so we can say that 102  100. etc. you will note that a minus index number after the 10 tells you how many places the decimal point has to be moved to the left in order to create the full number (remembering also to put in the 0s as well). it would be helpful to recap exactly what the index number means.43  102  0. Just as with indices we can also have numbers expressed in standard form using a negative index – for example.43  102.

Express the following numbers in standard form: (a) (b) (c) (d) (e) (f) 2.853  106 6.34 Basic Numerical Concepts Questions for Practice 10 1.653  1012 8.02  105 7.300.16  1010 Now check your answers with those given at the end of the unit.35  102 1.045 6. (a) (b) (c) (d) (e) (f) 4. 234 0.00000000000000003792 Write out in full the following numbers which are expressed in standard form.000 0.68  102 9. © ABE and RRC .420 0.0005032 845.

000.880 9.000 (b) 8. 9. 4. 5.Basic Numerical Concepts 35 ANSWERS TO QUESTIONS FOR PRACTICE Questions for Practice 1 1.000 5.000. 3 2 6 15 100 © ABE and RRC . 10.890 Four Four (b) (d) (b) (d) 3.000 Questions for Practice 2 1. 6.000 7.412 Totals 3. 4. The total of the check column should equal the sum of the four column totals.730 Three Two (b) 8. A 322 219 169 210 920 B 418 16 368 191 993 C 42 600 191 100 933 D 23 46 328 169 566 Check 805 881 1. 8. (a) (b) (c) (d) (e) (f) (6  3)  6  4  9  6  4  54  4  58 (6  3)  (6  4)  9  10  90 10  (5  2)  (10  5)  10  7  2  70  2  68 24  24  (2  2)  24  24  4  24  6  30 6  (24  [4  2])  6  (24  8)  6  3  18 (15  [3  2]  6  [4  3]  2)  11  (15  5  6  12  2)  11  (3  6  24)  11  33  11  3 Questions for Practice 3 1. (a) (c) (a) (c) (a) (c) (a) (c) 4. 7.140 554 The check on your answers is achieved by adding a column into which you put the sum of each row.840 7. 2.157 952 (b) (d) 232.056 670 3. 2.000. 3.000 2. 3.000 7. (a) (c) 57. 3 14 6 9 1 2.

4 15 11 30 5 12 27 40 3 8 1 3 20 1 2 8 21 4 7 1 4 11 30 8 1 5 8 1 12 © ABE and RRC . (a) (c) 1 4 3 10 2 7 4 9 1 8 2 10 36 81 (b) (d) (f) (h) (j) (b) (d) (b) (d) (b) (d) 1 8 1 3 1 6 1 3 3 8 12 20 72 88 2 4 2 5 1 6 2 1 8 1 3 10 8 3 33 8 27 8 39 4 Questions for Practice 5 1. 10. 5. 7. 11. 13. 12.36 Basic Numerical Concepts Questions for Practice 4 1. 9. 8. (a) (c) 4. 7 12 19 24 2. 4. (a) (c) 3. (a) (c) (e) (g) (i) 2. 3. 6. 14.

21 1 5 4 4 15 1 2 6 2 2 1  18 9 7 24 2.875 0. 2 2 2 9 7 24 16.75 0.375 0. 3 3 19 24 11 12 Questions for Practice 6 1. (a) (c) (e) (g) 3.47 23. 16 1 5 3 3 18 4 2 7 7 3 1  24 8 5 28 Questions for Practice 7 1.625 0.42 6.84 3.78 70 5.833 0.40 0. 3.092 5. 17.313 60 3  100 5 750 3  1000 4 © ABE and RRC .17 15. (a) (c) (e) (g) (i) (k) (m) (o) 2.178 9.276 1. 6.136 23. 18. (a) (c) 11. 4.68 26.413 5.571 125 1  1000 8 175 3 1 100 4 (b) (d) (f) (h) (j) (l) (n) (p) (b) (d) (f) (h) (b) (d) 19. 5. 7.6 50 0.591 10. 8.6 0.Basic Numerical Concepts 37 15.

33% 150% Questions for Practice 9 1.792  1015 0.653.700 H: £2.800 G: £2.000 Z: £25. 2.000000000816 7.453  108 468 1.000 Y: £20.0000602 0.15 20% 2. Firm A: 1 : 4 Firm B: 1 : 9 Firm C: 1 : 3 Firm D: 1 : 5 2. (a) (a) (c) 3. (a) (c) (e) 2.30 33.0935 0.000 (b) F: £1.032  104 3.38 Basic Numerical Concepts Questions for Practice 8 1.000 (b) (d) (f) (b) (d) 4.000. (a) X: £15.000 (f) © ABE and RRC .853. (a) (c) (e) 2. (a) (c) 3% 1 50 3 40 (b) (b) (d) (b) (d) (b) (d) 37. (a) (c) (e) 4.5% 2 25 9 50 £30 35 bottles £24.34  102 6.5  102 5.5% 90° £483.000.42  103 8.700 Questions for Practice 10 1.

39 Study Unit 2 Algebra. C. Equations and Formulae Contents Introduction A. Answers to Questions © ABE and RRC . Powers and Roots Brackets in Algebra Solving Equations Equality of Treatment to Both Sides of the Equation Transposition Equations with the Unknown Quantity on Both Sides Formulae Formulating a Problem Constants and Variables Substitution Rearranging Formulae Page 40 40 40 42 42 43 43 44 45 45 48 50 50 52 54 55 55 56 56 57 59 B. D. Some Introductory Definitions Algebra Equations Formulae Algebraic Notation Addition and Subtraction Multiplication Division Indices.

However. There are various rules and procedures which can be applied to equations to help this process. and is generally used to enable one or more unknown quantities to be worked out. and these will be examined in full. so that: 2a 3b We can then write: ab5 All that has happened is that we have replaced the numbers with letters. However. multiplication and division to algebraic notation. "a" could be 4 and "b" © ABE and RRC . There are formulae which describe a great many physical situations or mathematical problems and you will encounter these throughout your studies. usually expressed in the form of an equation. we look briefly at formulae. 5 is more than 4. the same basic rules are applied. but later in the course we shall consider more complex forms. It enables us to work out an unknown quantity or value provided we know certain specific quantities or values. in the previous expression. so it is important to understand these at the outset. A formula is one particular form of generalised mathematical statement. We all know that 2  3  5. that sets out a rule which may be applied in particular situations. algebra enables general expressions to be developed and general results to be obtained. and simplify algebraic expressions by the process of collecting like terms define equations and formulae and outline their uses find unknown quantities and values from simple equations by using transposition outline the principles of formulae and how they are constructed rearrange the terms in a formula to isolate different unknowns. Equations and Formulae INTRODUCTION In this study unit we continue our review of basic concepts by looking at algebra. Objectives When you have completed this study unit you will be able to:   outline the basic principles of algebra apply the basic arithmetic operations of addition. a number is a specific quantity – for example. subtraction.40 Algebra. though. Algebra essentially involves the substitution of letters for numbers in calculations. but less than 6 – whereas a letter can be used to represent any number. Thus. In this sense then. SOME INTRODUCTORY DEFINITIONS Algebra Algebra is a branch of mathematics in which. Here we shall be concerned with "simple" equations. both in this course and many others. Suppose. that we substitute letters for the first two numbers. instead of using numbers.     A. An equation is simply a statement of equality between two mathematical expressions. so that we can establish rules and procedures for carrying out mathematical operations which can be applied whatever the actual numbers involved. We shall then go on to introduce the subject of equations. we use letters to represent numbers. Finally.

Algebra. letters may be substituted to give us a general expression: std where: speed  s mph time  t hours distance travelled  d miles. Example 1: Suppose you have a piece of wood which is 7 metres long and from it you wish to cut a piece 4 metres long. Let us consider some examples to illustrate this further. Example 3: The distance travelled by a train in 3 hours at a speed of 60 miles per hour is easily calculated as: 3  30  180 miles. Here again. whereas arithmetic (using numbers) uses definite numbers and gives definite results. Arithmetic is specific. The main consequence of this is that algebra uses general expressions and gives general results. © ABE and RRC . this is a general expression which can be made specific by putting in particular values for "l" and "w". We only know that they are 2 and 3 respectively because we defined them as such before. we can reformulate the expression as: area  l  w Again. whereas algebra is general. calculated as follows: 743 This is a specific arithmetic statement relating to cutting a specific amount from this particular piece of wood. Equations and Formulae 41 could be 1. We could translate this into an algebraic expression by substituting letters for the specific lengths: let the original length of the piece of wood  x metres and the length of the piece cut off  y metres. Substituting the letters "l" and "w" to represent the actual length and width. Example 2: To find the area of a floor measuring 10 m long and 9 m wide. The calculation can now be shown as: xy3 This is now a general statement for cutting one length of wood from another to leave a piece 3 metres in length. The length of the remaining piece is 3 m. we multiply one dimension by the other: area  10  9 square metres  90 sq metres.

we can write this as: 3  the required number  6. x  4  20 x  4  20 x45 x31 The correct equations are: The point of constructing equations in this way is that it provides a means by which you can work out the unknown value (here. though – there are a number of simple rules which can be used to help solve them. equations can get very complicated. they are "2  2" and "4". using algebra – working with letters instead of numbers – allows us to construct general mathematical expressions. width. are. but is very important when we come to consider equations. a formula is an equation which always applies to a particular mathematical problem. Equations and Formulae Thus. However. It provides a set of rules which can be used in a particular situation in order to solve the problem. we have already introduced two formulae in the preceding section on algebra:   area  length  width distance travelled  speed  time. whatever the actual values. If a certain number is divided by 3 the result is 1. These examples are very simple and you can probably solve them (i. speed etc. The essence of these two statements is that they are always true. The two sides must always be in equality for the statement to be an equation. To save writing "the required number" (or the " unknown" number). Equations An equation is simply a mathematical statement that one expression is equal to another. Formulae A mathematical formula (plural "formulae") is a special type of equation which can be used for solving a particular problem. it is more convenient to call it "x".42 Algebra. for example. Here. In fact. If we now introduce the concept of algebra into this. work out the unknown) very easily. If 4 is taken from a certain number the result is 5. so that we can then write: 3x6 Here are a few simple statements which you can easily write as equations for practice: (a) (b) (c) (d) (a) (b) (c) (d) A certain number is added to 4 and the result is 20. whatever the actual values of length. "x"). the statement that "2  2  4" is an equation. For example. Note that an equation has two sides. This is not particularly helpful in itself. A certain number is multiplied by 4 and the result is 20. we have the makings of an extremely useful mathematical tool. Thus.e. © ABE and RRC . Don't worry. given that a certain number multiplied by 3 is 6. So.

we could put them into the equation. Addition and Subtraction Addition and subtraction follow the same rules as in arithmetic. subtraction. just as: 33333336 so y  y  y  y  y  y  y  6. using the full stop to represent multiplication). Having a formula for a particular problem set out as a general equation using algebra enables us to work out the answer by substituting actual values back into it for the unknowns. ALGEBRAIC NOTATION As algebraic letters simply represent numbers. can easily be used to solve the problem faced. we would write simply "ab" (or sometimes "a. © ABE and RRC . However. Equations and Formulae 43 If we know the actual values. in algebra it is not always necessary to write the multiplication sign. Again. we can formulate a general statement using algebra as follows:  Alw where: A  area l  length w  width. They are usually quite straightforward and. You will meet a good many in your studies – principally in this subject and in financial subjects.b".Algebra. but if not we would have to treat them as unknown. Thus. we can easily work out the distance travelled by substituting 60 for "s" and 2 for "t" in the above formula: d st  60  2  120 miles. by using the general rules governing equations. B.  dst where: speed  s mph time  t hours distance travelled  d miles. For example. all the terms to be added above were "y". Like terms are those of the same symbol or value – for example. or 6y You should note that this holds good only for like terms. So. if we know we have been driving for 2 hours at an average speed of 60 mph. the operations of addition. do not worry too much about them. There are many formulae used in mathematics and other physical sciences such as chemistry or physics. instead of "a  b". Thus. multiplication and division are still applicable in the same way.

44

Algebra, Equations and Formulae

When unlike terms are to be added, we can simplify an arithmetical statement, but not an algebraic statement. For example: 3  6  7  16 yxzyxz The three numbers, when added, can be reduced to a simpler form, i.e. 16. However, there is not a simpler means of expressing "x  y  3". We can, though, simplify expressions when they involve like terms. Just as the result of adding two numbers does not depend upon their order, so the result of addition in algebra is not affected by order. Thus: 27729 and x  y  3  3  y  x  y  3  x, etc. In simplifying an algebraic expression, we can collect like terms together whenever possible. For example: 5x  2y  x  3y  2  (5x  x)  (2y  3y)  2  4x  5y  2 We have collected together all the "x"s (5x  x) and all the "y"s (2y  3y), but cannot do anything with the "2" which is unlike the other terms. This distinction between like and unlike terms, and the way in which they can be treated, is an important point.

Multiplication
As we have seen, the product of a number and a letter, for example 3  a, may be written as 3a. Always place the number before the letter, thus 3a, not a3. Since multiplication by 1 does not alter the multiplicand, the expression 1a is not used; you just write a. To multiply unlike terms, we simply write them together with no sign. Thus: a  b  c  d  abcd If numerical factors are involved then they are multiplied together and placed in front of the simplified term for the letters: 3a  2b  4c  d  24abcd That this is correct can be shown by writing it in full and changing the order of multiplication: 3a2b4cd 324abcd  24abcd Multiplying like terms (for example, x  x) gives a result of the term being raised to the power of the multiplier (here, x2). We shall consider this in more detail later.

Algebra, Equations and Formulae

45

Division
Division should present little difficulty if we follow the rules used in arithmetic. Consequently, just as we write:
4 for 4  2 2

so we can write:
x for x  y y

Remember that any number goes into itself once. This will prevent you from saying that 6  6  0 instead of 6  6  1. In the same way x  x  1. It is then easy to cancel algebraic fractions. For example:
2xy 2y can be cancelled by dividing both the numerator and denominator by x to give 4x 4 2y can be further cancelled by dividing both the numerator and denominator by 2 to 4 y give 2 2xy or we could simply say we can cancel by dividing both parts by 2x. 4x

We can always cancel like terms in the numerator and denominator.

Indices, Powers and Roots
You will remember that indices are found when we multiply the same number by itself several times. The same rules apply in algebra, but we can investigate certain aspects further by using algebraic notation. To recap: xx xxx xxxx and so on. In, for example, x5, x is termed the base and 5 is called the exponent or index of the power. Note two particular points before we go on: x  x1 x0  1 (a) Multiplication of indices You will note that, for example: x2  x4  (x  x)  (x  x  x  x) xxxxxx  x6  x2  4 is written as x2 is written as x3 is written as x
4

(referred to as "x squared") (referred to as "x cubed") (referred to as "x to the power four" or "x raised to the fourth power")

46

Algebra, Equations and Formulae

and x3  x5  (x  x  x)  (x  x  x  x  x) xxxxxxxx  x8  x3  5 There is in fact a general rule that, when we have two expressions with the same base, multiplication is achieved by adding the indices. Thus: xm  xn  xm + n (b) Division of indices When we divide expressions with the same base, we find we can achieve this by subtracting the indices. For example:
xxxxxx x6  3 xxx x

(Rule 1)

xxx  x3  x6  3 and
x2 xx  x x

x  x1  x2  1 The general rule is:
xm  xm  n n x

(Rule 2)

(c)

Showing reciprocals In Rule 2 above, let us put m  0, so that the expression becomes :
x0  x0  n n x

We know that x0  1, so
1  x0  n  x  n n x

Thus, negatives denote reciprocals. (d) Raising one power by another To work out an expression such as (x2)3 – i.e. x squared all cubed – we have x2  x2  x2 as it is the cube of x2.

Algebra, Equations and Formulae

47

Thus (x2)3  x2  x2  x2 xxxxxx  x6  x2  3 In general, we have (xm)n  xmn (e) Roots Indices do not have to be whole numbers. They can be fractional. Let us consider x½ (i.e. x raised to the power half). We know from Rule (3) that: (x½)2  x½  2  x Thus, x½ when multiplied by itself gives x. However, this is exactly the property that defines the square root of x. Therefore, x½ is the square root of x which you will sometimes see denoted as:
2

(Rule 3)

x or, more commonly, as

x.

Therefore, x½ 
1

x . (The symbol "" means "identical with".)
1

Similarly, ( x 3 )3  x. Thus, x 3 is the cube root of x which is denoted by the symbol Similarly, x¼ is the fourth root of x denoted by Thus, in general, we can say that:
x
1 n

3

x.

4

x.

is the nth root of x or

n

x.

Note that in all these expressions the number to the left of the root sign – for example the 3 in
3

x – is written so that it is level with the top of the root sign. A number written

level with the foot of the root sign – for example, 3 x – would imply a multiple of the square root of x – in this example, 3 times the square root of x. We shall often need to use square roots in later work, particularly in statistics and in stock control. (f) Roots of powers There is one more type of index you may meet – an expression such as: or x 3 , or in general x n . To interpret these, remember that the numerator of a fractional index denotes a power and that the denominator denotes a root. Thus, x
3 2 5 m

is (x3)½, i.e. the square root of (x cubed) or

2

x 3 . Alternatively, it is (x½)3, i.e.

the cube of (the square root of x) or ( x )3. Let us consider its value when x  4:
4
3 2

2

43 

444 

64  8 (since 8  8  64)

or
4
3 2

 ( 4 ) 3  23  2  2  2  8

Therefore it does not matter in which order you perform the two operations, and we have:
x
m n

n

x m  ( n x )m

48

Algebra, Equations and Formulae

(g)

Collecting like terms We saw earlier how we can collect like items. Let us look at a couple of examples where indices are involved. It is no more complicated, even though it may look that way. a2  2a2  a2  a2  a2  3a2 These are like terms and can be added. b2  2b2  2b3  b2  b2  b2  b3  b3  3b2  2b3 Note that b2 and b3 are not like terms, so they cannot be combined.

Brackets in Algebra
The rules for brackets are exactly the same in algebra as in arithmetic. However, if we have unlike terms inside the brackets, it is not possible to collect them together before removing the brackets. In such cases, we need to be very careful when removing them and always obey the following three rules    If the bracket has a term (a number or a letter) in front of it, everything within the bracket must be multiplied by that term. If a bracket has a  sign before it, or before the term by which the terms inside the brackets are to be multiplied, the signs  or  within the brackets remain unchanged. If a bracket has a  sign before it, or before the term by which the terms inside the brackets are to be multiplied, the signs  or  within the brackets are changed so that  becomes  and vice versa.

These rules need a little explanation, so let us consider some examples. Example 1: 4x  2(x  y) The terms inside the brackets are unlike and cannot be further collected. All the terms within the brackets must be multiplied by the term outside before removing the brackets, so this gives: 4x  2x  2y  6x  2y Example 2: x  (y  3z) Again, the terms inside the brackets are unlike and cannot be further collected, so we can proceed to remove the brackets as the next step. Since there is a  sign before the brackets, we must reverse the signs of the terms inside when removing them. Thus: x  y  3z We can clarify this by imagining that there is an unwritten "1" before the bracket, making the expression effectively: x  1(y  3z) So "1" is multiplying the whole bracket: x  1  (y  3z)

Algebra, Equations and Formulae

49

This gives us three separate terms: x, 1  y, and 1  3z Thus, we have: x  y  3z Example 3: 3a(a  3)  4b(b  4) The first bracket is multiplied by 3a and the second by 4b. This gives us: 3a2  9a  4b2  16b Note that the  sign before the second bracket changes the  sign inside to . Often we find it necessary to insert brackets into an algebraic expression. This is known as factorisation because, when we include the brackets, we place any common factor outside the brackets. Thus: x  5y  5z  x  5(y  z) x  5y  5z  x  5(y  z) Notice how we must keep an eye on the signs as well.

Questions for Practice 1
Simplify the following expressions, writing "no simpler form" if you think any cannot be further simplified: 1. 3. 5. 7. 9. 11. 13 15. 17. aaaa 3x  2x  x 5x  3x  x 3a  2a x  y  xy a  2b  3a  4b 5 x + 2(y + z) x  (y  3z) a(b  c)  b(c  a) 2. 4. 6. 8. 10. 12 14. 16. 18. ab  ab 3x  2y  y 7xy  2x  y aaba3 xxxyyy 3x  (x  3) 4  3(2x  y)  (x y) 4x  2(x  y) 3a(a  3)  4b(b  4)

Now check your answers with those given at the end of the unit.

Looking at the rule in more detail. You can easily see that if 4 times the unknown  20 then the unknown must be 5. we can see a number of possible operations which may be carried out without changing the equality of the equation. in: 4x  20 the value of the unknown (x) can be discovered because of the stated relationship between 4 and 20. these are based on the simple rules which we set out here. (a) The same number may be added to both sides of the equation For example: x46 If we add 4 to each side. If the weights of a balance are equal on both sides it is "in balance".50 Algebra. and it will remain "in balance". we mean finding the value of the unknown or unknowns using the other numbers in the equation. So. or take an equal weight from each side. What you do to one side of the equation. we get: x  6  6  18  6 x  12 (c) Both sides of the equation may be multiplied by the same number. then we get: x4464 x  10 (b) The same number may be subtracted from both sides of the equation For example: x  6  18 If we deduct 6 from each side. you must also do to the other. There are a number of rules which can be used to help solve equations and we shall consider these next in relation to equations which have only one unknown. In a later study unit we shall examine more complex equations – those with two or more unknowns – and consider additional rules. Equality of Treatment to Both Sides of the Equation This is the golden rule that you must always follow when handling equations and there are no exceptions. anything we do to the equation must maintain this equality. For example: 1 3 of a number  10. SOLVING EQUATIONS When we talk about solving an equation. For example. or  10 3 3 © ABE and RRC . or 1 x x  10. However. Another way of thinking about this is that an equation is like a balance. Remember that an equation has two sides and those two sides must always be in equality for the whole statement to remain an equation. You can add an equal weight to each side. Equations and Formulae C.

4. © ABE and RRC . 3. we get: 2x 30  2 2 x  15 You should have noticed that the object of such operations is to isolate a single unknown. writing "no simpler form" if you think any cannot be further simplified: 1. 2. x  7  21 x  3  26 x + 9  21 x  11 5 x 7 8 3x  30 Now check your answers with those given at the end of the unit. x. 5. Questions for Practice 2 Simplify the following expressions.Algebra. or 2x  30 If we divide both sides by 2. For example: 2 times a number is 30. on one side of the equals sign. 6. Equations and Formulae 51 If we multiply both sides by 3. we get: x  3  10  3 3 3x  30 3 x  30 (d) Both sides of the equation may be divided by the same number.

Equations and Formulae Transposition When you understand the rules in the previous subsection you may shorten your working by using transposition instead.52 Algebra. Consider the following examples: © ABE and RRC . You must observe the following rules:  An added or subtracted term may be transposed from one side of an equation to the other if its sign is changed from  to . Consider the following examples: (a) x46 Transfer the 4 to the right-hand side of the equation and reverse its sign: x64 x  10 (b) x  6  18 Transfer the 6 to the right-hand side and reverse its sign: x  18  6 x  12 (c) 12  x  25 Transpose the 12 and reverse its sign: x  25  12 x  13 (d) 18  x  14 First transpose the x and reverse its sign: 18  14  x Then transpose the 14 and reverse its sign: 18  14  x x4 Note that we can keep transposing terms from one side to the other until we have isolated the unknown quantity. a divisor may be transposed from one side of an equation by changing it to the multiplier on the other. This is done so as to isolate an unknown quantity on one side.  A multiplier may be transposed from one side of an equation by changing it to the divisor on the other. Similarly. Transposition is a process of transferring a quantity from one side of an equation to another by changing its sign of operation. or from  to .

4. move 4 across the  sign and change it from divisor to multiplier: x  12  4 x  48 (b) 2x  20 To get x by itself. we need to transpose x and change it from divisor to multiplier: 24  6x Then. 6. transpose the 2 and change it from multiplier to divisor: x 20 2 x  10 (c) 24 6 x In this case. Equations and Formulae 53 (a) x  12 4 To get x by itself. 2. 5. Questions for Practice 3 Find the value of x by transposition: 1. we transpose the multiplier (6) and change it to a divisor: 24 x 6 x4 This rule is called cross multiplication. to get x by itself.Algebra. x  36 6 72  24 x 3x  3  21 5x  25 2x  3  15 21  x  15 Now check your answers with those given at the end of the unit. 3. © ABE and RRC .

54 Algebra. We simply keep transposing terms as necessary until we have collected all the unknown terms on one side of the equation. 6. x  3  8  2x 2x  24  x 1 2 x  15  x 5x  3  27  2x 18  x  5x  2 2x  24  6x 3 Now check your answers with those given at the end of the unit. For example. making it 6: 3x  x  8  6 Then transpose x from right to left. move the 2 to the right-hand side changing it from multiplier to divisor. consider 3x  6  x  8 First move the 6 to the right-hand side. 3. © ABE and RRC . 4. changing its sign: 3x  x  8  6 2x  14 Finally. Equations and Formulae Equations with the Unknown Quantity on Both Sides These equations are treated in the same way as the other equations we have met so far. 2. 5. and thus isolating x on the left: x 14 2 x7 Questions for Practice 4 Find the value of x: 1.

The amount of interest payable is determined by the annual rate of interest charged on the money borrowed. and then cycles at a certain number of kph for a certain number of hours. provided of course that we know all the others. The easiest way to explain this is to give an example. taking into account the time over which the loan is repaid.Algebra. whilst others describe key mathematical relationships which you will meet time and again throughout the course. What is his average speed? let: x  the speed at which he runs (in kph) y  the number of hours he runs p  the speed at which he cycles (in kph) q  the number of hours he cycles. we can say he covers (xy  pq) kilometres in (y  q) hours. Thus. then: average speed  xy + pq kph y+q © ABE and RRC . consider the following situation: A man runs at a certain number of kilometres per hour for a certain number of hours. Formulating a Problem When a person borrows money. it is possible to develop formulae to model any situation involving mathematical relationships. For example. he or she has to pay interest on the loan (called the "principal"). what we have done is express the real situation as a mathematical model using an algebraic equation. Some are just useful for helping to understand how to work out a particular problem or situation you are faced with. FORMULAE A formula is a mathematical model of a real situation. Equations and Formulae 55 D. multiplied by the number of time periods the interest is to apply. Therefore. It can be used for any simple interest calculation and enables us to find the value of any one of the elements. However. We could model this situation by using algebraic notation and then show the whole calculation as an equation: let:: I  simple interest P  principal r  rate of interest (%) n  number of periods then: IP r n 100 This equation is the formula for calculating simple interest. If we want to work out the amount of interest repayable on a particular loan. This particular formula is very important in financial studies and can be applied in many situations. the calculation is as follows: Interest  the principal multiplied by the percentage rate of interest.

For example. consider the following example: What is the simple interest on £1. The numbers in a formula always stay the same and are known as constants. sometimes. we have to give the variables the values which apply in that particular situation. The last step is to test the formula by going back to the problem and check that it works by using real number values in the formula through the process of substitution. see if you can work the following problem out by developing a formula.e. ensuring that all terms are expressed in the same units. Equations and Formulae Note that when we use a letter such as x to represent an unknown quantity. We shall look at substitution shortly. In order to use that model in a real situation. If she sells the mixture at v pence per kg. are as follows:    select a letter to represent each of the quantities required convert each block of information from the problem into an algebraic term or statement. the constant will always be 100. Turning back to the formula for simple interest. Constants and Variables As we have seen. for example. So. what profit. The steps to take in formulating the problem. numbers. to have said "let x  the speed at which he cycles" would have been inaccurate. Note that the unknown quantities have already been given letters. Activity As practice. we said "let x  the speed at which he cycles (in kph)". no matter what the values given to the P. it only represents that quantity and the units must be exactly specified. The letters on the other hand are variables. in the formula for simple interest. formulae consist of letters (algebraic notation) and. A grocer mixes m kg of tea at x pence per kg with n kg of tea at y pence per kg. per year) for two years? First of all we should restate the formula: IP r n 100 © ABE and RRC .000 invested at 10% per annum (i. These are combined together in particular ways to express the mathematical relationships between the terms such that a particular result is obtained. in pence.56 Algebra. The value that they have in working out the formula in any particular situation may vary – it depends on the particular circumstances of the problem. does she make? Bring your answer to £. then. r and n in a particular situation. using the letters selected form an equation by combining the symbolic terms. Substitution We defined a formula as a mathematical model of a real situation. Now check your answer with that given at the end of the unit. Doing this is the process of substitution.

we should note that a formula can be rearranged to find a different unknown quantity. We can do this by dividing both sides by P   Pr   n I  100    Pr   Pr       100   100  n I  Pr     100  Pr 100 I 100 Pr nI n 100 I Pr © ABE and RRC . but by using the rules we have discussed in relation to simple equations. Equations and Formulae 57 Now we substitute the values from the problem into the formula: P  £1. : 100 100 To find n.000  10  20 100  £200 This explains the principle of substitution. Rearranging Formulae Finally.000 r  10 n2 So: I  £1. So for example. The way in which the formula is expressed at present does not allow us to do that.000   10 2 100 £1.Algebra. as it may also be expressed. when considering simple interest. we start by restating the original formula: IP r n 100 Note that we can also express this as: I P rn 100 r Pr or. we need to rearrange the formula so that n is isolated on one side of the equation. we can rearrange the formula so that it does. In the next study unit you will get much more practice in this. Again. at a particular rate of interest. we may wish to know how long it would take to earn a specified amount of interest from a loan made to someone else.

2. © ABE and RRC . Questions for Practice 5 Rearrange the simple interest formula to allow the following calculations to be made: 1. Now check your answers with those given at the end of the unit. To find r – the rate of interest which would have to be charged to earn a particular amount of interest on a specified principal over a certain period of time. see if you can rearrange the simple interest formula to isolate the other two unknowns. However. To find P – the principal which would provide a certain amount of interest at a particular rate of interest over a specified period. Equations and Formulae Don't worry too much about the mechanics of this for now – again you will have plenty of practice at these types of operation as you work through the course.58 Algebra.

12 13 14. 3. 72  x. x  28 x  29 x  12 x  55 x  56 x  10 Questions for Practice 3 1. 5. 2. 7. 8. 3.x6 3 3x  21  3. 17. 4. 3. 11. 6. 10. Equations and Formulae 59 ANSWERS TO QUESTIONS FOR PRACTICE Questions for Practice 1 1. 16. 2. x  36  6  218 72  24x. 18. 4a 2ab 6x 3x  3y 3x No simpler form. 5. 2. 9.Algebra. 15. x  3 24 18 . x  © ABE and RRC . 6. 4. 3x  3y 4a  4b  5 3x  x  3  2x  3 x  2y  2z 4  6x  3y  x  y  4  5x  4y x  y  3z 4x  2x  2y  6x  2y ab  ac  bc  ab  2ab  ac  bc 3a2  9a  4b2  16b Questions for Practice 2 1. a 3a  b  3 No simpler form.

5. x  2x  x  24. 4. x  10 18  2  5x  x. or I  100 100 r . x  10 5x  2x  27  3. start with the original formula IP To find r P rn  n. we divide both sides by P  n: 100 © ABE and RRC . 6. 21  15  x. Equations and Formulae 4. or 100 100 To isolate P. x  8 1 2 5 2 1 3 3 x  x  15. We always start by restating the original formula: IP r P rn  n or I  100 100 r rn  n. 3. x  21  15  x.60 Algebra. we divide both sides by  rn  P  I  100    rn   rn       100   100  I P  rn     100  PI So 2. x 25 5 5 18 9 2 2x  15  3. 5. 6. x  2x  8  3. x  24  5  24     4½ 3 3 3 16 16 2 Questions for Practice 5 1. x  6 Questions for Practice 4 1. rn 100 I 100 rn 100 I rn P Once again. 2. x  5 24  6x  2x 1 1 3 72 9  5 x.

we simply divide the result by 100: profit  £ v (m + n)  (mx + ny) 100 © ABE and RRC . Equations and Formulae 61  r    Pn I  100   Pn Pn I r  Pn 100 r  r%. which is what we want to find 100 so r%  I Pn Activity The solution would be as follows: profit  selling price  cost price selling price  quantity sold  price  v(m  n) pence cost price  mx  ny pence profit  v(m  n)  (mx  ny) pence To convert this to £.Algebra.

Equations and Formulae © ABE and RRC .62 Algebra.

63 Study Unit 3 Basic Business Applications Contents Introduction A. Currency and Rates of Exchange Rates of Exchange Currency Conversion Discounts and Commission Discounts Commission Simple and Compound Interest Simple Interest Compound Interest Depreciation Straight-Line Depreciation Reducing Balance Depreciation Pay and Taxation Make Up of Pay Taxation Page 64 64 64 66 68 68 68 70 71 72 76 76 77 79 79 81 84 B. C. Answers to Questions for Practice © ABE and RRC . D. E.

Even then. the rate of exchange is usually shown as "x  y". and vice versa.98 (US dollars) £1  €1.9832 (US dollars) £1  €1.50  The rate of exchange is usually quoted in two forms – a selling price and a buying price. although it is a ratio. Before we look at the relevant calculations involved in converting one currency to another.7 Japanese yen Note that. the changes are very small. such large changes usually only occur over a long period or because of some particular economic problem. Companies may have many millions of pounds sterling invested in overseas currencies.00 Swiss francs or 1 Swiss franc  £0. (The difference between the two enables the bank or bureau de change etc.2845 (euros) © ABE and RRC .50). \$1  £0. or how much a particular amount of one currency is worth in another. and a small change in the value of those currencies can produce substantial losses or gains.00 Swiss francs £1  \$1. one side of the ratio is always "1". Here are some examples of typical rates of exchange of the British pound: £1  2.9830 (US dollars) €1.e. the rates quoted might be: Buying rate £1  2. It is this ratio that determines how much of one currency can be bought with another currency. Thus.43 (euros) £1  237. their effects can be very large when you consider that large amounts of money could be involved.Basic Business Applications 65 about 50p (i.0069 Swiss francs \$1. there are two further points to note about the rate of exchange. the rate of exchange can be expressed in two ways – how much one unit of one currency is worth in terms of the other. In general.  For any relationship between two different currencies.286 (euros) Selling rate 2. depending on whether the value of the currency rises or falls against the pound. undertaking the exchange to make a profit. The rate of exchange is expressed as a ratio.0076 Swiss francs £1  \$1. and the buying price determines the amount of pounds you will receive in return for your foreign currency. although the changes may be very slight. although they can take place quite rapidly. The selling price determines the amount of foreign currency you will be sold for your pounds. However. the rate of exchange between the British pound and the Swiss franc could be shown as: £1  2. Thus. It shows the value of one unit of currency in terms of another currency.) As an example of the two different prices. rather than "x : y".

02 units of the second currency (SF) is equivalent to 1 unit of the first (£). we would substitute the following values into the formula: x  100 e  1.29 (Note that the actual result of the calculation is £988. but we round down to express it correct to two decimal places as required for currency. to convert £100 into \$ at the rate of £1  \$1.) Example 2: Find the value of £63. as follows.29? The rate of exchange is expressed as 1. To convert a certain amount of the first currency (say.g. we would have to rearrange the above formula to isolate x.068  128. So. again x is the amount in pounds and y the amount in Swiss francs.275 euros if the rate of exchange is £1  1.66 Basic Business Applications Currency Conversion To find the value of a sum of money in one currency in terms of another currency. Thus the formula would be: yxe So. dollars) equivalent to one unit of a first currency (e. Substituting in the first formula (to find y) we get: y SF  63. we could develop a simple formula using the rate of exchange. Let: e  rate of exchange expressed as the number of units of a second currency (e. we simply divide each side of the equation by e – the rate of exchange: y xe  e e x y e We can now apply these formulae to any given problem. euros are the second currency and pounds are the first.98  £1) x  amount in first currency y  amount in second currency.07 Swiss francs (rounded up). as in \$1.275  £988. pounds.98.g.98 The formula gives us the following calculation: \$y  100  1.98  \$198 If we wanted to know how much a particular sum of the second (say \$y) currency would be worth in the first (say £x) currency. Substituting in the second formula (to find x) we get: £x  £ 1. £x) into the second currency (say \$y).40  2.37 1.29 units of one currency is equivalent to 1 unit of the other.37209. and x is the amount in pounds and y the amount in euros. © ABE and RRC . we multiply the amount of the first currency by the rate of exchange. Example 1: What is the cost in £ of an invoice for €1.40 in Swiss francs if the rate of exchange is £1  SF2.02. The rate of exchange is expressed as 2.02  128. To do this. So.

(a) (b) 3. The rate of exchange is expressed as 1. Calculate the cost of the meal in pounds.000  £1. Calculate: (a) (b) the number of Swiss francs they received the number of euros they bought.03 (rounded down). A bank buys Swedish kroner (SEK) at 10.35 per litre. © ABE and RRC .507. again x is the amount in pounds and y the amount in dollars. The procedure should be becoming clear to you now.01  £502.01 \$y  250  2. The Smiths spent a holiday touring in France. The meal for two cost £ (Cyprus) 35. How many Kroner did he receive? Sadly the excursion was cancelled and so he changed all his Kroner back into pounds.05 Swiss francs to the pound and 0. how much did the petrol cost them in pounds? what was the price per litre of the petrol in pence? 2. 195.Basic Business Applications 67 Example 3: Convert \$3. While in Switzerland they spent 720 Swiss francs and then changed the remaining francs into euros as they crossed the border into Italy.000 to pounds given that £1  \$1.05 kroner to the pound and sells at 9. 4.54 (rounded up).70. £x  £ Example 5: Find the value of £250 in dollars if the exchange rate is £1  \$2.3 3.642 euros to the Swiss franc.99. the Jones family changed £800 into Swiss francs. John and Paul ate a meal in a restaurant whilst on holiday in Cyprus.000 Japanese yen to pounds given that £1  195. The exchange rate was 2.7888 £ Cyprus to the British pound.3 yen.29 euros to the pound. Before going on holiday to Switzerland and Italy. The exchange rate was 0. While travelling they used 200 litres of petrol which cost €1. 1. So.99 units of the second currency (\$) is equivalent to 1 unit of the first (£). (a) (b) Mr Watson changed £100 into Swedish Kroner for a trip to Sweden.50 100. How much money did he lose because of the cancellation? Now check your answers with those given at the end of the unit.45 kroner to the pound.000  £512. The exchange rate was 1.99 Questions for Practice 1 1. Substituting in the second formula (to find x) we get: £x  £ Example 4: Convert 100.

600  £720  £2. She is paid commission of 12½%. Let us look in more detail at their exact meanings and the way in which they are calculated.600  20  £720 100 net price  £3.400  £6. Example 1: A three-piece suite was listed as £720. Pauline obtained £6. What is the net price? discount  £720  15  £108 100 net price  £720  £108  £612 Example 2: A car was listed at £3.800 Commission A commission is the amount of money paid to an agent. broker or salesman for buying or selling goods or services. The discount is expressed as a percentage of the list price (the list price is the original price) and what the buyer actually pays is the net price – the list price minus the discount. DISCOUNTS AND COMMISSION You've probably heard the terms "discount" and "commission".600 but was sold for cash less a discount of 20%.400   £800 100 200 © ABE and RRC . To find the commission.68 Basic Business Applications B. Thus: DB R 100 Consider the following examples. Calculate the amount due to her. To calculate a discount (D). commission  12½% of £6. What do they mean to you? Discounts and commissions are a practical use of percentage calculations. you use the following rule or formula: multiply the list price (or base. What was the final selling price? discount  £3. Consider the following example.5 25  £6. B) by the rate of the discount (R) and divide by 100.400  12. we simply multiply the value of the goods/services traded (known as the principal) by the rate of commission.400 of new business for the PR agency for which she works. usually offered as an inducement for quantity buying or for prompt payment. but was sold at a discount of 15%. This is expressed as a percentage of the value of the goods/services traded and is called the rate of commission. Discounts A discount is a reduction in the selling price of an item.

3. but consider the following example in relation to currency exchanges. We shall consider fixed and variable elements in more detail later. how many dollars would you get when converting: (a) (b) £100 to American dollars in cash? £200 to American dollars in cash and in traveller's cheques? In each case. © ABE and RRC .47. The fixed rate is the larger so that would be applied: £200  £3  £197 £197  1.47  \$142. the commission payable on converting £200 into \$ traveller's cheques is either £3 or £2 (1% of £200).12 However. The commission charged on traveller's cheques is commonly £3 or 1% whichever is larger. 2. What is the price paid for a £310 washing machine with a discount of 10%? What is the price paid for a £7.Basic Business Applications 69 Commissions are also usually charged by banks. and then apply whichever is the larger. travel agents and bureaux de change to cover their costs when carrying out a currency exchange. most people take a limited amount of cash in the currency of each country to be visited and take the remainder that they will need in the more secure form of traveller's cheques. In these cases. The commission on changing cash is commonly £3 or 2% whichever is larger.47  \$289. When going on holiday.47  \$288. we would need to work out the commission at both the fixed and the variable rate.000 car with a discount of 15%? An estate agent charges a commission of 1.59 Questions for Practice 2 1.5% of the value of each house she sells. Using the rates of commission quoted above. How much commission is earned by selling a house for £104.50 (b) The commission payable on converting £200 into dollars cash is either £3 or £4 (2% of £200). 4. The variable rate is the larger so that would be applied: £200  £4  £196 £196  1. The fixed rate is the larger so that would be applied: £100  £3  £97 £97  1. and given that £1  \$1.000? Calculate the commission earned by a shop assistant who sold goods to the value of £824 if the rate of commission is 3%. (a) The commission payable on converting £100 into dollars cash is either £3 or £2 (2% of £100). the person exchanging the currency or buying traveller's cheques is charged either:   a fixed cost – £3 per transaction or a variable cost – the value of 1% or 2% of the amount of the transaction.

in the bank for a number of years. SI  £270 and R  9. (Note that the way in which this formula is expressed is slightly different from that used in the previous unit.40 in an account that paid simple interest at a rate of 9%.) As we have seen. There are two ways of looking at interest – simple interest and compound interest. and then dividing by 100. with a rate of interest. R. that the notation is also slightly different. we can also rearrange this formula to allow us to work out any of the variables.000 to earn £270 if invested at 9%. So. we get: T 100  270 27.90 100 Example 2: Calculate the length of time taken for £2. Simple Interest Simple interest (SI) is calculated on the basis of having a principal amount. but is stated in a rather easier fashion.000   1. say P. Note. Here we need to work out T where P  £2.5 years 2. too. The principal (P) is £16.000. there is a simple formula to work out the amount of interest your money will earn: SI  PRT 100 This means that the interest is found by multiplying the principal by the rate.40  9  5  £5. the rate (R) is 9% and the time (T) is 5 years. substituting in the formula.40.000 © ABE and RRC .000  9 18. Example 1: John had £16. then by the time. It is often the case that formulae can be expressed in slightly different ways. As we have seen. It works in exactly the same way. Thus: P  100SI RT SI PT R%  T  100SI PR Consider the following examples. Calculate how much interest would be paid to John if he kept the money in the account for 5 years. Substituting in the formula. we get: SI  16.Basic Business Applications 71 loan/savings and also the period of the loan (or the amount of time you guarantee to leave the savings with them). T.

080. the amount invested at the end of the first year is £1.080 which is £86. usually annual) n  number of time periods. If an amount of money is invested and the interest is added to the investment. So you can multiply by 1.750 at 8.166. You can continue working out the amounts in this way.000 and the interest is 8% per year.240 in interest had been withdrawn. you are finding 100%  8%  108% of the amount at the beginning of the year. However. 2.72 Basic Business Applications Questions for Practice 3 1.08 each time. how much interest will you earn and how much will you have in your account after 4 years? If you invest £5. Now. to find the amount at the end of any year.50. © ABE and RRC . the interest is 8% of £1.40.75% interest rate? If you invest £6. At what annual rate of interest was the money invested? Now check your answers with those given at the end of the unit. Suppose you begin with £1.5% per annum. Consider the following examples.25% per annum for 6 months. After 4 years. 4. how much interest will you receive? £10. in the second year. 3. Adding this to the original principal. What is the simple interest on £250 over 3 years at 8. as follows: r   A  P 1    100  n where: A  accrued amount P  original principal r  rate of interest (for a particular time period. The interest in the first year is 8% of £1. The amount accrued at the end of the second year is £1.000 which is £80. There is a standard formula which can be used to calculate compound interest. a total of £3.800 was invested and at the end of each year the interest was withdrawn. Compound Interest Let us now look at how compound interest differs from simple interest.000 at 9. then the principal increases and so the interest earned gets bigger each year.

78 (to the nearest p).360  £403.763.215.51 interest at 5.000  £360  £3.58 value of the investment after 3 years  £3.000 interest at 12% (of £3.12  £3.000 is invested in an account paying 12% compound interest per year.51)  £116. Therefore the rate of interest for 6 months  5.Basic Business Applications 73 Example 1: £3.5%.105 interest at 5.51  £2.78 Alternatively.25%.20  £451.000)  £360 value of the investment after 1 year  £3.58  £4.31 value of the investment after 1½ years  £2. Example 2: £2.20  £3.20 value of investment after 2 years  £3.51 value of investment after one year  £2. Working this out year by year.5% per annum.214.25% (of £2.000  £105  £2. Calculate the value of the investment after this time and the amount of interest earned. we get: investment at the beginning of year 1  £2.360)  £403. Find the value of the investment after 3 years.000 is invested in a high interest account for 1½ years. The rate of interest for 1 year  10.000  1.404928  100  3  £4. and it is paid into the account every six months.105  £110.25% (of £2.82 © ABE and RRC . The interest rate is 10.000  1    £3.360 interest at 12% (of £3.331.51  £116.000 r  12% n3 Substituting these values in the formula gives: r   A  P 1   100   n 12   3  £3.000)  £105 value of the investment after 6 months  £2.763.20)  £451. we could use the formula: r   A  P 1    100  n where: P  £3. we get: investment at the beginning of year 1  £3. Working this out by each six month period.215.214.31  £2.25% (of £2.20 interest at 12% (of £3.000 interest at 5.215.763.105)  £110.000  1.

Consider the following example: A manufacturing company is considering replacing its production line in three years' time.83 So far we have used the compound interest formula r   A  P 1    100  n to work out how much a sum of money today (i.000 r  5. looking at our formula.g.e.74 Basic Business Applications Total compound interest earned  £2.0525  £2.82 Note that at each stage the amount of interest has been rounded to the nearest penny.e. the rate of interest (r) and the number of time periods (n) are known with certainty and the uncertain variable this time is the original principal (P).000.25% n3 Substituting these values in the formula gives: r   A  P 1    100  n  5. Calculate how much must be invested now in a bank account to purchase the new production line in three years' time.331. if the interest rate is 12% per annum.331.e. the rate of interest (r) and the number of time periods (n) are known and the uncertain variable is the accrued amount (A).82  £2. the replacement of an asset – such as machinery and equipment – where the future cost of the asset is known). © ABE and RRC . the accrued amount (A). the original principal P) would be worth at some point in the future (i.000  1.25  3  £2.000  1    £2.000  £331.e. In other words.000  £331.83  £2. a financial manager might want to know how much to invest now in a bank account in order to accrue a certain sum of money for a known future expenditure (e. the accrued amount (A)).1659134 100   3  £2. Alternatively. The cost of this is expected to total £40. there are some situations when we might want to know how much we would need to invest today (i. the original principal (P)) to achieve a certain level of known return at some point in the future (i. the accrued amount (A)). In other words. looking at our formula. However. we could use the formula: r   A  P 1    100  n where: P  £2.83 (to the nearest p). For example. (Interest earned each year will be reinvested and no money will be withdrawn). the original principal (P).000  1. Total compound interest earned  £2.331.

000 required to replace its production line in three year's time.471.471.000  P 1   becomes:  100  3 P  40.21 today at an interest rate of 12% per annum if it is to accrue the £40.000  P 1    100  3 To solve for P.000 r  12. You will recall from Study Unit 2 Section D how we go about rearranging formulae.0% n3 Substituting these values in the formula gives: 12   40.000 12   1    100  3   40. we need to rearrange our compound interest formula. © ABE and RRC .000 1. the manufacturing company needs to invest £28.21 (to the nearest p) Thus.404928  £28. Thus: r   A  P 1   becomes:  100  n P A r   1    100  n So 12   40.000 1.123 40.Basic Business Applications 75 Using our compound interest formula: r   A  P 1    100  n We know that: P  £40.

60 100 Therefore.500  £500 3 end of year 1: value  £1. How much will it be worth after 3 years if it depreciates at £600 per year.646. depreciation is calculated each year by applying a fixed percentage to the diminishing balance.280 Amount of depreciation in year 2  12% of £5.800  £600  8 years. the period over which the machine is depreciated to zero will be: £4.000  0.000  £6.280  0.000 If the annual depreciation is £600. but assets which quickly depreciate.60  £4.40  12  £4. such as motor vehicles.12  £633. the final balance never reaches zero).280  £633.280  12  £5.e. value of machine at the end of year 1  £6.646. and how many years will it be before the machine is valueless? Depreciation after 3 years will be 3  £600  £1.Basic Business Applications 77 Example 1: A computer costing £1. value of machine at the end of year 2  £5. the total provision for depreciation will never equal the cost price at the end of the asset's useful life (i.40  0.000  12  £6.800  £1. A company bought a machine for £6000.12  £557.800  £3. Show the process of depreciation.57 100 © ABE and RRC .000 end of year 2: value  £1.40  £4. such as buildings. will bear a large percentage (say 15%).646.000  £720  £5. assets having long lives. The percentage to be applied will depend on the type of asset. For example.500 will have a useful life of 3 years.500  £500  £1. If it depreciates at 12% per annum what will it be worth after 3 years? Amount of depreciation in year 1  12% of £6.800.12  £720 100 Therefore. Unlike the straight-line method then.800 The value of the machine will be £4.40 Amount of depreciation in year 3  12% of £4.000  £500  £500 end of year 3: value  £500  £500  £0 Example 2: A machine cost £4.646. Reducing Balance Depreciation In the reducing-balance method. annual depreciation  £1.280  £5. will have a small percentage (say 2½%). Consider the following example.

12 (12%). A launderette buys a washing machine which cost £2. Questions for Practice 5 1. We can demonstrate this in respect of the above example as follows: Amount of depreciation in year 1  12% of £6.88  £5.12 and then subtracting this from the original amount.088. How much will it be worth after 2 years if it depreciates at £400 per year? How many years before the machine is valueless? A factory buys a machine for £7.000  88%  £6.000. the value of the machine  £5.088.88 will give the amount left each year. Repeatedly multiplying by 0.000. Thus: At the end of the year 2. If you are reducing the amount by a factor of 0. A quicker and simpler way is to realise that 12% is 12/100.88  £4.88  £4. you are left with 0.646. the value of the machine  £4.88 (88%) of the amount.000  0.40  0.000  £6.57  £4. (a) (b) 2.83 We can express this method by the following formula: D  B(1  i)n where: D  depreciated value at the end of the nth time period B  original value at beginning of time period i  depreciation rate (as a proportion) n  number of time periods (normally years).12  £720 100 Therefore. value of machine at the end of year 1  £6. which is 0.646. If the machine will be valueless after 4 years.646.40 At the end of the year 3. The rate of depreciation is 20% per annum.280 This is the same as: value of machine at the end of year 1  £6.83 However this is obviously a time-consuming method. So.500.88 gives the same result as multiplying the amount by 0.40  £557.280 x 0.78 Basic Business Applications Therefore.000  0.000  £720  £5. 3. multiplying the amount by 0. value of machine at the end of year 3  £4.000  12  £6.12. What is its value after 3 years? © ABE and RRC . what should be the amount of depreciation per year? A small business buys a computer costing £4.280 This is both quicker and easier and you are less likely to make a mistake.

An insurance representative is paid purely on a commission basis. their basic salary or wage can be quite small. Calculate his total pay for the 4 weeks in which his sales were £3.700. Overtime worked at the weekend is paid at double time. He is paid overtime at time and half.500. a salesman earns a basic salary of £8.84  £210. If the value of insurance sold exceeds £4. Commission is paid at 8% on all insurance sold up to a value of £4. 40-hour week. What is the basic hourly rate? A trainee mechanic works a basic 37.100 and £2.04  £289.000.750 how much is she paid per month? A basic working week is 36 hours and the weekly wage is £306. £4. and some shop assistants.84  1. If she works overtime during the week.000. Now check your answers with those given at the end of the unit. she is paid at time and a half.000 per week. Calculate her wage for a week when she worked five hours overtime during the week and four hours overtime on Saturday. 4. Calculate the weekly wage of a factory worker if she works 42 hours per week at a rate of pay of £5.5)  £43. £5.08 People who are employed as salespersons or representatives.25 per hour If an employee's annual salary is £18.24  5 hours  (£5. How much does he earn in a week in which he does 9 hours overtime? An employee's basic wage is £6. 6. In some cases.000) worth of sales: commission  5% of (£9. 3. 2.900. and the commission on their sales forms the largest part of their pay.400  £5. 5.90.400  £5. are often paid a basic salary or wage plus a percentage of the value of the goods they have sold.000)  0.20 per hour. or even non-existent. he is paid a commission of 18% on the excess. and she works a basic 5-day.05  £4.280   £690 12 12 He earns commission on (£9.400  £220 Total monthly pay  basic pay  commission  £690  £220  £910 Questions for Practice 6 1.280 per year plus a monthly commission of 5% on all sales over £5.400? basic monthly pay  annual salary £8. What would be his income for a month in which he sold goods to the value of £9.84  2)  £35.5 hour week at an hourly rate of £4. For example.80 Basic Business Applications Basic pay: Overtime at time and a half: Overtime at double time: Total pay:  36 hours  £5. © ABE and RRC .80  3 hours  (£5.

The two main deductions are:   income tax and pension contributions.000 £36.1: Taxation bands Annual taxable income Tax rate 0 – £36. The tax is a percentage of gross pay.000 For a person with a total taxable pay of £45. the total tax liability would be: £20. the UK system divides taxable pay into two bands of income.Basic Business Applications 81 Taxation The total earnings made up in the way just discussed is known as gross pay.000)  40%  £7.600 £10. The amount of gross income on which tax is payable (known as taxable pay) may be affected by the provision of certain allowances within the tax system and also by certain elements of pay or deductions being exempted from tax (non-taxable items). based on the amount a person earns in a tax year that begins on 6th April and ends on 5th April the following year. This method of paying income tax is called PAYE (Pay As You Earn).800 These annual tax bands can be converted into monthly or weekly amounts for calculating tax on a PAYE basis and deducting the correct amount from an employees gross pay: © ABE and RRC . the total tax liability would be: £36.000  20% (£45.200 £3. Unfortunately. with the following rates of tax applied to each: Table 3. Most people have income tax deducted from their pay by their employer before they receive it. Certain amounts of money are deducted from gross pay and the pay the employee actually ends up with is the net pay or take-home pay. for a person with a total taxable pay of £20. The percentage rate at which the tax is applied varies according to the level of income. Income tax is a major source of finance for the government and is used to fund all types of public expenditure.001 and above 20% 40% This means that.000  £36. There are often other deductions – some being further national or local taxes on pay (such as the UK system of National Insurance) and others being deductions made by the employer for work-specific reasons (such as repayments of a loan for a car or for a public transport season ticket). very few employees actually receive all the money they have earned. The rates of tax applied to particular bands of income.000 in a year. are decided each year by the government as part of the Budget. the employer then pays the tax to the government.000 in a year.000  20%  £4. and the levels of allowances. From April 2008.

000 £3. using the bands and tax rates set out here.800? What is the monthly tax deduction? There are no non-taxable items specified. etc.800 tax liability  £12. an adjustment may be made to a person's personal allowances to allow for them. such as an employer's occupational pension scheme. © ABE and RRC .600 net taxable pay  gross taxable pay  personal allowances  £18.001 and above Monthly taxable income 0 – £3. whether they are single or married. i.82 Basic Business Applications Table 3. Here we shall consider just two. are not taxable.  Personal allowances – the amount of money an individual is allowed to earn each year before he or she starts to pay tax. If your personal allowances are greater than your actual pay.560 divided by 12.800  20%  £2. so: gross taxable pay  gross pay  non-taxable items  £18.000 of net taxable pay  20% plus remainder of net taxable pay  40% Consider the following examples. monthly and weekly taxable income Tax rate 20% 40% Annual taxable income 0 – £36. This amount (sometimes called free-pay) depends on the individual's circumstances – for example. but this can be converted into monthly or weekly amounts of "free-pay". The level of personal allowances are generally set as an annual amount. Therefore.001 and above Weekly taxable income 0 – £692 £693 and above There are many possible allowances and non-taxable items which affect the level of taxable pay.000 £36.  Pension contributions – the amounts paid by an individual into a pension scheme. (Note that this does not necessarily apply to any other types of deductions from pay. may also be deducted from gross pay before calculating tax.2: Annual. using the tax bands set out here.e. pension contributions.560 The monthly income tax deducted will be £2. number of children. Example 1: How much annual tax will be paid by someone who earns £18.800  £12.600  £5.) Now we can develop a framework for calculating the amount of tax payable annually. then you would pay no income tax at all. £213 rounded down. Where pension contributions are not paid through the employer. where they are deducted from pay by the employer. gross taxable pay  gross pay  non-taxable items net taxable pay  gross taxable pay  personal allowances tax liability  first £36.600 per annum and has personal allowances of £5.

Calculate her net pay in a week when she works 48 hours if the tax bands and rates are as set out above (i. 5.Basic Business Applications 83 Example 2: How much annual tax will be paid by someone who earns £43.000  £2.000 per annum. Always answer the question using the information supplied.000.000  20%  (£36. If tax is payable at the rate of 24%. 2.380 Note that taxation questions are often set without reference to income bands with different taxation rates applying. the pension contributions are non-taxable.400 and pays pension contributions through his employer at a rate of 5% of his annual salary. how much less tax will be paid by the employee in Question 1 above. For example: How much tax will be paid each month by someone who earns £24. Overtime is paid at a rate of time and a half for the first five hours.000 per annum if the tax rate is 25%? Net taxable pay per month. Here. He has personal allowances amounting to £4. based on the information supplied  Tax liability  £2.e. £24. She also belongs to an occupational pension scheme. or without reference to deductions from gross pay or personal allowances.850. 3.000  (£43. as per 2008 in the UK). An employee is paid a salary of £24. Now check your answers with those given at the end of the unit. Someone whose personal tax allowance is £4.000  25%  £500 per month. Would this person have to pay tax? An employee has a gross income of £23. the Chancellor changes the rate of income tax from 24% to 23%. She pays £3.800 is paid a weekly wage of £80. and at double time for any additional hours worked. © ABE and RRC . making contributions at a rate of 6% of basic earnings only. calculate his monthly pay. in the next Budget. she has a personal taxation allowance of £5. has personal allowances of £4.200  £180  £7.850 net taxable pay  £40.850  £4. so: gross taxable pay  £43.450 tax liability  £36.800.600 tax. If.450  £36.000 and personal allowances of £5. but wages remain the same. What is the rate of income tax? A weekly paid worker receives a basic wage of £5.50 per hour for a 37.000  5%)  £40.000 12 Questions for Practice 7 1.400  £36.5 hour week.492 per annum which he receives in 12 monthly payments.000)  40%  £7. As a single mother. 4.

2.64 euros.315.30 105p (rounded). £65.986 £2.26 (a) (b) (a) (b) (a) (b) £209.72 £304 There are three parts to this question.295 interest and £9.950 £1.68. 4. 1640 Swiss francs 590.05 HKD. they would be changing: £107 into NZD  299. £31 £5. £231. 3. 4. Spread between the four members of the family.97 Questions for Practice 2 1. 3.5% Questions for Practice 4 1. 2. 5.70 for the travellers cheques  £914.045 in the account. £7. Questions for Practice 3 1. 3. cost per person is £228.43 AUD £117 into HKD  2299. (b) (c) (i) (ii) £570 £5.39 NZD £97 into AUD  238. 2. 4. (a) They would be charged £3 for each cash transaction. 6.63 £2.70 Total cost is the £900 worth of foreign currency plus the commission of £9 for the cash transactions and £5. £45. Therefore.25 © ABE and RRC . 945 Swedish kroner £5.84 Basic Business Applications ANSWERS TO QUESTIONS Questions for Practice 1 1.25 7. 2.560 £24.70.

000 required to replace some of its machinery in three year's time.10 9. (a) (b) £1. Using the formula: P  A r   1    100  9. if the office is to accrue the £9.1659134 100   2  £2. Thus: value at the end of year 1  £4.200 5 years £1.762 would need to be invested today at an interest rate of 10% per annum. 3.173. © ABE and RRC .0875 100   1  £2.600 value at the end of year 2  £3.880 value at the end of year 3  £2.880  80%  £2.000  1.000 10   1    100  3 n   1.00 Therefore.5% payable six monthly equates to 4. To compare the effects of the two rates.331  £6. is the better investment.75% paid annually.75  A  £2. Questions for Practice 5 1. the rate of 8. 4.000 1.25% as a six-monthly rate.61 (to the nearest p) The calculation for the second building society rate of 8.75% paid annually is:  8.175. Thus.000  1    £2.750 The value at the end of each year is 80% of the value at the beginning of the year. with the rate of 8.600  80%  £2.500  80%  £3.000 3  9. 2.304. the second building society.000  1.0425  £2. £6. work out the accrued amount after one year in each case: For the first building society. using the formula for calculating the accrued amount over two six month periods is: r   A  P 1    100  n 4.25   2  £2.762 Therefore.000  1    £2.000  1.Basic Business Applications 85 3.

000 Company B: The value at the end of each year is 80% of the value at the beginning of the year.000 5 after 3 years. Questions for Practice 6 1.000  £61. Thus: value at the end of year 1  £305.00 Non-taxable items are pension contributions amounting to: © ABE and RRC .50 £308.5) per hour 5. 5.000. 2.160. total depreciation  £61.50 £1. 3.38 £206.50 £8.50  1.10 £1. The value at the end of each year is 89% of the value at the beginning of the year.41 per month No 20% Gross pay is made up as follows: 37. 3.000  £183.456 Questions for Practice 7 1.50  2) per hour gross pay £206.25 £41.247. 2.55 value at the end of year 2  £707.200 value at the end of year 3  £195.90 £344.000  £122. 5.86 Basic Business Applications 4.200  80%  £156. 6.562. £1.72  £630 to the nearest pound. Thus: value at the end of year 1  £795  89%  £707. £220.55  89%  £629.000  80%  £195.5 hours at (£5.25  6%  £12.000  80%  £244. 4.25 £60. 4. 5.000 value at the end of year 2  £244.50 £249.50 per hour 5 hours at (£5.000 value of the fleet after 3 years  £305.250  20  £305.16 per month £16.5 hours at the basic rate of £5. Total cost of the fleets to both companies  £15. Company A: annual amount of depreciation  £305.000  3  £183.

62  £5.16  £247.81  20%  £48.81  £240.46 © ABE and RRC .62 Net taxable pay  gross taxable pay  weekly personal allowances  £295.38  £295.38  £48.850 52  £295.00  £12.00  £12.81 Tax liability is based on the weekly income tax bands. and this level of weekly net taxable income falls within the 20% liability band (0 – £692).62  £54. tax liability  £240.16 Net pay  gross pay  pension contribution  tax  £308.Basic Business Applications 87 Gross taxable pay  gross pay  non-taxable items  £308.

Simultaneous Equations Solution by Elimination Formulating Problems as Simultaneous Equations Quadratic Equations Solving a Quadratic Equation by Factorisation Solving Quadratic Equations with no Terms in x Solving Quadratic Equations by Formula Logarithmic and Exponential Functions The Exponential Number e Logarithms Relationship between Logarithms and the Exponential Number Laws of Logs and Exponents Solving Equations using Logarithms and the Exponential Number Page 90 91 91 96 98 99 102 103 106 106 106 108 108 110 111 B. Answers to Questions for Practice © ABE and RRC . C. Exponential and Logarithmic Equations Contents Introduction A. D. Quadratic.89 Study Unit 4 Simultaneous.

so if you are unsure of any points. a multiplier may be transferred from one side of an equation by changing it to the divisor on the other. Quadratic. you must also do to the other. the same number may be added or subtracted from both sides of the equation without changing its equality. in order to collect all the unknown terms on the same side of the equation. and both sides may be multiplied or divided by the same number without changing its equality. Now we shall take our study of equations further by examining equations where there is more than one unknown quantity – simultaneous equations. In that study unit. Thus. We shall also consider the distinction between linear and quadratic equations. and learn how to solve equations using them.    Objectives When you have completed this study unit you will be able to:      use the elimination method to solve simultaneous equations including two unknowns formulate problems as simultaneous equations define a quadratic equation and distinguish it from a linear equation solve quadratic equations by the factorisation method simplify and solve equations using both base 10 and natural logarithms. using the same rules. go back to Study Unit 2 and revise the section on equations there. we considered simple equations where there was only one unknown quantity. Exponential and Logarithmic Equations INTRODUCTION In Study Unit 2 we introduced the concept of equations and looked at the basic principles of solving them – i. Quantities may be transposed – transferred from one side of the equation to the other – by changing their signs. Before we start though.e. Similarly. To do this. What you do to one side of the equation.   An equation is a statement of equality between two expressions. a difference which will be developed in the next study unit when we look at drawing graphs to represent equations. the constants. The golden rule in manipulating equations is equality of treatment to both sides of the equation. © ABE and RRC . Finally. an added or subtracted term may be transposed if its sign is changed from  to . we should recap the key principles already examined in relation to equations. It is essential that you fully understand these before studying this unit. Solving an equation means finding the value of the unknown quantity (or quantities) in the equation – the variable – by using the other numbers. we will examine the principles and rules of logarithms and exponential values. Unknowns may also be transposed. the expressions which form the equation may be manipulated with the objective of isolating the unknown on one side of the equation. and a divisor may be transferred from one side by changing it to the multiplier on the other. It has two sides and they must always be equal.90 Simultaneous. finding out the value of the unknown quantity. Thus. or from  to .

that for equations to be simultaneous they must satisfy two conditions – they must be consistent and independent. we had only one unknown quantity to find. Once we have done that. if x  y  7 and x  y  2. we can eliminate one of the unknowns and obtain an equation containing only one unknown. up until now. in practice. To determine a particular pair of unknown values then. we may find any number of pairs of values to satisfy it. Exponential and Logarithmic Equations 91 A. Similarly.3. SIMULTANEOUS EQUATIONS As we said in the introduction. Study them carefully. Note though. the resulting equation can be solved using the methods we already know. etc. Quadratic. Solution by Elimination If two equations are true simultaneously. etc. we need more information. and then the other unknown may be discovered by substitution in the original equation. Example 1: Consider the following pair of equations: 3x  y  11 x  y  3 (1) (2) We can add the two equations together to derive a third equation which will eliminate y: © ABE and RRC . there is only one solution: x  4½ and y  2½. One way of getting more information is to have another equation – for example: x  y  7 and 2x  y  12 Now there is only one solution pair: x  5 and y  2. These groups of equations are known as simultaneous equations. The following examples explain the method.3 and y  8. Consider the following example: xy7 This may be satisfied by: x  3 and y  4 x  2½ and y  4½ x  1. and no pair of values which satisfies one will satisfy the other the equations 3x  3y  21 and 2x  2y  14 are not independent – they are really the same equations. to find three unknowns we need three equations. many situations will arise when there is more than one unknown. If an equation involves two unknown quantities. and all pairs of values which satisfy the one will satisfy the other. By adding or subtracting suitable multiples of the given equations. Therefore to find two unknowns we need two equations. any other equation obtained from them will also be true. For example:   the equations x  y  7 and 2x  2y  5 are inconsistent – they cannot be true simultaneously.Simultaneous. However. The method of solution by elimination depends on this fact. when solving equations.

providing we apply the same proportionate change to each term in the equation (by multiplying or dividing each by the same number). independent equation. (The coefficient is simply the factor or number applied to an algebraic term.92 Simultaneous. If we now substitute that value for x in (1). Example 2: Consider the following pair of equations: 6x  5y  6 18x  7y  6 (1) (2) We want one of the unknowns to have equal coefficients. the coefficients were equal. but unlike in Example 1. but simply a different way of expressing the same equation. we get: 6x  6  15  9  x  1. y) was possible because both the "y"s in the original two equations had a coefficient of 1 – i.) If we want to eliminate one of the unknowns by adding or subtracting the two equations. However. We can check that the values we have determined for x and y are correct by substituting both in equation (2): x  y  2  5  3 Note that eliminating one of the unknowns (in this case. It is not a new.e. the equation has not been changed. we derive a changed version of the same equation by multiplying (1) by 3 and then subtract (2) from this equation: Multiplying (6x  5y  6) by 3 gives: 18x  15y  18 Note that. Quadratic. To make the coefficients equal. for example.5 © ABE and RRC . by adding the two equations. the coefficients of the unknowns must be equal. so that 3x becomes 6: 6  y  11  y  5 (the symbol  means "therefore"). x  2. we substitute the value of y  3 in (1) so that 5y becomes 15: 6x  (15)  6 By transposition. so the coefficient of the term 3x is 3. Exponential and Logarithmic Equations 3x  y  11 x  y  3 4x  8 Thus. Therefore. we can see that the coefficient of x in (2) is a multiple of the coefficient of x in (1). we are left with 4x  8. this is not the case here. We can now subtract equation (2) from this: 18x  15y  18 18x  7y  6 8y  24  y  3 To find x.

we get: 2y  16  18  2  y  1 Important note: When working through the processes of transposition and substitution. Quadratic. there are no equal or multiple coefficients. to avoid large numerical terms.. We can then eliminate y by adding the two equations: 26x  14y  66 63x  14y  112 89x  178  x  2 To find y. We can then choose which unknown to eliminate and. it is not necessary to write  (7) and  (15). Where there are no equal or multiple coefficients. so there is no obvious unknown term which can be easily eliminated. Example 3: Consider the following pair of equations: 13x  7y  33 9x  2y  16 (1) (2) This time. 5x  3y  14 7x  5y  38 3. but we will skip that step from now on. we shall decide to eliminate y. Questions for Practice 1 Solve the following simultaneous equations by finding values for x and y. we need to change the way in which both equations are expressed so that we can make the coefficients of one of the unknowns equal. 3x  y  8 x  5y  4 4.Simultaneous. we substitute the value of x  2 in (2) so that 9x becomes 18: 18  2y  16 By transposition. 1. so that the coefficient of y in both equations is 14. 2x  11y  23 7x  8y  50 © ABE and RRC . but be sure to pay special attention to all calculations involving negative numbers. To do this here. Exponential and Logarithmic Equations 93 It is always good practice to check the values by substitution in the other equation. etc. we multiply (1) by 2 and (2) by 7. 3x  4y  1 9x  5y  10 2.

x 1 2 2 3 y  16 x  2y  14 3 5 7. multiply (1) by 7 and (2) by 3 and then subtract: y 28 x 9 x 19 x   21  28 y 21  y (1) (2) 6  22   22 19 19 22  1 x x Substituting for 1 in (1) gives: x 4  22 3  4 y 19 88 3 4 y 19 © ABE and RRC . Quadratic. To eliminate 1 . 1 12 x  y  12 x 7 15 y  21 Now check your answers with those given at the end of the unit. Exponential and Logarithmic Equations 5. 9x  10y  12 2x  15y  5 6. Now we shall look at some harder examples! (a) Where the unknowns are denominators Consider the following equations: 4 3  4 x y 3 7 – 2 x y (1) (2) 1 1 and as the unknowns and then solve the problem in the usual x y Here we can treat way.94 Simultaneous.

we need to reformulate the equation as two simultaneous equations. but remember to write your answer as x  ? and y  ?. (b) Forming simultaneous equations from single equations Equations are sometimes given in the form: x  5y  2x  4y  3  3x  2y  4 In order to solve this. Quadratic. not as m and n. Taking the first pair of expressions. multiply (2) by 3 and add to (1): © ABE and RRC . To eliminate y. you may find it easier to put 4m  3n  4 3m  7n  2 (1) (2) Solve these in the usual way.Simultaneous. Exponential and Logarithmic Equations 95 88  76 12 3   y 19 19 4 1  y 19 y  19  43 4 4 1 1  m and  n. and rewrite the equations as: x y Alternatively. we can rearrange the terms to form one equation in the usual form: x  5y  2x  4y  3  x  9y  3 Now take the second pair of expressions and rearrange them: 2x  4y  3  3x  2y  4  2x  3y  4 We now have two simultaneous equations as follows: x  9y  3 2x  3y  4 5x  15 x3 y 2 3 (1) (2) These can be solved in the usual way.

Quadratic. we get: x5 The required number is 54. Let: x  the tens figure y  the units figure Then the number is 10x  y. Exponential and Logarithmic Equations 97 Example 2: In a bag containing black and white balls. we get: 10x  y  9  10y  x and 10x  y  6(x  y) for (1): 9x  9y  9 or xy1 for (2): 4x  5y 4x  4y  4 We can then substitute the value of 4x from (B) to get: 5y  4y  4 y4 Substituting for y in (A). and the reversed number is 10y  x. of two digits. and twice the total number of balls exceeds three times the number of black balls by four. Note how the equations are built up. (B) If we now multiply (A) by 4. we get: (A) (2) Rearranging these equations to isolate the constant in (1) and simplify both. Expressing the information given as equations. half the number of white is equal to a third the number of black. Find the number. A certain number. we get: (1) © ABE and RRC . is such that it is diminished by 9 if the digits are reversed.Simultaneous. How many balls did the bag contain? Let: x  number of white balls y  number of black balls. The number itself is six times the sum of its digits. Expressing the information given as equations. we get: 1 2 x 1 3 y and 2(x  y)  3y  4 Example 3: Pay special attention to the following problem.

we do not need to give it in full every time. since the product of two numbers is zero only when one of them is itself zero. Example 1: Consider the following equation: x(x  3)  0 Here. but one of them must be. we have to expand the product of the factors: x2  7x  10  4 © ABE and RRC . This is true whatever the value of the product. or (x  5)  0  x  0. y may have any value. then x  3 so the equation is satisfied if x  0. Again. Quadratic. x  0. or x may be ½. They need not both be 0. x may have any value. Example 2: Consider the following equation: x(x  5)  0 Either. reflecting this. we shall examine this later. as is shown in the following examples. if x and y are two quantities and all we are told about them is that their product is 12. etc.Simultaneous. in which case y must be 4. For example. First of all. the above method of solving an equation applies only when the RHS (righthand side) of the equation is zero. or if x  3. one of the factors must be zero: either: x0 or: (x  3)  0 further. Therefore. with the one exception that if xy  0. Example 3: Consider the following equation: (x  5)(x  2)  4 We must rearrange this so that the RHS is zero. in which case y must be 24. while the LHS can be expressed in factors. if (x  3)  0. we cannot tell either of their values – x may be 3. we cannot determine their values unless we are told something more. is that when we draw a graph of the outcome of a quadratic equation for all the possible values of the unknown. Exponential and Logarithmic Equations 99 A second feature. and if y  0. The following example simplifies the steps. Solving a Quadratic Equation by Factorisation If an equation contains two unknown numbers. the result is a curve. Therefore. the product of the two factors x and (x  3) is zero. or x  5 The important thing to remember is that this applies only to a product the value of which is zero. Having understood the principle of this procedure. then either x  0 or y  0. If x  0. This important fact provides the easiest method of solving a quadratic equation.

therefore. 5. or the terms are not already collected on one side of the equation.. remember to change the sign of the number as you transfer it to the RHS of the equation. or 2x  5  x  ¾. 6. For example: (4x  3)(2x  5)  0  4x  3. or x  1 Check this in the original equation just to justify to ourselves that the method works: (x  5)(x  2)  4 if x  6: (6  5)(6  2)  1  4  4 if x  1: (1  5)(1  2)  (4)(1)  4 Questions for Practice 4 Solve the following equations: 1. or (x  1)  0. However. Quadratic. have to work as follows: © ABE and RRC .. no matter how simple an equation is. On the other hand. equations are not so simple – the factors may not already have been found." step.. x(x  1)  0 (x  1)(x  2)  0 3x(x  1)  0 4x2  0 (2x  1)(x  3)  0 (3x  2)(2x  3)  0 Now check your answers with those given at the end of the unit. or .100 Simultaneous. you may prefer to omit the "either . but do not just guess. 3. or x  2½ One of the reasons that the questions just given were so easy is that the LHS of the equations were already expressed in factor terms. 2. Then. Exponential and Logarithmic Equations Then we rearrange the equation so that the RHS is zero: x2  7x  6  0 Next we have to re-express the LHS in factors: (x  6)(x  1)  0 Now we can say that either (x  6)  0. If you do this. though. You will. always copy it down first exactly as it is set. You should have found these questions very simple and you may even have been able to work them out in your head. you may give the answer at once.  x  6. if you are sure of the working. Commonly.. 4.

give your answers as alternative values for x: 5  x  1 7 or 5 Example 5: Solve x(5x  2)  3(2x  1) Expand the product of the factors: 5x2  2x  6x  3 Collect the terms: 5x2  8x  3  0 Factorise the LHS: (5x  3)(x  1)  0  5x  3. Exponential and Logarithmic Equations 101     Collect all the terms on the LHS. Example 4: Consider the following equation: 7x2  23x  60 First. Quadratic. or x  5 Finally. Equate each factor in turn to zero. collect all the terms on the LHS.Simultaneous. Finally. or x  1 x 3 5 or 1 © ABE and RRC . so that the RHS is zero: 7x2  23x  60  0 Next factorise the LHS expression: (7x  12)(x  5)  0 Then equate each factor in turn to zero: either (7x  12)  0. or (x  5)  0 which may be expressed as:  7x  12. where possible. Factorise the LHS expression – the product of the factors is then zero. give your answers as alternative values for x. so that the RHS is zero. Work through the following two examples carefully to make sure that you understand the process thoroughly.

5. or 5x  6  x  1. 8. Quadratic. There is. Consider the following equation: 25x2  36  0 Using the factor method. 4. 6. a simpler method of obtaining this answer and you should use it whenever there is no term in x in the given equation: 25x2  36  0 25x2  36 x2  36 25 © ABE and RRC . however. Solving Quadratic Equations with no Terms in x In particular circumstances. 9.2. this would be solved as follows: (5x  6)(5x  6)  0  5x  6. 7. 3. 7  x2  5  x x2  4x  4  0 x2  10x  24 x(x  2)  2(2  x) x2  11x  30  0 2x2  5  11x 10x2  11x  3  0 (x  3)(x  3)  8x (x  1)2  (x  1)2  34 (x  1)2  (x  1)2  x2 9x2  25 12x2  12x  9 Now check your answers with those given at the end of the unit. there is a quicker and easier method than the factor method of solving a quadratic equation. 2.102 Simultaneous. Exponential and Logarithmic Equations Questions for Practice 5 Solve the following equations: 1.2 or 1. 12. 10. 11.

4. 5. which means adding or subtracting the expression that is needed in order to be able to factorise the LHS. 2. Consider the equation ax2  bx  c  0. take the square root of each side. but opposite in sign: x2  36 25 x 6 or 1. Solving Quadratic Equations by Formula Since the method just used for solving a quadratic equation applies to all cases. Quadratic. 3. it can be used to establish a formula. b c b2  b  x  x     a a 4a 2  2a  2 2 © ABE and RRC . even where no real roots can be found. where a. Exponential and Logarithmic Equations 103 To solve this. In this case. it is:  b     2a  2 Remember that this needs to be added to both sides of the equation to keep it balanced. b and c may have any numerical value.Simultaneous. Rearranging the equation we get: ax2  bx  c  0 ax2  bx  c x2  b c x a a At this point. we need to "complete the square".2 5 Questions for Practice 6 Solve the following equations: 1. x2  16 4x2  9 3x2  27  0 x2  a2  0 9x2  1 Now check your answers with those given at the end of the unit. remembering that there are two square roots numerically equal.

When using it.4ac  x+    2a  4a 2  2 x b 2  4ac b  2a 4a 2 b b 2  4ac  2a 2a x x  b  b 2  4ac 2a This is an important formula. It can be used to solve any quadratic equation. c in the particular case.28 (to two decimal places) Example 2: Consider the following equation: 3x2  4x  5  0 x  b  b 2  4ac 2a where: a  3. although you should use the factor method whenever possible. Quadratic.236  10 10  1. paying great attention to the signs. c  11 Therefore: x  15  225  220 15  2.718  12. then write the values of a. Exponential and Logarithmic Equations Factorising the LHS we get: b  b 2 . c  5 Therefore: x   4  16  60  4  8. b. b  4.104 Simultaneous.718 or 6 6  0.718  6 6 4. Example 1: Consider the following equation: 5x2  15x  11  x  b  b 2  4ac 2a 0 where: a  5.79 or 2.12 (to two decimal places) © ABE and RRC . b  15.72 or 1. always write it down first.

Quadratic.Simultaneous. © ABE and RRC . if the term b2  4ac is negative. Exponential and Logarithmic Equations 105 Example 3: Consider the following equation: 3  2x2  4x First arrange equation with x2 term positive: 2x2  4x  3  0 Then apply the formula as usual. If b2  4ac is a perfect square. and that no real square root can be found for a negative number. no real roots of the equation can be found. Remember that the square of a number is always positive. Therefore. b  4. 2x2  13x  4  0 5x2  3x  4  0 x2  5x  2  0 4x2  7x  1  0 x2  x  1  0 Now check your answers with those given at the end of the unit. 5. For example: 2x2  4x  5  0 Applying the formula: x  b  b 2  4ac 2a where: a  2. then the square root will be a whole number and you should have solved the equation by the factor method. 4. 2. 3. c  5 Therefore: x   4  16  40 4  4   24 4 There are no real roots to this expression. Questions for Practice 7 Solve the following by means of the formula: 1.

Using common logarithms we can rewrite this expression as: log10 100  2 © ABE and RRC . To do so.7182818. simply locate and press the ex button on your calculator and enter 1. so you may have to experiment with your own calculator to find out how to obtain the value of e. e0 e3 e-2 e1/2 e-3 Now check your answers with those given at the end of the unit. it is known as the exponential constant or exponential number. (Calculators differ. It is common to work with logarithms using the number 10 as the base. LOGARITHMIC AND EXPONENTIAL FUNCTIONS So far in this study unit we have considered simultaneous and quadratic equations.e. 5. The Exponential Number e There is a particular constant used in many mathematical calculations whose value (to seven decimal places) is 2.106 Simultaneous. you should have no problem handling logarithms! In essence. In other words. you have just demonstrated that: e1  2. Let us first consider the expression 102. Logarithms A logarithm (also known simply as a log) is used to write an expression involving powers. 102  100 (i. 10 is the base and 2 is the power or index. Let us work through some examples to aid our understanding of the application of logarithms. Have a go on your scientific calculator now to obtain the value of e. Having mastered the application of powers in Study Unit 2.7182818 Questions for Practice 8 Using your scientific calculator. find the following values: 1.) By performing this calculation. 4. a logarithm is the power to which a fixed number (base) must be raised in order to produce a given number. 3. 2. In this section we will now move on to consider logarithmic and exponential functions. Some scientific calculators may require you to enter 1 before pressing the ex button. referred to as common logarithms. Exponential and Logarithmic Equations D. Quadratic. From our understanding of powers. This value occurs so frequently when mathematics is used to model physical and economic phenomena that it is convenient to write it simply as e. 10  10  100).

76) ln(0. ln(1. Quadratic.e.3) ln(2002) ln(10) ln(38) Now check your answers with those given at the end of the unit. using this value for x. do not forget that: © ABE and RRC . think of a number for y and then calculate the natural logarithm of this number to find x (i. thus demonstrating that an inverse relationship exists between the natural logarithm function f(x)  ex and the exponential function f(y)  ln(x). Since logarithms are basically powers.e. Notice that the variables x and y have been reversed. Relationship between Logarithms and the Exponential Number From this discussion of natural logarithms and the exponential number. 4. so if y  ex. The resulting value of y should be the same as the initial number you chose for y. there should be some logic to the following rules. This is not a mistake! When we take the natural logarithm of an exponential function the result is the exponent. 5. find the natural logarithms of the following values: 1. First. 3. Thus. Thus. Use your calculator to confirm this natural logarithmic and exponential relationship. we can conclude that: ln(ex)  x Laws of Logs and Exponents Numbers expressed in logarithmic or exponential form obey the same rules applicable to powers and indices. for any base:       log(p  q)  log p  log q p log    log p  log q  q log pn  n log p if y  axn then n  (log y  log a) ÷ log x log 1  0 ln(ex)  x and finally. calculate the value of y using the exponential function y  ex. Here we give the various rules of logarithms (which can be equally applied to common logarithms (i. which have already been described in Study Unit 2. Then. to the base 10) and natural logarithms (to the base e)). 2. using the formula y  ex). you may have identified that a relationship exists between the two. The natural logarithm function f(x)  ex is the inverse of the exponential function f(y)  ln(x).108 Simultaneous. Exponential and Logarithmic Equations Questions for Practice 10 Using your scientific calculator. then ln(y)  x.

© ABE and RRC . Example 1: Simplify the following logarithm expression into a single log term: log(x)  2 log(y) Using the rule [log pn  n log p] we can rearrange the second part of the expression: log(x)  log(y2) Using the rule [log(p  q)  log p  log q] we can rearrange the expression: log(xy2) Example 2: Simplify the following logarithm expression into a single log term: 3ln(A)  4ln(B) Using the rule [log pn  n log p] we can rearrange both parts of the expression: ln(A3) – ln(B4) p Using the rule [log    log p  log q] we can rearrange the expression:  q ln  A4    B  3 Questions for Practice 11 Simplify the following logarithmic expressions to a single log term: 1. 2.Simultaneous. 3. Quadratic. we can simplify some expressions involving logarithms to a single log term. Exponential and Logarithmic Equations 109 Using these laws. log(x)  log(x  4) 4 log(2)  2 log(3)  log(12) ln(x2y)  ln(y)  3 ln(x) Now check your answers with those given at the end of the unit.

3. 4x  100 e2x  10 log 3x  3 Now check your answers with those given at the end of the unit. by answering the following questions. Quadratic. Questions for Practice 12 Solve the following equations. © ABE and RRC . 2. let us bring together everything we have learnt about them.110 Simultaneous. giving your answer correct to 2 decimal places (you will need a scientific calculator to evaluate logs and natural logs): 1. as well as how to solve equations in general. Exponential and Logarithmic Equations Solving Equations using Logarithms and the Exponential Number Before completing this section on logarithms and exponents.

multiply (i) by 5 and add the equations: 16x  44.  x  2 3 4 or to eliminate x. multiply (i) by 5 and (ii) by 3 so that the coefficients are equal and then add: 25x  15y  70 21x  15y  114 46x  184  x  4 Substituting x  4 in (i) we get: 20  3y  14 3y  6 y2 (Check in (ii): 28  10  38) The full workings for the next three answers are not given.Simultaneous. y  1 4 © ABE and RRC . multiply (i) by 3 so that the coefficients are equal and then subtract (ii): 9x  12y  3 9x  5y  10 7y  7  y  1 Substituting y  1 in (i) we get: 3x  4  1 3x  5 2  x  13 (Check in (ii): 15  5  10) 2. multiply (ii) by 3 and subtract (i): 6y  4. Quadratic. 3. 5x  3y  14 7x  5y  38 (i) (ii) To eliminate y. 3x  y  8 x  5y  4 (i) (ii) To eliminate y. but the method is indicated. Exponential and Logarithmic Equations 111 ANSWERS TO QUESTIONS FOR PRACTICE Questions for Practice 1 1. 3x  4y  1 9x  5y  10 (i) (ii) To eliminate x.

2x  11y  23 7x  8y  50 or (i) (ii) To eliminate y.112 Simultaneous. 1 1 2 x  3 y  12 5 (i) (ii) x  15 y  21 1 2 1 7 Notice that (i) has a factor 3. so divide by this to get a simpler equation: x5y4 (i) 13 15 7 2 To eliminate x. x  3 y  16 1 2 2 (i) (ii) To eliminate y. y  13. Quadratic. subtract 7 times (i) from 2 times (ii): y  1 5. subtract 2 times (ii) from (i): 3 3 y  12. subtract 2 times (i) from 9 times (ii): y   3 5 6. 9x  10y  12 2x  15y  5 or to eliminate x. 10y  36. subtract 2 times (i) from (ii): ( 15  5 )y  13. y  3 10 substituting for y in (ii): 1 2 1 2 1 2 6 x  6 5  14 x  14  7 5 4 x 65 1 x  13 3 5 7. y  15 Substituting in (i) for y gives: 1 2 1 2 x34 x7 x  14 © ABE and RRC . subtract 2 times (ii) from 3 times (i): x  2 (i) (ii) x  2y  14 6 1 To eliminate x. subtract 8 times (i) from 11 times (ii): x  6 to eliminate x. Exponential and Logarithmic Equations 4.

and y  2 Questions for Practice 3 1. if three times their difference is 18: Substituting for y in (i). 2. and b  . Then 6x  11y  1. change each amount of money to pence. Let: x  cost of tea per kg in pence y  cost of sugar per kg in pence. As the units must all be the same.Simultaneous. we get: x  17  40 x  23 Therefore. Exponential and Logarithmic Equations 113 Questions for Practice 2 Let a  1 1 . and b  Therefore: x  1. Then express the equations using these terms: x y 3a  20b  13 7a  10b  2 Solving these in the normal way gives: a  1. the numbers are 17 and 23.330 11x  6y  1. Let: x  first number y  second number Then. we get: (i) and.730 (i) (ii) © ABE and RRC . Quadratic. if half the sum of two numbers is 20: 1 2 1 2 (x  y)  20 or x  y  40 3(x  y)  18 or xy6 2y y  34  17 (ii) Eliminating x by subtraction.

250 y  50 Substituting for y in (i) we get: 6x  550  1.780 20x  20y  13. 4. 3. subtract 6 times (ii) from 11 times (i) to get: 85y  4. Thus.50 (i) (x  9 8 1 8 x)  (y  5 6 1 6 1 y)  657 2 x 1 y  657 2 Multiplying this by 24 to eliminate the fractions. the original income from: advertisements  £340 sales  £330.380 x  340 Substituting for x in (i): 340  y  670 y  330 Thus. Let: x  the original income from advertisements (in £) y  the original income from sales (in £).50 per kilo.330 x 1. in the first year: x  y  670 In the second year: 1 2 (x  12 2 % x)  (y  16 3 % y)  670  12.114 Simultaneous. the cost of tea  £1. we get: 27x  20y  15. Exponential and Logarithmic Equations To eliminate x. multiply (i) by 20: © ABE and RRC .400 and subtract this from (ii) to get: 7x  2. and the cost of sugar  £0. Let: x  the average speed (in miles per hour) y  the distance (in miles) We can use a formula for the time taken: distance  time taken speed (ii) To eliminate y from the pair of equations. Quadratic.30 per kilo.330  550 6 x  130 Therefore.

4. 6. 5. x  0 or 1 x  1 or 2 x  0 or 1 x0 x x 1 2 2 3 or 3 1 or 1 2 Questions for Practice 5 1. 6. 3. x  2 or 1 x  2 x  12 or 2 x  2 or 2 x  5 or 6 x 1 2 or 5 1 x   3 or  2 5 © ABE and RRC . Quadratic. 3. 4. 2. we get: y  110 Thus: average speed for the journey  55 mph distance from London to Birmingham  110 miles. Exponential and Logarithmic Equations 115 So at average speed: y  2 hours x and at the slower speed: y 1  2 hours 12 minutes  2 5 hours (x  5) Rearranging these equations into standard linear form. Questions for Practice 4 1. we get: y y  2x (i) 1  2 5 x  11 (ii) Subtracting (ii) from (i) gives: 0 1 5 x  11 1 5 11  x 55  x  x  55 Substituting for x in (i). 2. 5.Simultaneous. 7.

2. x  9 or 1 x  4 or 4 (written as x  4) x  0 or 4 2 x  13 1 x  1 2 or 1 2 Questions for Practice 6 1. 9.16.32.59 1.301 log 10 0. 6.62 Questions for Practice 8 1.649 e-3  0.4  -0.643 © ABE and RRC .135 e1/2  1.050 Questions for Practice 9 1. x  4 1 x  1 2 x  3 x  a 1 x3 Questions for Practice 7 1. 11. 5. Quadratic. 10.699 log 10 45  1. 5. 4.24.398 log 10 4. 4.4  0. 3. 3. 1.653 log 10 2  0. Exponential and Logarithmic Equations 8. 3. 0. 12. 0. log 10 50  1.18 1. 0. 3.44 0.086 e-2  0. 2. 2.56. 0. 2.64 4.116 Simultaneous. 4. 5.62. 5. 4. e0  1 e3  20.

5.3)  1. Using the rule [log(p  q)  log p  log q]: log(x)  log(x  4)  log(x (x  4))  log(x2  4x) 2. Exponential and Logarithmic Equations 117 Questions for Practice 10 1.  2x  In(10)  x  In10  2.602 ln(10)  2.638 Questions for Practice 11 1. 4.Simultaneous. Using the rule [log pn  n log p]: 4 log(2)  log(24)  log(16).204 ln(2002)  7. 2. Using the rules [log(p  q)  log p  log q] and [log pn  n log p]: ln(x2y)  ln(y)  3ln(x)  ln(x2)  ln(y)  ln(y)  ln(x3)  ln(x5) Questions for Practice 12 1. 3.  x  3.  3x  10 3 .32 e2x  10. 2. Quadratic. 3.  x  1 15 log 3x  3.303 ln(38)  3. and 2 log(3)  log(32)  log(9)  p Using the rules [log(p  q)  log p  log q] and [log    log p  log q]:  q 4 log(2)  2 log(3)  log(12)  log(16)  log(9)  log(12)  16  9   log    12   log(12) 3. 4x  10.5653 ln(0.  x  log 100  log 4 .76)  0.33 © ABE and RRC . ln(1.  x  10 3 ÷3  x  333.

D. Answers to Questions for Practice © ABE and RRC . C. Basic Principles of Graphs Framework of a Graph Drawing a Graph Graphing the Functions of a Variable Representation Ordered Pairs Graphs of Equations Linear Equations Quadratic Equations Equations with x as the Denominator Exponential and Logarithmic Equations Gradients Gradient of a Straight Line Equation of a Straight Line and Gradient Using Graphs in Business Problems Business Costs Break-even Analysis Economic Ordering Quantities Plotting Time Series Data Page 120 121 121 122 124 125 126 126 127 132 136 139 141 142 145 147 147 151 152 155 160 B. E.119 Study Unit 5 Introduction to Graphs Contents Introduction A.

One element of this is the need to use graph paper. The picture may. For example. You will immediately realise that since equations are a mathematical expression of the relationship between two variables. and we cover some of the key design aspects here. since the answers must be read from the graph and hence cannot be more accurate than the thickness of the pencil. it is possible to represent equations by means of graphs.120 Introduction to Graphs INTRODUCTION For many problems and issues which fall within the remit of quantitative methods. There are many different types of graph. quadratic. The graph might also be used to approximate a solution to a particular problem. The importance of graphs is that they give a visual picture of the behaviour of the two variables. whether two lines are parallel describe a shaded region on a graph in terms of the lines bounding it use graphs to solve a variety of problems. many graphs involve time as one of the variables. showing graphically what happens to one variable when the other changes. which way its graph will slope identify. We shall spend some time on this in the unit. from looking at their equations. and thus provide a measure of something that changes with time. quadratic (curved line). Line graphs illustrate the relationships between two variables. You will find it very useful to have a supply to hand as you work through this unit. from an equation. indicate a trend in the data from which we might be able to make further predictions. However. f(x)) calculate and tabulate the values of y from an equation in x draw graphs from linear. For this reason. From this. it is possible to analyse and solve them both mathematically and graphically. it is essential that you are very careful when drawing graphs. for example. it is possible to draw a graph. exponential and logarithmic graphs. but here we will only be concerned with linear (straight line). by knowing how to interpret it. and calculate its gradient from its graph or its equation decide. information can very quickly be found. Objectives When you have completed this study unit you will be able to:          explain the construction of graphs calculate functions of x (that is. This study unit introduces the principles and role of graphs in such problems. they are less accurate. Whenever one measurable quantity changes as a result of another quantity changing. Graphical approaches to problem solving are often easier to carry out than mathematical ones. © ABE and RRC . exponential and logarithmic equations identify the characteristics of a straight line in terms of its equation.

We have shown the main gridlines here as grey. This grid may actually be marked on the paper (as is the case with graph paper. We shall start by considering the framework of the graph. Not all values need to be marked. dotted lines. The first co-ordinate number always refers to a value on the scale along the x-axis and the second to the scale on the y- © ABE and RRC . the variable which causes change in the other variable. The area within the axes (known as the plot area) is divided into a grid by reference to the scale on the axes. The x-axis is the horizontal axis and usually displays the independent variable – i. but the dividing marks do need to be equally spaced along the axis and the scale used must be clear to the reader.e. 0–9 on the x-axis and 0–6 on the y-axis). BASIC PRINCIPLES OF GRAPHS A graph is a diagram illustrating one or more mathematical relationships between two variables. Each axis should be titled to show clearly both what it represents and the units of measurement.2) x-axis Time since planting (days) A graph has two axes along which the values of the variables are shown:  The y-axis is the vertical axis and usually displays the dependent variable – i. Here this is the height of the plant (which changes in accordance with the time since planting). Framework of a Graph Figure 5. The scale need not be the same on both axes.3) co-ordinate (6.1: Sample graph framework Height of plant (cm) 6 5 4 3 y-axis 2 1 origin 0 1 2 3 4 5 6 7 8 9   co-ordinate (3. Particular points within the graph are identified by a pair of numbers called co-ordinates which fix a point by reference to the scales along the axes.Introduction to Graphs 121 A. The following figures show the relationship between the height of a plant and the time since it was planted.e. which is why it is much easier to draw graphs using graph paper) or may be invisible. We shall use this to explain some of the basic features of graphs and their construction. Here this is the time since planting (which causes the height of the plant to increase). the variable which changes according to changes in the other variable.  The values of the variables are marked along each axis according to a scale (here.

In this case. you may be given the following data about the growth of a plant: © ABE and RRC . note that the x and y-axes may extend to include negative numbers. Remember that the scales must be consistent along each axis. If you are not using graph paper. since the range of values for the independent variable will determine the range of values needed for the dependent variable on the yaxis. the framework of the graph will look like that shown in Figure 5. for the dependent variable. If you are using graph paper. Points on the graph are identified by their co-ordinates.1) x-axis Drawing a Graph When drawing a graph. Finally. The point at which the two axes meet is called the origin. Figure 5. You should start by considering the x-axis.122 Introduction to Graphs axis.2)  2 1 y-axis 3 2 1 0 1  co-ordinate (1. We have shown two points on the framework in Figure 5. use the thick lines as your main divisions. Draw the axes and mark off the scales along them. use a ruler to mark the scale and select easy-to-identify measurements for the divisions (such as one cm for each main unit). The aim is usually to draw as large a graph as possible since this will make it easier to plot the co-ordinates accurately. secondly. This has the co-ordinate reference (0.1 – make sure you understand their referencing. but do not need to be the same on both. Give each axis a title.2: Graph framework including negative scales 3 co-ordinate (2.2) 3 2 1 2 3  co-ordinate (3. number the main divisions along each scale and give the whole graph a title. and also make it easier to read.  Determine the range of values which will need to be shown on each of the axes.2. Thus. follow this procedure for constructing the framework.0). the independent variable and. These are the values given (or determined) for firstly.   Now you are ready to start plotting the graph itself.

you should join them all up. When you are satisfied that the path of the curve is smooth.1) you count one unit across (along the x-axis) and then count one unit up (along the y-axis).3: Graph of height of plant against time 6 Height of plant (cm) 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 Time since planting (days) Now that we have reviewed the basic principles.7). Figure 5. then the plot line will form a curve. give the complete graph for all values of the independent variable. if your graph is a quadratic graph. Be careful not to get these in the wrong order or when you come to plot the co-ordinate (7. which is somewhere completely different! (Remember that if the x co-ordinate is negative you count to the left and if the y co-ordinate is negative you count downwards. If your graph is a linear graph.3.) Once you have plotted all the given points. draw the curve through the points in one movement. © ABE and RRC . to find the shape of the curve. The easiest way of doing this is to trace through the points with a pencil.) However. all the points will be in a straight line and you can join them using a ruler. you only need to plot two points which. when joined and extended across the whole scale of the x-axis. The full graph derived from the previous data on the height of a plant and extended over nine days is given in Figure 5. The points should be joined with a smooth curve.Introduction to Graphs 123 Height of plant (cm) Day 1 Day 3 Day 5 Day 7 1 2 3 4 The points should be plotted with a small x using a sharp pencil. to draw a linear graph. we shall move on to consider in detail the way in which variables and equations are represented on graphs. (In fact. To plot the co-ordinate (1. but without touching the paper.4) you will have plotted the point (4.

Suppose you are given the equation y  x2  3 and are asked to find the values of y from x  2 to x  4. For example. if y  x  3. GRAPHING THE FUNCTIONS OF A VARIABLE The statement that something is a "function of x" simply means that the value of that something depends upon the value of a variable quantity. you could tabulate just the first and the last lines. for example: f(x)  x2  2x Replace x by 3 and find the value: f(3)  32  2  3  15 Replace x by 2 and find the value f(2)  (2)2  2(2)  0 © ABE and RRC . as this is a simple problem. working the rest in your head as you go along. y  4. y  6.124 Introduction to Graphs B. and so on. You could start by making the following table: x x2 3 y  x2  3 2 4 3 7 1 1 3 4 0 0 3 3 1 1 3 4 2 4 3 7 3 9 3 12 4 16 3 19 Alternatively. when x  1. If y  f(x). when x  2. then when x is 0. x. y will have a different value for each value of x. when x  3. so that we can obtain a succession of values for y by giving successive values to x. We can also find a single function of x by substituting that value directly into the given equation. we usually tabulate our values for x and y. For convenience. as follows: x yx3 0 3 1 4 2 5 3 6 4 7 5 8 Let's look at another example. y will be equal to 3. It is denoted by f(x). y  5.

For example. in Figure 1 1 5. B is (4.1). 4. Complete the table for y  3x  2.4. If f(x)  5x  2 find: (a) (b) f(2) f(3).Introduction to Graphs 125 Questions for Practice 1 1.1). One way of showing how these values change is by means of a graph.1 2 ). Representation From the tables we see that as the value of x changes.5 3.5 1 1. so does the value of y. x y  3x  2 2. D is (5.5 2 2. © ABE and RRC . Which one of the following functions could be represented by the following table? yx4 y  2x  1 x y 3 7 y  x2  6x  16 y  x3  1 5 11 7 15 9 19 11 23 0 0.2). E is (1. C is (3. If f(x)  x2  2 find: (a) (b) f(2) f(3) Now check your answers with those given at the end of the unit. A is the point (2. 2 ).0) and F is (0. and the second figure is the value of y along the scale of the y (vertical) axis. The values in the table can be expressed as a pair of co-ordinates which may be plotted on a graph. The first figure is the value of x along the scale of the x (horizontal) axis.

(0. if we had to write down the ordered pairs of the function 2x  1 for integral (whole number) values of x from 1 to 2 inclusive. (0.7) (5. As the name implies. Thus. logarithmic and hyperbolic equations.5).e. This is a very useful and helpful way of noting the points we required in the graphical representation of a function. containing the square of the independent variable. exponential. y  ½x2  3). called a parabola.e. C. Such equations always give a curved graph with a characteristic "U" shape. (2. you will normally be told the values of x with which you are concerned.1). and if the function was x2  3 for the same values of x it would be (1.15) (9.19) (11.4).23) These are then the points which we would plot if we were going to draw the graph of the function in question. When dealing with ordered pairs. (1.1). (1.11) (7.3). such as y  2x  3. quadratic. y  5x  1).126 Introduction to Graphs Figure 5. such as y  2x2 x  1. a Hyperbolic equations – we shall look at graphs from the expression y  where a is a x constant.3).  Linear equations – equations of the first degree in x and y (i.7). but no higher power. Such graphs have a different curved shape – a hyperbola. (2. The table in Question 2 of Questions for Practice 1 could be written as a set of ordered pairs as follows: (3. Quadratic equations – those of the second degree in x and y (i. we would write (1.   © ABE and RRC . containing no higher power of x than the first.4).4: Co-ordinates for y  f(x) 3  B 2 1   E 2 A 6 4 2 0 1 4  D 6  C  2 F Ordered Pairs Another way of describing the varying values of x and y is by means of a set of ordered pairs. the graph of expressions like these will always be a straight line. GRAPHS OF EQUATIONS In this section we shall consider graphs of linear.

Suppose we choose x  0. It is even better if we know three points. using the third as a check. where m and c are constants. If we choose any three values of x we can find the corresponding values of y and this will give us the co-ordinates of three points which lie on the line. y  b  ax.5: Graph of y  2x – 3 4 2 0 -3 -2 -1 -2 -4 -6 -8 0 1 2 y x 3 © ABE and RRC .) To draw a line we must know two points on it. equations of both exponentials and logarithms produce non-linear lines. Taking the given equation. When plotted. This will always give a straight line. So. Example 1: Draw the graph of y  2x  3. x  2 and x  2.Introduction to Graphs 127  Exponential and logarithmic equations – an exponential or logarithmic function occurs when the independent variable appears as the exponent or as a logarithm. on joining them.g. Linear Equations A linear equation is of the form y  mx  c. we have the graph of y  2x  3. e. y  2x  3: when x  0: y  0  3  3 when x  2: y  (4  3)  1 when x  2: y  (4 3)  7 This is better tabulated as: x y  2x  3 0 3 2 1 2 7 We now have three points which we can plot and. the minimum number of points you ought to plot for a linear graph is three. (Note that such an equation can be written in different forms. Figure 5.

0 and 2. for values of x from 3 to 2. Then compare your result with Figure 5. (0. mark the scale. Figure 5. This is because the constant c in the general equation y  mx  c is always the point where the line passes through the y-axis (i. Draw the graph of y  3x  2. Now draw your two axes to cover the range of values for x specified (although it is good practice to extend them beyond this – leaving some spare room at each end of each axis).6.e. x y  3x  2 You should have found the points: (3.6 you will notice that the line y  2x  2 meets the y-axis at the point where y  2. A further fact which emerges is that when x goes from 0 to 2 (i. The y intercept is where the line crosses the y-axis and the x intercept is where the line crosses the x-axis. increases by four). (2. 7). The point at which the line crosses an axis is called the intercept. 2). plot your points and join them up. © ABE and RRC . We have thus found that in the equation y  mx  c: the constant c  the point where the line crosses the y-axis the constant m  the slope of the line. increases by two). We shall return to these points in some detail later. Then draw up a table of values – fill in the spaces for yourself to find the co-ordinates of the points to be plotted. y rises from 2 to 8 (i. 8).6: Graph of y  2x + 2 8 6 4 2 3 0 2 y x -4 -2 0 0 -2 -4 -6 -8 2 4 If you look carefully at Figure 5. say 3.128 Introduction to Graphs Example 2: Try this one yourself. where x  0). The change in the value of y for a change of one in the value of x is given in the equation y  mx  c by the constant m.e.e. First choose three values of x.

from x  4 to 2. having divided both sides of it by 3: y 2x  1 3 Now choose some values of x – let's say 4. So we rewrite the equation.7. 1. 2.Introduction to Graphs 129 Example 3: Suppose we are asked to draw the graph of 3y  2x  1. we must change the equation to the form y  mx  c. as in the title to Figure 5.7: Graph of 3y  2x + 1 3 y 1 x -2 -1 -1 0 1 2 3 4 -3 © ABE and RRC . Before we can work out the table of values in this case. Figure 5. Note that. Working these out gives us the following table of values: x y 4 3 1 1 2 1 Now plot these points and join them. this is still a graph of the original equation.

Taking values of x between 4 and 10. Figure 5.e. cross) at A (6. we get the following tables of values: For y  1 3 x x y 3 1 2 10 0 0 4 4 3 1 10 2 For y  8  x x y The graphs are shown in Figure 5.2) © ABE and RRC .8: Graphs of y  ⅓ x and y  8 – x 10 8 6 4 2 y A x -4 -2 0 0 -2 2 4 6 8 10 You will notice that the lines intersect (i. we could draw the graphs of y  1 3 x and y  8  x on the same axes.8.130 Introduction to Graphs Example 4: We are not limited to having just one line plotted on a graph. For example. We can draw two or more graphs using the same axes – a very useful method of comparing different equations of the same variables.

Draw the graph of y  5x  4. For values of x from 3 to 6. and draw the line through them: (i) (ii) (iii) (iv) (3. The graph of y  c must. draw on the same axes the graphs of: (i) 7.3) (1. and plays a very important part in calculations concerning a straight-line graph. For values of x from 3 to 6. x  0 is the y-axis. y  2x (ii) y  2x  3 (iii) y  2x  4 Describe briefly the resultant lines and their intercepts. Now check your answers with those given at the end of the unit.3) (0. be a line parallel to the x-axis at a distance c from it. from x  1 to x  2. y is always equal to c. © ABE and RRC . This means y  c. 2. from x  4 to x  2. plot the following pairs of points on squared paper. and those you drew in the Questions for Practice. 4. Note that y  0 is the x-axis itself. for values of x from 4 to 6. For values of x from 3 to 6. we can start to identify a number of characteristics of straight-line graphs. Whatever value x has. draw on the same axes the graphs of: (i) y5 (ii) x  5 (iii) y  5x (iv) y1x 5 Describe briefly the resultant lines. On the same axes.0) (3. 3. draw on the same axes the graphs of: (i) yx2 (ii) 3y  6  x Identify the point of intersection. Draw the graph of y  2x  3. In the same way.6) (0.0) (1. Draw the graph of y  2  x. x  c is a line parallel to the y-axis at a distance c from it.5) (2. Characteristics of Straight-Line Graphs If you consider the graphs we have drawn as examples.  Lines parallel to an axis Consider the case where m  0. The value of m is known as the coefficient of x. 5. including 0.Introduction to Graphs 131 Questions for Practice 2 1.2) (4. then.2) Write down the co-ordinates of the points where (i) and (ii) each cut the lines (iii) and (iv). 6. The general equation for a straight line is y  mx  c where m and c are constants which may have any numerical value.

in order to draw a quadratic graph accurately. b and c are constants. © ABE and RRC . This will always give you a curved graph.9: General shape of a quadratic equation y y x or x where a is negative where a is negative In contrast to linear graphs. you need as many points as possible – especially around the dip. y  4  3x) the line will slope down from left to right instead of up. When x  0 it is clear that y  0. the lines are parallel. if m is kept the same and c has different values. Thus.  Intercepts The distances from the origin to the point where the line cuts the axes are called the intercepts on the axes. if m is negative the line slopes down. if m is positive the line slopes up from left to right. The general shape of a quadratic equation will be as shown in Figure 5. as it does when the coefficient of x is positive. Quadratic Equations A quadratic equation is of the form y  ax2  bx  c. Figure 5.9. then the equation of the line is y  mx.0). the intercept on the y-axis is always at c. They are found by putting x  0 and y  0 in the equation of the graph.  Direction of slope If the coefficient of x is negative (for example. we get a set of parallel straight lines. where a.  Parallel lines When the coefficient of x is the same for several lines.132 Introduction to Graphs  Lines through the origin If c  0. In y  mx  c. so the graph of y  mx is a line through the origin (0. in y  mx  c. and the part that is usually asked for is the part that does a "U" turn. Thus. The maximum and minimum values of the quadratic equation can be found by careful inspection of the graph.

the first step is to draw up a table of values for the given range of x.0)). What are the co-ordinates at which the function y  x2 is minimised? As ever. x y  x2 3 9 2 4 1 1 0 0 1 1 2 4 3 9 Plot these seven points on the graph and then join the points with a smooth curve.10. From the graph you will see that function y  x2 is minimised where y  0 and x  0 (i.e. Figure 5. the co-ordinates where y  x2 is minimised are (0. as shown in Figure 5.10: Graph of y  x2 10 y 8 6 y  x2 4 2 x 0 -3 -2 -1 -2 0 1 2 3 © ABE and RRC .Introduction to Graphs 133 Example 1: Draw the graph of y  x2 taking whole number values of x from x  3 to x  3 inclusive.

x  0 and y  0.25 the values of x and y where the function y  2x2 is minimised. Table of values: x x2 y  2x2 3 9 18 2 4 8 1 1 2 0 0 0 1 1 2 2 4 8 3 9 18 Plot the points and draw the graph as in Figure 5. the value of y is at the point C: y  3.3 when the function y  2x2 is minimised.25 and x  2.134 Introduction to Graphs Example 2: Draw the graph of y  2x2 from x  3 to 3. Figure 5.25. at points A and B: x  2.25 when x  1.11: Graph of y  2x2 18 16 14 12 y y  2x2 A 10 8 6 4 B y  10 y  1. Then from the graph find: (a) (b) (c) the values of x when y  10 the value of y when x  1.25 C 2 x 0 -3 -2 -1 0 1 2 3 Reading off from this: (a) (b) (c) when y  10. there are two values of x.11. © ABE and RRC .

Introduction to Graphs 135 Example 3: Draw the graph of y  x2  2x  1 from x  2 to 3. we can see that the minimum value of y occurs when x  1. At what value of x does the minimum value of y occur? Table of values: x x2 2x 1 y  x2  2x  1 2 4 4 1 9 1 1 2 1 4 0 0 0 1 1 1 1 2 1 0 2 4 4 1 1 3 9 6 1 4 Plot the points and draw the graph as in Figure 5. © ABE and RRC .12. Figure 5.12: Graph of y  x2 – 2x + 1 10 y 8 6 4 2 x -2 -1 0 0 1 2 3 Reading from the graph.

3. Equations with x as the Denominator a where a is a constant. 4. since the expression meaning). the direction of which may vary.136 Introduction to Graphs Questions for Practice 3 1.33 4 0.e. © ABE and RRC . From the graph find: (a) (b) the output at which profit is maximised the output range at which profits generated amount to at least £8. 1 2 x2 from x  4 to 4.000. x They result in a characteristic curve.2 the values of x when y  7. Draw the graph of y  From the graph find: (a) (b) 2. the value of y when x  3. Now check your answers with those given at the end of the unit.5 3 0. Find the values of x when y  9.13.17 1 is not defined (i. x y 1 x 1 1 2 0.25 5 0. These equations (hyperbolic equations) take the form of y  Example 1: Let's look at the case where a  1: y 1 x We'll take x to have values from x  1 to x  6. Draw the graph of y  x2 from x  3 to 3. which gives the table following and the graph in Figure 5. where x ('000) represents output.2 6 0. Draw the graph of the profit (£000s) function x2  8x  1 from x  0 to 8. it has no 0 Note that we have avoided taking x  0. Draw the graph of y  4x2 from x  2 to 2.

6 0.2 0 0 2 4 1 x 6 a . There is also no point corresponding to x  0.8 4 x 6 0. x  1.5.Introduction to Graphs 137 Figure 5. find the value of x when y  2. In each case.14: Graph of y  4 3 2 1 0 0 2 4 6 8 Reading from the graph. Table of values: x y 4 x 1 4 2 2 3 1.5.2 1 0. This graph illustrates the general shape of graphs representing y  Example 2: Draw the graph of y  4 for values of x from x  1 to x  8. © ABE and RRC . as x x gets bigger.13: Graph of y  1.67 7 0.7.8 0.57 8 0.33 4 1 5 0. x From the graph.4 0. when y  2.5 Figure 5. y gets smaller.

Just as we saw that the shape of the quadratic graphs changed according to whether the coefficient of x2 is positive or negative. Figure 5.138 Introduction to Graphs Questions for Practice 4 1.15. so the shape of graphs developed from equations where x is the denominator change:  a (i. x From your graph.e.5. Draw the graph of y  2 from x  1 to x  6. find the value of x when y  1. x a x    These are illustrated in Figure 5.15: Shape of graphs from the equation y a x a x 0 x a x a x © ABE and RRC . where both the a and x are positive) gives a graph which slopes down from left x to right a gives a graph which slopes down from left to right x a gives a graph which slopes up from right to left x a gives a graph which slopes up from right to left. Now check your answers with those given at the end of the unit.

00 1 2.16: Graph of equation y  ex 160 140 120 100 80 60 40 20 0 -2 -1 0 1 2 3 4 5 Reading from the graph.39 3 20. In the final part of this section we will now plot graphs for exponential and logarithmic equations. © ABE and RRC .72 2 7. x  2. Using a scientific calculator.1 4 54. The mathematics behind exponentials and logarithms was discussed in Study Unit 4. Example 1: Let us begin by plotting the graph of an exponential function.37 0 1.14 -1 0. when y  10.Introduction to Graphs 139 Exponential and Logarithmic Equations So far in this unit we plotted and interpreted graphs for linear and quadratic equations (also known as polynomial functions). or otherwise. Plot a graph of this exponential function and from the graph find the value of x when y  10. calculate the value of y  ex for the values of x from x  2 to x  5.60 5 148.5 (approximately). Table of values: x y  ex -2 0.41 Figure 5.

0 0.405 2.00 1.099 3.5 0 0 -0.5 0. Use a scientific calculator to construct a table of values for y  log10(x) when x  0.5 0.386 Figure 5.5.693 1. 10.5 1 0. Plot a graph of this logarithmic function and from the graph find the value of x when y  0.0 1.5 to x  4. 20 and 25 (give your answers to 2 decimal places). (a) (b) Construct a table of values for S for t  0. x  1. 20 and 25 (give S values to the nearest whole number).1625t). 10.0 0. © ABE and RRC .0 1. when y  0.5.0.5 1. calculate the values of y  ln(x) for the values of x from x  0.5 1 2 3 4 -1 Reading from the graph. Using a scientific calculator.253 4. Now check your answers with those given at the end of the unit. 2. Plot these values on a graph. Use the table to draw a graph of S against t and from your graph estimate the value of S when t is very large.17: Graph of equation ln(x) 1. Table of values: x ln (x) 0.916 3. 15. Questions for Practice 5 1. The sales S (in thousands of units) of a product after it has been on the market for t months is given by the equation: S  100(1  e0.693 2.5 0. 15.140 Introduction to Graphs Example 2: In this example we will plot a graph of the natural logarithmic function.

then the gradient is negative. Figure 5.3) and (4.8) and find the gradient of the line joining them.20. This is denoted by introducing a negative sign. 8) y 9 8 7 6 5 5 4 3 2 1 3 1 2 3 4 5 x (b) Make this line the hypotenuse of a right-angled triangle by drawing in a horizontal "base" and a vertical "height". we shall examine the method by reference to plotting the line on a graph.20: Gradient of the line connecting points (1. The procedure would be as follows: (a) Plot the two points on a graph and join them to give a sloping line. as shown in Figure 5. Figure 5.19. Firstly.142 Introduction to Graphs If a line slopes down from left to right (instead of up). 2 Gradient of a Straight Line We can easily calculate the gradient of a straight line joining two points from the co-ordinates of those points.19: Negative gradient 2 s 2 The gradient of the line s is  2  1. Example 1: Plot the points (1. 3) and (4. © ABE and RRC . as illustrated by Figure 5.

at step (c). we could have worked out the lengths of the base and the height from the co-ordinates of the points: length of the base  difference between x co-ordinates  4  1  3 length of the height  difference between y co-ordinates  8  3  5 Example 2: Plot the points (1. 8) is 1 3 Note that. 4) and (2. Plot the points.Introduction to Graphs 143 (c) Work out the lengths of the base and the height from the graph: base  3 height  5.3) and find the gradient of the line joining them. the gradient is given by the equation Note that the slope is positive.4) and (2. Therefore the gradient   height 1   7  2 3 3 base © ABE and RRC .21: Gradient of the line connecting points (–1. join them and construct a right-angled triangle with the line forming the hypotenuse: Figure 5. Note that the slope is negative. height 5 2   13 base 3 As we have seen. –3) y 4 3 2 1 7 2 1 1 2 3 x 1 2 3 3 This time we shall work out the lengths of the base and the height from the co-ordinates of the points. 2 Therefore. 3) and 4. the gradient of the straight line joining (1. base  2  (1)  3 height  4  (3)  7 We can check that this is correct by reference to the plotted triangle on the graph.

22: Gradient of a straight line y B B (i) (ii) A C C A O x Note that these lengths are directed lengths: AC is positive if the movement from A to C is in the same direction as O to x. we find the following:   In line (i). take any two points A and B on the line. We can further see that AC is the difference between the co-ordinates along the x-axis and CB is the difference between the co-ordinates along the y-axis. (0. and CB parallel to the y-axis.e.2) and (2. say. The gradient is then given by the ratio CB : AC. Therefore. AC and CB are both positive.3). CB is positive if the movement from point C to B is in the same direction as O to y. (The point at O is the origin. Figure 5.0).) Taking the lines in Figure 5. CB is positive. AC is negative. (5. AC In line (ii). the gradient would be: 32 1  3  ( 2)   2  ( 1) 21 3 © ABE and RRC . i.2) as follows: 2 1 1  75 2 For points (1. In the same way. To find the gradient of a line. we could calculate the gradient where the points are. and negative if the movement from A to C is in the opposite direction to O to x.1) and (7. and the gradient is negative. Draw AC parallel to the x-axis. hence the gradient CB is positive.144 Introduction to Graphs General formula We can now develop a general formula for the gradient of a straight line by reference to its co-ordinates.22.

23. c. and P is another point along the line.23: Proving the equation of a straight line y B D P G O A x OAPB is a rectangle and OAGD is a rectangle. Note that c is the intercept that the line makes with the y-axis. we can find the gradient of the line (m) immediately. We need to put the equation into the general form y  mx  c: y  6x  19 3 © ABE and RRC . This is shown in Figure 5. Let: OA  x. given any linear equation. Using this method. the intercept OD  c and the gradient of DP  m. Also. OB  y.Introduction to Graphs 145 Equation of a Straight Line and Gradient We can now go back to consider aspects of the equation of a straight line which we have met before. Consider a general straight line drawn on a graph. its equation can be found immediately. Then: m   PG PG BD   DG OA OA BO  DO yc  OA x yc we get: x Rearranging the equation m  mx  y  c y  mx  c This is the general equation of any straight line. given any straight-line graph. its intercept is negative. D is the point on that line where it crosses the y-axis. If the line cuts the y-axis below the origin. Figure 5. For example: Find the gradient of a line with the equation 6x  3y  19.

the equation of this line is: y or 6y  5x  30 5 x5 6 Questions for Practice 6 1. the gradient is 6  2 3 Finally. (a) (b) (c) (d) (e) (f) 2. consider the graph in Figure 5.1) and (5.146 Introduction to Graphs Therefore.0) and (1.0) (7.3) and (3.4) (0. Figure 5. Find the gradients of the straight lines joining the following points.7) (2. 10  5 5  6 6 Therefore.7) Find the gradients of the lines with the following equations: (a) (b) (c) 2y  3x  6  0 7y  4x 7y  1  2x © ABE and RRC .5) and (2.2) and (4. c  5.24: Deriving the equation of a straight line 10 8 6 4 2 0 0 2 4 6 8 The gradient of the line.24.1) (4.6) and (1. (1. m  The intercept.6) (2.

USING GRAPHS IN BUSINESS PROBLEMS In this final part of the unit. Now check your answers with those given at the end of the unit. Business Costs All businesses incur costs in the production and supply of the goods and/or services which they provide. there are many ways of looking at costs. which at heart is simply: revenue from sales less costs of production and supply (or cost of sales) The analysis of costs is a subject of its own and.   Fixed costs are those which the business must always meet. (Note that fixed costs are only fixed in the short term – over a long period. Draw the graph of y  3x for values of x from 3 to 3. They are actual examination questions.) If we consider variable costs. as the level of production changes. they may change. Whether they print six or six hundred books a week. 3.5) and (3. we can say that they are a function of the level of production. and the variable costs would include paper and ink. not least in calculating profit. we shall consider how we can use the properties of graphs we have been examining to analyse and solve business problems. On the same axes draw the line y  5. E. Draw an x-axis from 3 to 5 and a y-axis from 4 to 4. Find (a) the gradient and (b) the equation of the straight line joining them.3). 5. Write down the co-ordinates of the point where the two lines intersect. This means that. its fixed costs would include the rent on its offices. Plot the points (6. We can turn this observation into an algebraic expression: let: x  number of units produced (level of production) b  variable costs © ABE and RRC . so do the variable costs. Thus. These are an important consideration in many aspects of business finance. Variable costs are those which vary with output – such as the cost of raw materials used as part of the production process. regardless of the number of units produced – such as rent on buildings. for example by acquiring different premises which may be cheaper. if we take the case of a printing company. within that. printing room and warehouse. On them draw the graphs of y  x  1 and 2y  4  x. Write down the co-ordinates of the points where the line and the curve intersect.Introduction to Graphs 147 The next three questions bring together aspects of the whole unit so far. the fixed costs remain the same. One of these is to divide them into fixed and variable costs. 2 2 4. but the variable costs rise with the number of books produced.

If we know that one other point on the graph is (9. Let us suppose that we do have such values and can produce such a graph (Figure 5.25): Figure 5. if we know the level of fixed costs.25: Graph of relationship between variable costs and level of production Costs 6 (£000s) 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 Variable costs Production (thousands of units) We can work out the gradient of the line on this graph. no variable costs will be incurred. since when there is no production.26. we can show this on the same graph as in Figure 5. 3). 0). then the gradient (which is m) is: m 30 3 1   90 9 3 We also know that the intercept (c) of the line is 0. One of those co-ordinates must be (0. so we can express the relationship of variable costs to units of production by the following equation in the form of y  mx  c: a or 3a  x Now consider fixed costs. based on two co-ordinates. 1 3 x0 © ABE and RRC . So.148 Introduction to Graphs Then: b  f(x) This is where we started the unit! We know that we can draw a graph of this relationship if we know some of the values for x and y. These do not vary with the level of production.

with an intercept at 2. in Figure 5.Introduction to Graphs 149 Figure 5.26. we get the following equation: yab Substituting from the equations established for a and b. and using the notation from above. We can say that: total costs  variable costs  fixed costs If total costs  y.27. Note that the line for total costs is in fact a combination of the lines for variable and fixed costs. We could work out a table of values for it (which we shall skip here) and draw a graph. but in their total costs. we get: y 1 3 x2 This is a standard linear equation. This has been done. This is represented by the equation: b  2 (where b is fixed costs) Businesses are not only interested in costs as either fixed or variable. © ABE and RRC .26: Graph of fixed costs and level of production Costs 6 (£000s) 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 Variable costs Fixed costs Production (thousands of units) We can see that this line is parallel to the x-axis. again on the same graph as in Figure 5. Its intercept is at the level of fixed costs and its gradient is that of variable costs.

for company A. it is a useful way of looking at costs. Company A charge a call out charge of £5. In real life. company A is cheaper. For example:   at 7 miles.27: Graph of total costs and level of production Costs 6 (£000s) 5 4 3 2 Total costs Variable costs Fixed costs 1 0 0 1 2 3 4 5 6 7 8 9 Production (thousands of units) Note that the chart is only a model. the costs is: y  2x The graph is shown in Figure 5. the cost of using each company is the same. Consider the following situation.28.150 Introduction to Graphs Figure 5. it is cheaper to use company B. variable costs do not necessarily rise at a constant rate with increases in production. © ABE and RRC . whilst company B charge £2. For any distance above this. Thus at that point. which is why we can show the variable and total costs lines as linear.00 per mile but no call out charge. A business wishes to use a courier company on a regular basis. The point at which the two lines intersect means that the values of x and y are the same for each line. This takes place at 5 miles. Some of these situations are shown by dotted lines on the graph. A charges £7 and B charges £4. whereas for distances less than five miles. total cost may be defined as : yx5 For company B. Which one is more economical? We can develop equations for these charges and then use a graph to show which company would be more economical at different journey lengths: Let: x  number of miles y  total cost (£) Then. A charges £12 and B charges £14 at 2 miles.00 per mile. Despite this.00 plus £1.

(Note that we are assuming that all units produced are sold.Introduction to Graphs 151 Figure 5.27. Adding the graph of this to the costs graph shown in Figure 5. revenue exactly equals costs.29: Costs and revenue graph Costs (£000s) Revenue 8 6 Break-even point 4 Total costs 2 0 0 1 2 3 4 5 6 7 8 9 Sales(thousands of units) The point at which the revenue and cost lines intersect is known as the break-even point.) Figure 5.000. Revenue is the total income from sales of the units produced. with both costs and revenue being £3. we can say that y  x where y is the income from sales (in £) and x is the number of units sold.28: Comparison of courier company charges Costs 20 (£) 15 Company B Company A 10 5 0 0 2 4 6 8 10 Journey length (miles) Break-even Analysis We can further develop the costs graph by including revenue on it. this takes place at sales of 3. Here.000 units. © ABE and RRC . If we assume that each unit sells for £1.29. we get Figure 5. At that point.

giving the average holding of 50.000Q sprockets.) (36  £30)  (£2  50)  £1. We order as soon as existing stock runs out. monthly orders of 300.630. If it buys in lots of 1. Let us consider another problem. (As we place an order for 3. we receive the whole order instantaneously – unrealistic but let us keep the problem simple. The relevant costs for each option above are: (a) £30 for one order  storage costs of £2  1.) (6  £30)  (£2  300)  £780. i. The graph of stock level is a saw-tooth.200 each. then the stock will last for Q days.) We can see that the cheapest option is (b).) (3  £30)  (£2  600)  £1. Make 36 orders at 10-day intervals.30).000 sprockets every day. giving the average holding of 300.) (b) (c) (d) (e) (12  £30) for 12 orders  (£2  150) storage  £660 (In this case. the average stock held in storage will be 1.800 units.000Q sprockets in stock immediately after a purchase has been made. Q days later this will be down to zero. We have several options as to how we could tackle this: There are many other possibilities but these will serve our purpose. The cost of keeping one unit in stock for a year is £2.800  £3.290. As soon as we order.Introduction to Graphs 153 Suppose we have the following problem:      (a) (b) (c) (d) (e) We want to order 3. Place a single order for the whole year. The average stock level is 500Q (see Figure 5. so the average holding is 150. The cost of placing the order is £30. (600 is the average holding on orders of 1. It could purchase daily.180 (36 orders of 100 each. when another purchase will be made.600 units and use them evenly through the year. There will be 1. but considers this to be unacceptable as the cost of raising an order and sending a van to a cash-and-carry to collect the goods has been estimated at £100. Make three orders at four-monthly intervals.e. the orders are for 300 each.600 units over one year and these units are used at a constant rate. Make six orders bimonthly. © ABE and RRC . (Six orders of 600 each. But is it really the cheapest? We would need to try many other options before we could be sure. A company called Oily Engineers Ltd uses 1. Make 12 monthly orders.

500 Sprockets purchased in batches of 4. For example. assume that there are 250 working days each year: (a) (b) (c) Sprockets purchased in batches of 1. as it is a fixed amount per order.32.000.5  £6.000.250. the cost per annum halves.000.000 Sprockets purchased in batches of 2. i.30: Stock levels over time Stock level 1. on average. then the warehousing costs are doubled as there are.31. a batch every other day: cost  £100  125  £12. a batch every day: cost  £100  250  £25. © ABE and RRC .e.000Q 0 Q 2Q 3Q Days The warehousing cost is linear – if the amount purchased in each order is doubled. i. Figure 5. a batch every fourth day: cost  £100  62. twice as many items in the warehouse.e. Each time the amount purchased is doubled. i.154 Introduction to Graphs Figure 5.e.31: Warehousing costs Cost Size of purchased lot The cost of purchasing is not linear. The cost per annum depends on the number of orders placed. The graph is shown in Figure 5. This shown in Figure 5.

quarterly or monthly intervals could be examined in an attempt to predict their future behaviour.33: Total costs of purchasing Cost Total cost Minimum cost Storage or warehousing cost Purchasing cost Size of purchased lot Thus. The first step in analysing the data is to plot the observations on a scattergram.32: Cost of purchasing (per annum) Cost Size of purchased lot These costs can be added to the warehouse costs to give a total cost per annum. © ABE and RRC . This can give us a good visual guide of the trend in the time series data. Such sets of values observed at regular intervals over a period of time are called time series. with time evenly spaced across the x-axis and measures of the dependent variable forming the y-axis. Figure 5.33. we can identify the economic order quantity from the minimum cost position on the graph. to plan future strategies and policies.Introduction to Graphs 155 Figure 5. sales values or unemployment levels recorded at yearly. The analysis of this data is a complex problem as many variable factors may influence the changes which take place. For example. The graph of this is shown in Figure 5. Plotting Time Series Data Businesses and governments statistically analyse information and data collected at regular intervals over extensive periods of time.

The following table shows the number of days lost through sickness over the last five years. The scattergram of a time series is often called a historigram. (Do not confuse this with a histogram.34.34. say thermometers. It is possible to change the scale of the y-axis to highlight smaller changes. Each year has been broken down into four quarters of three months. We have assumed that the number of employees at the factory remained constant over the five years.) Note the following characteristics of a historigram:     It is usual to join the points by straight lines. A historigram is simpler than other scattergrams since no time value can have more than one corresponding value of the dependent variable. Intermediate values of the variables cannot be read from the historigram. at such a factory during the course of a year some employees will be absent for various reasons.1: Days lost (per quarter) through sickness at a thermometer factory We can plot this information using a particular type of scattergram. Naturally.156 Introduction to Graphs Let us consider a factory employing a number of people in producing a particular commodity. as shown in Figure 5. depending on the scale of the yaxis. © ABE and RRC . which is a type of bar chart. Year 2003 Quarter 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Days Lost 30 20 15 35 40 25 18 45 45 30 22 55 50 32 28 60 60 35 30 70 2004 2005 2006 2007 Table 5. Every historigram will look similar to the one in Figure 5. The only function of these lines is to help your eyes to see the pattern formed by the points.

seasonal variations. The trend is the change in general level over the whole time period and is often referred to as the secular trend.34. cyclical fluctuations. we can clearly see that lost days from work (per quarter) have increased over the period.34 that the trend is definitely upwards. you only need to concern yourself with the trend. A trend can thus be defined as a clear tendency for the time series data to travel in a particular direction in spite of inter-period fluctuations (be they large or small).35. from approximately 25 days lost per quarter in 2003 to approximately 50 days lost per quarter in 2007. Based on this trend line. and irregular or random fluctuations. However. These are the trend. You can see in Figure 5.34: Historigram (scattergram) of days lost (per quarter) through sickness (2003–2007) Lost 70 days 60 50 40 30 20 10 0 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 2003 2004 2005 2006 2007 There are four factors that influence the changes in a time series over time. in spite of the obvious fluctuations from one quarter to the next. © ABE and RRC . then we would get a trend line similar to that shown in Figure 5. If we were to draw a simple linear trend line through the time series data shown in Figure 5.Introduction to Graphs 157 Figure 5.

35: Historigram (scattergram) of days lost (per quarter) through sickness (2003–2007). with linear trend line Lost 70 days 60 50 40 30 20 10 0 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 2003 2004 2005 2006 2007 © ABE and RRC .158 Introduction to Graphs Figure 5.

Sales (£000) 320 200 150 400 400 210 180 450 800 400 340 900 950 500 400 980 Plot the time series on a historigram. © ABE and RRC . Now check your answers with those given at the end of the unit. Using a ruler. 2. draw a linear trend line through the time series plotted in Question 1 and comment on the trend.Introduction to Graphs 159 Questions for Practice 8 The following table shows sales of premium bonds at a large post office over five years: Year and Quarter 2004 1 2 3 4 2005 1 2 3 4 2006 1 2 3 4 2007 1 2 3 4 1.

3.5 y  2x  1 is the only function which satisfies all values in the table. Table of values: x y  5x  4 1 9 0 4 2 6 Graph of y  5x – 4 6 4 2 0 -2 -1 -2 -4 -6 -8 -10 0 1 2 y x 3 © ABE and RRC .160 Introduction to Graphs ANSWERS TO QUESTIONS FOR PRACTICE Questions for Practice 1 1. 4.5 1 1 1.5 2.5 5.5 0. 0 2 0. (a) (b) (a) (b) f(2)  10  2  8 f(3)  15  2  17 f(2)  4  2  6 f(3)  9  2  11 Questions for Practice 2 1. x y  3x  2 2.5 2 4 2.

Introduction to Graphs 161 2. Table of values: x y  2x  3 4 5 0 3 2 7 Graph of y  2x + 3 8 6 4 2 0 -4 -3 -2 -1 -2 -4 -6 0 1 y x 2 3. Table of values: x y2x 4 6 1 1 6 4 Graph of y  2 – x 6 y 4 2 0 -4 -2 -2 0 2 4 x 6 -4 © ABE and RRC .

e.4.1 when y  7. makes an intercept of 3 on the y-axis and 1½ on the x-axis (i. crosses the yaxis at 3 and the x-axis at 1½).7) to 1 decimal place. y  5x is a line through the origin. cuts (iii) at (1. (i) (i) (ii) (ii) 5.7 and 3. The lines intersect at (3.1) to 1 decimal place. 1 y   5 x is a line through the origin. makes an intercept of 4 on the y-axis and 2 on the x-axis. y  5. y  5 is a line parallel to the x-axis and 5 units above it.7. 4. sloping up from left to right with a gradient 5 in 1 (i. 0). 2 3 ) or (1. cuts (iv) at (2. x  5 is a line parallel to the y-axis and 5 units to the left of it.e. we get: (a) (b) when x  3.2. 2. These three graphs are parallel lines: (i) (ii) (iii) passes through the origin.7 © ABE and RRC . 0. (i) (ii) (iii) (iv) 6. sloping down from left to right with a gradient of 1 in 5 (i. if you go along 1 unit you go up 5 units). Questions for Practice 3 1. along 5 and down 1). Table of values: x x2 y 1 2 4 16 x2 8 3 9 4½ 2 4 2 1 1 ½ 0 0 0 1 1 ½ 2 4 2 3 9 4½ 4 16 8 The graph is shown on the next page.6. 2. 2 2 cuts (iii) at ( 1 3 . 7. x  3. cuts (iv) at (1. Reading off from it. we do not show the graphs themselves.8). 1).e.162 Introduction to Graphs Note that for the next four questions.

Table of values: x y  x2 3 9 2 4 1 1 0 0 1 1 2 4 3 9 Graph of y  –x2 -3 -2 -1 0 0 1 2 3 -2 -4 -6 -8 -10 © ABE and RRC .Introduction to Graphs 163 Graph of y  8 1 2 x2 6 4 2 0 -4 -2 0 2 4 2.

Reading off from it.000 units of output. i. Table of values: x x2 4x2 y  4x2 2 4 16 16 1 1 4 4 0 0 0 0 1 1 4 4 2 4 16 16 Graph of y  –4x2 -2 -1 0 0 1 2 -4 -8 -12 -16 Reading from the graph.e.5 and 1. Table of values: x -x2 8x 1 y  x2  8x  1 0 0 0 1 -1 8 2 -4 16 3 -9 24 4 -16 32 5 -25 40 6 -36 48 7 -49 56 8 -64 64 1 1 1 8 1 13 1 16 1 17 1 16 1 13 1 8 1 1 The graph is shown on the next page.5. 4. y  8 or above). between x  1and x  7). when y  9. when x  4. profits generated are in excess of £8. © ABE and RRC .000) at 4. At an output of between 1.000 and 7.000 (i. we get: (a) (b) Profit is maximised (£17.e.e. x  1.164 Introduction to Graphs 3.000 units (i.

33 Graph of y  2 1.3. Table of values: x y 2 x 1 2 2 1 3 0.67 4 0.5 0 0 1 2 3 4 5 6 When y  1.5 2 x 5 0. x  1.5.5 1 0. © ABE and RRC .Introduction to Graphs 165 Graph of y  x2 + 8x + 1 18 Profit (£000s) 16 14 12 10 8 6 4 2 0 0 1 2 3 4 5 6 7 8 Output (£000s) Questions for Practice 4 1.4 6 0.

8 0.70 10 1.1625t ): t S 0 0 5 56 10 80 15 91 20 96 25 98 Graph of S 100(1  e0.2 0 0 5 10 15 20 25 2.00 15 1.18 20 1.4 0.1625t) 120 100 80 60 40 20 0 0 5 10 15 20 25 From the graph.6 0.40 Graph of y  log10(x) 1.2 1 0.30 25 1. Table of values for S 100(1  e0.00 5 0.166 Introduction to Graphs Questions for Practice 5 1. © ABE and RRC .000. when t is very large S will be approximately 100.4 1. Table of values for y  log10(x): x y  log10(x) 0 0.

© ABE and RRC . 73 4  2 3 1 2 6  (1) 7 1   23 53 3 7 6  ( 1) 1   23 ( 2)  1 3 5 5 50   ( 4 )  2 6 6 4 0  ( 4 )   1 2 7  ( 1) 8 7  ( 2) 9 1   22 40 4 y y y 3x  6 2 4 x 7 2x  1 7  gradient   gradient   gradient  3 2 4 7 2 7 Tables of values: x yx1 2y  4  x y2 x 2 1 2 1 2 3 4 1 0 5 4 x 2 2 0 2 0 2 4 2 2 0 x 1 2 1 x 3 y2 The graph is shown on the following page. (a) (b) (c) 3. (a) (b) (c) (d) (e) (f) 2.Introduction to Graphs 167 Questions for Practice 6 1.

5 2 6 1 1.5 0 0 1 1.168 Introduction to Graphs Graph of y  x – 1 and 2y  4 – x 4 3 yx1 2y  4  x 2 1 0 -3 -2 -1 -1 -2 -3 -4 0 1 2 3 4 5 Reading from this.5 2 6 3 13. 5). 5) and (1. you will see that the line and curve intersect at (1. Table of values: x y  3x 2 2 3 13. the lines intersect at point (2.8. 1) 4.5 Graph of y  14 12 10 8 6 3x 2 and y  5 2 3x 2 2 y  y5 4 2 0 -3 -2 -1 0 1 2 3 From the graph. © ABE and RRC .8.

1).Introduction to Graphs 169 5. Therefore. the equation is: y  x  1 or 3y  2x  3 Questions for Practice 7 1. © ABE and RRC .100 Standard teapots. 5 4 3 2 1 0 0 1 2 3 4 5 6 (a) (b) Gradient  53  2 3 63 2 3 The line cuts the y-axis at (0. Break-even chart: Revenue/ Costs (£000s) 50 40 30 20 10 0 0 1 2 3 4 Sales (000s) The break-even point is at sales of 3.

The historigram clearly shows that sales of premium bonds have increased over the period. Break-even chart Revenue/ Costs (£000s) 250 200 150 100 50 0 0 1 2 3 Sales (000s) The break-even point is at sales of 1. © ABE and RRC .170 Introduction to Graphs 2. Linear trend is shown on the historigram above.800 bicycles (to the nearest 100). Questions for Practice 8 1.000 per quarter in 2007. from approximately £250. Time series historigram: Sales (£000s) 1000 800 600 400 200 0 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 2004 2005 2006 2007 2.000 per quarter in 2003 to approximately £700.

Data for Statistics Preliminary Considerations Types of Data Requirements of Statistical Data Methods of Collecting Data Published Statistics Personal Investigation/Interview Delegated Personal Investigation/Interview Questionnaire Classification and Tabulation of Data Forms of Tabulation Secondary Statistical Tabulation Rules for Tabulation Frequency Distributions Simple Frequency Distributions Grouped Frequency Distributions Choice of Class Interval Cumulative Frequency Distributions Relative Frequency Distributions Page 172 173 173 174 174 175 175 176 176 176 177 179 181 183 185 186 186 188 189 189 191 B. C. D.171 Study Unit 6 Introduction to Statistics and Data Analysis Contents Introduction A. Answers to Questions for Practice © ABE and RRC .

then the data must all be in the same units. class boundary.  Definition of the units The results must appear in comparable units for any analysis to be valid. Preliminary Considerations It is important to be aware of these issues as they impact on the data which is to be collected and analysed. grouped. the more readings we need to take. the more precisely we wish to estimate a value. © ABE and RRC . In general. If the analysis is going to involve comparisons. class limits and class interval prepare simple. The degree of precision required in an estimate might affect the amount of data we need to collect. cumulative and relative frequency distributions from given data. and that effort is not wasted by collecting irrelevant data. A. are ages of individuals to be given to the nearest year or to the nearest month or as the number of completed years? If some of the data is to come from measurements. and prepare tables from given data explain the concept of a frequency distribution and define the terms class frequency. there are some important points to consider when planning a statistical investigation. then the accuracy of the measuring instrument will determine the accuracy of the results. DATA FOR STATISTICS A statistical investigation involves a number of stages:     definition of the problem or issue collection of relevant data classification and analysis of the collected data presentation of the results. some in number of inspected batches and so on. or with others of the same kind?  Accuracy of the data To what degree of accuracy is data to be recorded? For example. Are we interested in all departments of our business. So even before the collection of data starts. some in weight of items. The problem as originally put to the statistician is often of a very general type and it needs to be specified precisely before work can begin.  Scope of the enquiry No investigation should be commenced without defining the field to be covered.Introduction to Statistics and Data Analysis 173    explain the reasons for classifying and tabulating data.  Exact definition of the problem This is necessary in order to ensure that nothing important is omitted from the enquiry. It is no use just asking for "output" from several factories – some may give their answers in numbers of items. or only some? Are we to concern ourselves with our own business only.

 Quantitative and qualitative data Variables may be either quantitative or qualitative. etc. primary data is data which is both original and has been obtained in order to solve the specific problem in hand. 1½. such as colour of hair. A continuous variable may take any value between two stated limits (which may possibly be minus and plus infinity). 2. 2. to which we shall restrict discussion here. for example. if the variable is the number of children per family. for example. Alternatively. in Britain. (a) (b) Break down a non-homogeneous group into smaller component groups which are homogeneous. However each possible non-numeric value (hair colour say) may be associated with a numeric frequency of occurrence. and study them separately. Requirements of Statistical Data Having decided upon the preliminary matters about the investigation. is a continuous variable.  Homogeneity The data must be in properly comparable units. then the only possible values are 0. the statistician must look in more detail at the actual data to be collected. © ABE and RRC .  Primary and secondary data In its strictest sense. A discrete variable however can take only certain values occurring at intervals between stated limits. This is because a person's height may (with appropriately accurate equipment) be measured to any minute fraction of a millimetre. these interval values are the set of integers (whole numbers). Use units such as "output per person-hour" to compare the output of two factories of very different size. a child may be considered equal to half an adult. 2½. The desirable qualities of statistical data are as follows. Height. Houses cannot be compared unless they are of a similar size or value.174 Introduction to Statistics and Data Analysis Types of Data There are a number of different ways in which data may be categorised. are those for which observations are numerical in nature (for example a person's height or the number of children in a school). Standardise the data. there are two methods of adjustment possible. Secondary data is any data other than primary data.  Continuous and discrete data Quantitative variables may be either continuous or discrete. For example. shoe sizes are stated in half-units. etc. and so here we have an example of a discrete variable which can take the values 1. determine a relationship between the different units so that all may be expressed in terms of one – in food consumption surveys. Primary data is therefore raw data. 1. If the data is found not to be homogeneous. it includes all published statistics. "Five houses" means little since five slum hovels are very different from five ancestral mansions. Thus it includes any data which has been subject to the processes of classification or tabulation or which has resulted from the application of statistical methods to primary data. However. because it is impossible to have other than a whole number of children. and has to be classified and processed using appropriate statistical methods in order to reach a solution to the problem. Quantitative variables. For most (but not all) discrete variables. Qualitative variables relate to non-numeric observations.

Their instructions must be very carefully prepared to ensure that the results are in accordance with the requirements described in the previous section of this study unit. © ABE and RRC . Care in allocating the duties to the investigators can reduce the risks of bias. Delegated Personal Investigation/Interview When the field to be covered is extensive. Do not put investigator A to explore town X and investigator B to explore town Y. If you do. Questionnaire In some enquiries the data consists of information which must be supplied by a large number of people. it is essential to include a reply-paid envelope to encourage people to respond. asking them to fill in the answers to a set of printed questions. personal biases may tend to cancel out. This method is usually cheaper than delegated personal investigation and can cover a wider field. Additionally. The personal investigation method is also useful if a pilot survey is carried out prior to the main survey. there is sometimes a danger that the investigator may be tempted to select data that accords with preconceived notions. the task of collecting information may be too great for one person. For this reason. as personal investigation will reveal the problems that are likely to occur. in the UK Population Census. any difference that is revealed might be due to the towns being different.  If the forms are distributed and collected by interviewers. you would try to get both people to do part of each town. The result is that we end up with completed forms only from a certain kind of person and thus have a biased sample.) However care must be taken that the interviewers do not lead respondents in any way. (A carefully thought out questionnaire is also often also used in personal or delegated interview in order to reduce the effect of personal bias. (This is the method used. In such a case. The most convenient way to collect such data is to issue questionnaires to the people concerned. or it might be due to different personal biases on the part of the two investigators. If there are many investigators. For example. a greater response is likely and queries can be answered.) The distribution and collection of questionnaires by post suffers from two main drawbacks:  The forms are completed by people who may be unaware of some of the requirements and who may place different interpretations on the questions – even the most carefully worded ones! There may be a large number of forms not returned. This method has the advantage that the data will be collected in a uniform manner and with the subsequent analysis in mind. for example. and these may be mainly by people who are not interested in the subject or who are hostile to the enquiry. Then a team of selected and trained investigators or interviewers may be used. However the field that may be covered by personal investigation may be limited. The people employed should be properly trained and informed of the purposes of the investigation.176 Introduction to Statistics and Data Analysis Personal Investigation/Interview In this method the investigator collects the data personally. suppose you were investigating the public attitude to a new drug in two towns.

type of work. CLASSIFICATION AND TABULATION OF DATA Having completed the survey and collected the data. W Mr Jackson. The age of each employee is stated in years. W Miss Carter.000 employees in a large factory.Introduction to Statistics and Data Analysis 177 C. C Mrs Edwards. technical qualifications and so on. we need to organise it so that we can extract useful information and then present our results. R Mr Gray.e. One way would be to classify the individual cases by sex: © ABE and RRC . If we are required to present to the factory management a statement about the age structure of the workforce (both male and female). What is needed is to classify the cards according to the age and sex of the employee and then present the results of the classification as a tabulation. as follows. Archer. is usually known as raw data (i. For example. In order to do this. J Mr Jones. but they will contain a large amount of other data such as wage rates. so we will look at a shortened version. J Miss Johnson.000 cards. D. L Mr Black. age. and no one could possibly gain any idea about the topic from merely looking through the cards as they are. The data in its original form. W Mrs Brown. Let us assume that our raw data about employees consists of 15 cards in alphabetical order. this data does not allow us to draw any useful conclusions about the group of employees. Example: Of course we cannot give an example involving 3. then the alphabetical arrangement does not help us. L Mr 39 20 22 22 32 30 37 33 45 42 24 27 28 44 39 As it stands. F Miss Gunn. J Mrs Green. S Miss Dawson. we may have a card index of the 3. Very often the data will consist of a mass of figures in no special order. Mr Edwards. it is unprocessed). sex. before classification. T Mr Dee. we need to classify the data in some way. involving only a small number. A Mr Brown. The cards are probably kept in alphabetical order of names.

and we can do this by tabulation. An insurance company. J Mr Jones. W Mrs Brown. but it is not concerned with the age at which John Smith or Mary Brown. R Mr Gray. it is still a collection of data about individuals. D Mr Edwards. splitting them into three groups as follows: Age between 20 and 29 Black. will die. to give another example. A Mr Dawson. T Mr Dee. In comparing the height of Frenchmen with the height of Englishmen. we are concerned with things in groups rather than with individuals. S Miss Dawson. C Mrs Edwards. We need to transform the data into information about the group. after classification. W Miss Carter. W Mrs Brown. J Miss 20 22 32 37 42 24 28 An alternative way of classifying the group would be by age. J Mr 45 42 44 20 22 22 24 27 28 Age between 30 and 39 Archer. W Miss Green. The number of cards in each group. L Mr Carter. F Miss Jackson. R Mr Gunn. S Miss Dee. is counted and the results presented in a table. L Mr 39 32 30 37 33 39 Again. A Mr Brown. T Mr Edwards. but not with Marcel and John as individuals. F Miss Gunn. J Mrs Johnson. © ABE and RRC . we are concerned with Frenchmen in general and Englishmen in general. J Mrs Green. W Mr Johnson. In statistics though. although the data has now acquired some of the characteristics of useful information. C Mrs Gray. as individuals. D Mr Jones. L Mr Brown. W Mr Jackson.178 Introduction to Statistics and Data Analysis Male Archer. J Miss Age over 40 Edwards. L Mr 39 22 30 33 45 27 44 39 Female Black. is interested in the proportion of men or women who die at certain ages.

© ABE and RRC .1: Employees by age and sex Sex Male 2 4 2 8 Female 4 2 1 7 Age group 20 – 29 30 – 39 40 – 49 Total Total 6 6 3 15 You should look through this example again to make quite sure that you understand what has been done.e. where a more precise breakdown of a person’s age is given). may be slightly older than 39 years. You should note that we did not use the age groups 20 – 30.). therefore they should not be included in the age group 40 – 49! Thus. 20 to less than 30). Let us assume that Mr Archer is 39 years. In which age group in Table 6. Forms of Tabulation We can classify the process of tabulation into simple tabulation and complex or matrix tabulation. as this person would fall into both the 30 – 40 and 40 – 50 age groups! So choice of date grouping (often referred to as "class intervals") is an important aspect of the classification and tabulation of data. for example. 30 – 40 and 40 – 50 years. This is because an employee that is. 30 to < 40). but they are not as old as (exactly) 40 years. 30 – 39 and 40 – 49. minutes and seconds. could not be classified into a single group.000 employees mentioned earlier.2. days.it is to condense an unwieldy mass of raw data to manageable proportions and then to present the results in a readily understandable form. Mr Archer and Mr Jones would now clearly fall into the 30 to less than 40 years age group (i. 4 months and 3 days old and Mr Jones is 39 years. then it is possible that Mr Archer and Mr Jones. you are now in a position to appreciate the purpose behind classification and tabulation –. etc. As we learnt in section A of this unit. 30 to < 40 and 40 to < 50. A question as to how the workforce is distributed can be answered by sorting the cards and preparing a simple table as in Table 6. days. The idea is best conveyed by an example. therefore they should not be included in the age group 30 – 39.e. for example. 2 months and 20 days old. age is a continuous variable.Introduction to Statistics and Data Analysis 179 Table 6. exactly 40 years old.1 would Mr Archer and Mr Jones now be classified? Clearly they are both older than (exactly) 39 years. Having understood this. we could use the age groups 20 to < 30 (i. which means it is possible to measure a person’s age broken down into months. Of particular importance to note is the presentation of the original age data. which is given to the nearest year (and not broken down further into months. Consider the card index of 3. We shall look in more detail at the use of class intervals in section D of this unit when we look "frequency distribution". when tabulating continuous variables such as a person’s age. In most cases (i. we need to ensure that the data groups chosen are appropriate. This has allowed the use of the age groups 20 – 29.e. If the age of employees had been presented more precisely in the original data. Assume that each card carries the name of the workshop in which the person works. (a) Simple tabulation This covers only one aspect of a set of figures.

180

Introduction to Statistics and Data Analysis

Table 6.2: Distribution of employees by workshop Workshop A B C D E Total Number employed 600 360 660 840 540 3,000

Another question might have been "What is the wage distribution of the employees?" The answer can be given in another simple table, Table 6.3. Table 6.3: Wage distribution Wages group £40 but less than £60 but less than £80 but less than £100 but less than £120 but less than £140 but less than Total £60 £80 £100 £120 £140 £160 Number of employees 105 510 920 1,015 300 150 3,000

Note that such simple tables do not tell us very much – although it may be enough to answer the initial question at the moment. (b) Complex tabulation This deals with two or more aspects of a problem at the same time. In the problem just studied, it is very likely that the two questions presented in Tables 6.2 and 6.3 would be asked at the same time. We could therefore present the answers in a single, although more complex, table or matrix (as shown in Table 6.4).

ABE and RRC

Introduction to Statistics and Data Analysis

181

Table 6.4: Distribution of employees and wages by workshop Wages group (£ per week) Workshop 40 – 59.99* 20 11 19 34 21 105 60 – 79.99 101 52 103 167 87 510 80 – 99.99 202 90 210 303 115 920 100 – 119.99 219 120 200 317 159 1,015 120 – 139.99 29 29 88 18 136 300 140 – 159.99 29 58 40 1 22 150

Total no. employed 600 360 660 840 540 3,000

A B C D E Total

* Note: 40 – 59.99 is the same as "40 but less than 60" and similarly for the other columns.

This table is much more informative than are the two previous simple tables, although it is more complicated. We could have complicated it further by dividing the groups into male and female employees, or into age groups. Later in this unit, we will look at a list of the rules you should try to follow in compiling statistical tables. At the end of that list you will find a table relating to our 3,000 employees, which you should study as you read the rules.

Secondary Statistical Tabulation
So far, our tables have merely classified the already available figures – the primary statistics. However, we can go further than this and do some simple calculations to produce other figures – secondary statistics. As an example consider again Table 6.2, and this time calculate how many employees there are on average per workshop. This is obtained by dividing the total (3,000) by the number of workshops (5), and the table appears as follows: Table 6.5: Distribution of employees by workshop Workshop A B C D E Total Average number of employees per workshop: The average is a secondary statistic. Number employed 600 360 660 840 540 3,000 600

ABE and RRC

182

Introduction to Statistics and Data Analysis

For another example, we may take Table 6.3 and calculate the proportion of employees in each wage group, as follows: Table 6.6: Wage distribution Wages group £40.00 £60.00 £80.00 £100.00 £120.00 £140.00 – – – – – – £59.99 £79.99 £99.99 £119.99 £139.99 £159.99 Number of employees 105 510 920 1,015 300 150 3,000 Proportion of employees 0.035 0.170 0.307 0.338 0.100 0.050 1.000

Total

The proportions given here are also secondary statistics. In commercial and business statistics it is more usual to use percentages than proportions, and so in Table 6.6 these would be 3.5%, 17%, 30.7%, 33.8%, 10% and 5%. Of course secondary statistics are not confined to simple tables. They are used in complex tables too, as in the example presented in Table 6.7. Table 6.7: Inspection results for a factory product in two successive years Year 1 Machine no. Output 800 600 300 500 2,200 550 No. of rejects 40 30 12 10 92 23 % of rejects 5.0 5.0 4.0 2.0 4.2 4.2 Output 1,000 500 900 400 2,800 700 Year 2 No. of rejects 100 100 45 20 265 66.2 % of rejects 10.0 20.0 5.0 5.0 9.5 9.5

1 2 3 4 Total Average per machine

The percentage columns and the average row show secondary statistics. All the other figures are primary statistics. Note carefully that percentages cannot be added or averaged to get the percentage of a total or of an average. You must work out such percentages on the totals or averages themselves. Another danger in the use of percentages has to be watched: you must not forget the size of the original numbers. Take the case of two doctors treating a particular disease. One doctor has only one patient and the doctor cures the patient – 100% success! The other doctor has 100 patients of whom 80 are cured – only 80% success! You can see how very unfair it would be to make a comparison based on the percentages alone.

ABE and RRC

Introduction to Statistics and Data Analysis

183

Rules for Tabulation
There are no absolute rules for drawing up statistical tables, but there are a few general principles which, if borne in mind, will help you to present your data in the best possible way.           Try not to include too many features in any one table (say, not more than four or five) as otherwise it becomes rather clumsy. It is better to use two or more separate tables. Each table should have a clear and concise title to indicate its purpose. It should be very clear what units are being used in the table (tonnes, £, people, £000, etc.). Appropriate data groups (or class intervals) for continuous and discrete data should be chosen to ensure that all data can be included within a single group (or class). Blank spaces and long numbers should be avoided, the latter by a sensible degree of approximation. Columns should be numbered to facilitate reference. Try to have some order to the table, for example using size, time, geographical location or alphabetical order. Figures to be compared or contrasted should be placed as close together as possible. Percentages should be placed near to the numbers on which they are based. Use a ruler to draw the tables neatly – scribbled tables with freehand lines nearly always result in mistakes and are difficult to follow. However, it is useful to draw a rough sketch first so that you can choose the best layout and decide on the widths of the columns. Insert totals where these are meaningful, but avoid "nonsense totals". Ask yourself what the total will tell you before you decide to include it. An example of such a "nonsense total" is given in Table 6.8. Table 6.8: Election results Party A B Total Year 1 Seats 350 120 470 Year 2 Seats 200 270 470 Total Seats 550 390 940

The totals (470) at the foot of the two columns make sense because they tell us the total number of seats being contested, but the totals in the final column (550, 390, 940) are nonsense totals for they tell us nothing of value.   If numbers need to be totalled, try to place them in a column rather than along a row for easier computation. If you need to emphasise particular numbers, then underlining, significant spacing or bold type can be used. If data is lacking in a particular instance, then insert an asterisk in the empty space and give the reasons for the lack of data in a footnote. Footnotes can also be used to indicate, for example, the source of secondary data, a change in the way the data has been recorded, or any special circumstances which make the data seem odd.

ABE and RRC

184

Introduction to Statistics and Data Analysis

It is not always possible to obey all of these rules on any one occasion, and there may be times when you have a good reason for disregarding some of them. But only do so if the reason is really good – not just to save you the bother of thinking! Now study the layout of the following table (based on our previous example of 3,000 employees) and check through the list of rules to see how they have been applied. Table 6.9: ABC & Co. – Wage structure of labour force Numbers of persons in specified categories Wage group (£ per week) Workshop 40 – 59.99 (2) 20 11 19 34 21 105 3.5 60 – 79.99 (3) 101 52 103 167 87 510 17.0 80 – 99.99 (4) 202 90 210 303 115 920 30.7 100 – 119.99 (5) 219 120 200 317 159 1,015 33.8 120 – 140 139.99 159.99 (6) 29 29 88 18 136 300 10.0 (7) 29 58 40 1 22 150 5.0 Total no. employed (8) 600 360 660 840 540 3,000 100.0 %
See note (a)

(1) A B C D E Total no. in wage group %
See note (b)

(9) 20.0 12.0 22.0 28.0 18.0 100.0 –

Note (a): Total number employed in each workshop as a percentage of the total workforce. Note (b): Total number in each wage group as a percentage of the total workforce.

This table can be called a "twofold" table as the workforce is broken down by wage and workshop.

Questions for Practice 1
These questions are typical of those likely to be set in an examination. 1. The human resources/personnel manager of a firm submits the following report on labour and staff to the managing director: "The total number of employees is 136, of whom 62 are male. Of these men, 43 are industrial labourers and the rest are staff. The age distribution of the 43 is 'under 18' (seven), '18–45' (22) and the others are over 45. Our total industrial labour force is 113, of whom 25 are 'under 18' and 19 are 'over 45'. The total number of 'under-18s' on the site is 29 (10 male, 19 female). There are 23 'over-45s', of whom only 4 are male staff." The managing director, quite rightly, thinks that this is gibberish and instructs you to present it to him in a tabular form which he can readily understand. Show the table you would produce.

ABE and RRC

Introduction to Statistics and Data Analysis

185

2.

Consider the following report. "In the period Year 1 – Year 6 there were 1,775 stoppages of work arising from industrial disputes on construction sites. Of these, 677 arose from pay disputes, 111 from conflict over demarcation, 144 on working conditions and 843 were unofficial walk-outs. 10% of all stoppages lasted only one day. Of these, the corresponding proportions were: 11% for pay disputes, 6% in the case of demarcation and 12½% in the case of working conditions. In the subsequent period Year 7 – Year 12, the number of stoppages was 2,807 higher. But, whereas the number of stoppages arising from pay disputes fell by 351 and those concerning working conditions by 35, demarcation disputes rose by 748. The number of pay disputes resolved within one day decreased by 45 compared with the earlier period, but only by 2 in the case of working conditions. During the later period 52 stoppages over demarcation were settled within the day, but the number of stoppages from all causes which lasted only one day was 64 greater in the later period than in the earlier." You are required to: (a) (b) Tabulate the information in the above passage, making whatever calculations are needed to complete the entries in the table. Write a brief comment on the main features disclosed by the completed table.

Now check your answers with those given at the end of the unit.

D. FREQUENCY DISTRIBUTIONS
A frequency distribution is a tabulation which shows the number of times (i.e. the frequency) each different value occurs. For example, the following figures are the times (in minutes) taken by a shop-floor employee to perform a given repetitive task on 20 specified occasions during the working day: 3.5 3.6 3.4 3.7 3.8 3.8 3.7 3.7 3.8 3.9 3.6 3.7 3.4 3.7 3.8 3.5 3.6 3.5 3.6 3.9

If we now assemble and tabulate these figures, we can obtain a frequency distribution as follows:

ABE and RRC

186 Introduction to Statistics and Data Analysis Table 6.6 3.7 3.6 3.9 Total Tally marks || ||| |||| |||| |||| || Frequency 2 3 4 5 4 2 20 Grouped Frequency Distributions Sometimes the data is so extensive that a simple frequency distribution is too cumbersome and. Then we make use of a grouped frequency distribution. perhaps. First look at the figures to see the highest and lowest values so as to decide the range to be covered and then prepare a blank table.8 3. In this case.11 shows the procedure for the data in Table 6.7 3.4 3.4 3.5 3.10: Time taken for one employee to perform task Length of time (minutes) 3. Table 6.8 3. the "length of time" column consists not of separate values.10 after all 20 items have been entered. but of groups of values: © ABE and RRC . Then mark the items on your table by means of a tally mark. Table 6. uninformative.9 Total Frequency 2 3 4 5 4 2 20 Simple Frequency Distributions A useful way of preparing a frequency distribution from raw data is to go through the records as they stand and mark off the items by the "tally mark" or "five-bar gate" method. To illustrate the procedure.11: Time taken for one employee to perform task Length of time (minutes) 3.5 3.

12: Time taken for one employee to perform task Length of time (minutes) 3. 3. 3. A time such as 3. In the length of time example (see table 6. 26.Introduction to Statistics and Data Analysis 187 Table 6. in practice.12).6 minutes when corrected to one decimal place and it goes in the 3. .7 3.5 3. The number of items in each class (for example. as grouped earlier in this section. "26 to < 31") are called "classes" and the range of values covered by a class (five years in this example) is called the "class interval"..5 we put times which could in fact be anywhere between 3.9 Total Tally marks |||| |||| |||| |||| | Frequency 5 9 6 20 Grouped frequency distributions are only needed when there is a large number of values and. Table 6. In the age group example.13: Age distribution of employees in an office Age group (years) 16 to < 21 21 to < 26 26 to < 31 31 to < 36 36 to < 41 41 to < 46 46 to < 51 51 to < 56 56 to < 60 Total Number of employees 10 17 28 42 38 30 25 20 10 220 The various groups (for example. 220) is called the "total frequency".55.95 minutes..57 minutes would not have been in this group as it equals 3. Table 6. 21. The term "class limits" is used to stand for the lowest and highest values that can actually occur in a class.7 group.4 to 3.6 to 3.35. years.8 to 3.4 to 3.75.55 if we had been able to measure them more precisely. 3.6 to 3. so in the age group example (see Table 6. 28 in the 26 to < 31 class) is called the "class frequency" and the total number of items (in this example. when an ungrouped table would not have been of much use. The term "class boundary" is used to denote the dividing line between adjacent classes. If we had had a more precise stopwatch. the times could have been measured more precisely. This needs some explanation. would not have been required for the small amount of data in our example. we assume that is the precision to which they were measured.35 and 3. In the first group of 3.13 shows a grouped frequency distribution used in a more realistic situation. the class boundaries are 3. these would be 16 years and 20 years 364 days © ABE and RRC .13) the class boundaries are 16. As the original readings were given correct to one decimal place.

however.400 6. As a practical guide.801 – 5.401 – 7. of persons 50. 800.000 6.188 Introduction to Statistics and Data Analysis for the first class.001 – 5. you will find that somewhere between about five and 10 groups will usually be suitable. A superficial look at the original table (first two columns only) might have suggested that the most frequent incomes were at the middle of the scale.001 – 6. 200.14: Annual income statistics Annual income to the nearest £ 4.6 and 3. 400. In the length of time example.000 12.000 23. Sometimes.500 6. make the length of the class interval equal for all classes.000 Frequency per £200 interval 50.000 40. Choice of Class Interval When compiling a frequency distribution you should. 21 years and 25 years 364 days for the second class and so on. and column three shows that. But this apparent preponderance of the middle class is due solely to the change in the length of the class interval. assuming that the ages were measured correct to the nearest day.000 192.201 – 5. this rule has to be broken – for example.000 5. i.4 and 3. These intervals let you see how the last column was compiled. if possible. before using the information. because of the appearance of the figure 55. some grouped frequency distributions bring together the first few classes and the last few classes because there are few instances in each of the individual classes. the most frequent incomes are at the bottom end of the scale.200 5. Table 6. nor do we want too few groups or we will fail to see the significant features of the distribution. You should remember that the purpose of compiling a grouped frequency distribution is to make sense of an otherwise troublesome mass of figures.5 minutes for the first class and 3. as in the example for wage statistics in Table 6. This is so that fair comparisons can be made between one class and another. the top of the table.e.000 12. the class limits are 3. 400.000 40. 400.14. It therefore follows that we do not want to have too many groups or we will be little better off.200 Total No.000 3.500 11. In such cases. it is as well to make the classes comparable by calculating a column showing "frequency per interval of so much".000 – Notice that the intervals in the first column are: 200. in fact. © ABE and RRC .7 minutes for the second class.000.601 – 6.000 55.600 5. You should make yourself quite familiar with these terms.000 27.

However. but in the number of items above or below a certain value.Introduction to Statistics and Data Analysis 189 Cumulative Frequency Distributions Very often we are not especially interested in the separate class frequencies. the distribution used in Table 6. When this is the case. for example. and as an exercise you should now compile the "more than" cumulative frequency column for Table 6.15 gives us the number of items "less than" a certain amount.8 3.9 Frequency 2 3 4 5 4 2 Cumulative frequency 2 5 9 14 18 20 The cumulative frequency tells us the number of items equal to or less than the specified value. the result is known as a "relative frequency distribution" and the separate proportions or percentages are the "relative frequencies". The example in Table 6. Table 6.15: Time taken for one employee to perform task Length of time (minutes) 3.5 3. the number of persons having more than some quantity.7 3. always 1. This can easily be done by doing the cumulative additions from the bottom of the table instead of the top. and it is formed by the successive addition of the separate frequencies. The total relative frequency is.4 3.15. When the frequency in each class of a frequency distribution is given as a proportion or percentage of the total frequency. However we may wish to know. Cumulative relative frequency distributions may be compiled in the same way as ordinary cumulative frequency distributions. © ABE and RRC .6 3. As an example. we form a cumulative frequency distribution as illustrated in column three of Table 6. remember that proportions or percentages are useful secondary statistics.15. Relative Frequency Distributions All the frequency distributions which we have looked at so far in this study unit have had their class frequencies expressed simply as numbers of items.14 is set out as a relative frequency distribution in Table 6. A cumulative frequency column may also be formed for a grouped distribution.0 (or 100%).16 for you to study. of course.

000 180.000 – Relative cumulative frequency (%) 26 47 76 88 94 100 – Frequency 50.000 5.000 192. Now check your answers with those given at the end of the unit. the cumulative frequency and the relative cumulative frequency.200 5.000 23.000 12. © ABE and RRC .801 – 5.000 This example is in the "less than" form.000 192.000 40.001 – 5.000 145.201 – 5.001 – 6.000 12. Questions for Practice 2 The following figures show the number of faulty items made on successive days by a machine in an engineering workshop: 2 1 4 2 3 1 2 4 1 5 0 4 5 4 1 1 6 4 1 6 2 1 5 0 5 10 5 1 2 1 4 2 6 6 3 1 3 3 6 3 3 1 5 1 2 2 1 7 1 2 2 4 2 0 2 3 2 4 0 2 11 1 4 1 1 4 5 1 8 1 4 7 2 5 3 3 7 7 5 2 0 3 1 1 5 10 2 3 8 6 1 2 2 2 3 6 3 7 1 4 Tabulate these figures as a frequency distribution.400 6.16: Annual income statistics Annual income to the nearest £ 4.200 Total Relative frequency (%) 26 21 29 12 6 26 100 Cumulative frequency 50.601 – 6. and you should now compile the "more than" form in the same way as you did for the non-relative distribution.190 Introduction to Statistics and Data Analysis Table 6.000 6.000 90.600 5. giving columns containing the frequency.000 55. the relative frequency.000 168.401 – 7.

143.53 104. and 4. Obviously. 1. © ABE and RRC .5 Total 326 859 109 3. 241. and you should round off each figure to the nearest whole number.63 1.37 177.Introduction to Statistics and Data Analysis 191 ANSWERS TO QUESTIONS FOR PRACTICE Questions for Practice 1 1.53 807 93 3. This was particularly remarkable in the later period.340. XYZ Company: Age of Staff and Labour Industrial Male Female Total (2) 7 22 14 43 (3) 18 47 5 70 (4) 25 69 19 113 Staff Male Female Total (5) 3 12 4 19 (6) 1 3 0 4 (7) 4 15 4 23 Total Male Female Total (8) 10 34 18 62 (9) 19 50 5 74 (10) 29 84 23 136 Age group (1) Under 18 18 – 45 Over 45 TOTAL 2. with those concerning working conditions being less likely to last more than one day. In both periods.47 6.598.5 Total 677 111 144 843 1.34 126 764.97 4. fractional numbers of stoppages cannot occur.47 52 16 144. when they amounted to more than twice the number of all other stoppages put together. although stoppages over pay and working conditions both declined.5 More than 1 day 296.341.288 4.775 Year 7 – Year 12 1 day 29.5 More than 1 day 602. compared with 90% in the previous period.582 Pay Demarcation Working conditions Unofficial* All * Note that we have given the figures as worked out from the information supplied. (a) The complete table is as follows: Number of Stoppages of Work Arising from Industrial Disputeson Construction Sites Year 1 – Year 6 Cause 1 day 74. Demarcation disputes rose in number dramatically. (b) The overall number of stoppages in the later period was more than 2½ times greater than in the earlier period. the number of unofficial stoppages was the highest category.66 18 78. though.03 241. The subtotals of the columns will then be 177.597. About 95% of stoppages in the later period lasted more than one day.

00 0.192 Introduction to Statistics and Data Analysis Questions for Practice 2 Faulty items made on successive days by a machine in an engineering workshop Number of faulty items 0 1 2 3 4 5 6 7 8 9 10 11 Total Number of days with these faulty items 5 23 20 13 12 10 7 5 2 0 2 1 100 Relative frequency 0.00 – © ABE and RRC .05 0.73 0.07 0.28 0.83 0.05 0.61 0.99 1.13 0.10 0.97 0.23 0.12 0.97 0.00 Cumulative frequency 5 28 48 61 73 83 90 95 97 97 99 100 – Relative cumulative frequency 0.90 0.02 0.95 0.05 0.20 0.01 1.48 0.02 0.

F. D. E. Answers to Questions for Practice © ABE and RRC . C.193 Study Unit 7 Graphical Representation of Information Contents Introduction A. Graphical Representation of Frequency Distributions Frequency Dot Diagram Frequency Bar Chart Frequency Polygon Histogram The Ogive Stem and Leaf Diagram Pictograms Limited Form Accurate Form Pie Charts Bar Charts Component Bar Chart Horizontal Bar Chart Scatter Diagrams General Rules for Graphical Presentation Page 194 194 195 195 196 197 200 203 204 204 205 206 207 207 208 209 210 215 B.

pie charts. ogives and stem and leaf diagrams to represent frequency distributions prepare pictograms. Diagrams are particularly effective as a way of conveying quite complex information. histograms. In the first part of this unit we shall examine the specific graphical forms used for presenting frequency distributions of the type considered in the previous study unit.10). The following is a diagram that to some extent overcomes this limitation:  stem and leaf diagram. presentations that are not necessarily used by statisticians (e. and discuss the circumstances in which each form of presentation is most effective describe the general rules and principles for using graphical representations of data. Objectives When you have completed this study unit you will be able to:    use frequency dot diagrams. The methods of presenting frequency distributions graphically are as follows:      frequency dot diagram frequency bar chart frequency polygon histogram ogive. However. and enables broad distributions and trends to be taken in quickly. We then move on to discuss a number of popular graphical presentations that are used in everyday life. initially using the data on the times (in minutes) taken by an employee to perform a given task on 20 specified occasions during the working day which we previously considered in Study Unit 6 (Table 6. and details are likely to be lost. such graphs and charts cannot usually be used for accurate analysis of data.g. We will now examine each of these graphical methods in turn. bar charts and scatter diagrams from given data. Presenting information visually is easy to understand. Graphs and charts are normally much superior to tables (especially lengthy complex tables) for showing general states and trends. diagrams are rarely as accurate as the figures themselves. frequency polygons. The tabulated frequency distribution for this data is as follows: © ABE and RRC . frequency bar charts. those used in business journals and the mass media of newspapers and television). A.194 Graphical Representation of Information INTRODUCTION In this study unit we shall move on and look at further methods of summarising and presenting data as a means of providing useful information. However. GRAPHICAL REPRESENTATION OF FREQUENCY DISTRIBUTIONS Tabulated frequency distributions are sometimes more readily understood if represented by a diagram.

Loosely.1 would be as shown in Figure 7.) A horizontal scale is used for the variate and a vertical scale for the frequency. Above each value on the variate scale we mark a dot for each occasion on which that value occurs.6 3.1: Time taken for one employee to perform task Length of time (minutes) 3. (In statistics. roughly speaking.Graphical Representation of Information 195 Table 7. Thus.4 3.7 3.9 Total Frequency 2 3 4 5 4 2 20 Frequency Dot Diagram This is a simple form of graphical representation for the frequency distribution of a discrete variate. they can be used interchangeably. Whilst the two terms do not mean exactly the same thing.8 3.4 3.5 3.5 3. a frequency dot diagram of the information in Table 7.8 3. the term "variate" is used when you might have expected the term "variable". a variate is a quantity that can have any of a set of numerical values.7 3. Figure 7.6 3. at this level.1.1: Frequency dot diagram to show length of time taken by operator to complete a given task Frequency 3.9 Length of time (minutes) © ABE and RRC . Do not worry too much about the difference between these two terms.

the length of which represents the number of dots which should be there.3.4 3.3 3.2. the result is a frequency polygon.5 3. i. we could merely mark the position of the top end of each bar and then join up these points with straight lines. Figure 7.7 3.9 Length of time (minutes) Frequency Polygon Instead of drawing vertical bars as we do for a frequency bar chart. © ABE and RRC .0.8 3. as in Figure 7.4 3. as in Figure 7.9 4 Length of time (minutes) Note that we have added two fictitious classes at each end of the distribution.1 now becomes a frequency bar chart. Figure 7.6 3. When we do this.e.5 3.7 3. This is done to ensure that the area enclosed by the polygon and the horizontal axis is the same as the area under the corresponding histogram which we shall consider in the next section.3: Frequency polygon to show length of time taken by operator to complete a given task Frequency 6 5 4 3 2 1 0 3.6 3.3 and 4. we have marked in groups with zero frequency at 3.8 3.2: Frequency bar chart to show length of time taken by operator to complete a given task Frequency 6 5 4 3 2 1 0 3. The frequency dot diagram in Figure 7.196 Graphical Representation of Information Frequency Bar Chart We can avoid the business of marking every dot in such a diagram by drawing instead a vertical line.

we shall use an example of a grouped frequency distribution similar to one from Study Unit 6 (Table 6. They are mostly used in those cases where the variate is discrete and where the values are not grouped.2: Age distribution of employees in an office Age group (years) 15 to < 20 20 to < 25 25 to < 30 30 to < 35 35 to < 40 40 to < 45 45 to < 50 50 to < 55 55 to < 60 Total Number of employees 10 17 28 42 38 30 25 20 10 220 Figure 7. The information in the table may be shown graphically by the histogram in Figure 7.Graphical Representation of Information 197 These three kinds of diagram are all commonly used as a means of making frequency distributions more readily comprehensible. Sometimes. by drawing the vertical line (or marking its top end) at the centre point of the group. To illustrate this.13). frequency bar charts and polygons are used with grouped data.2.4: Histogram showing age distribution of employees in an office Frequency 40 30 20 10 0 15 20 25 30 35 40 45 50 55 60 Age (years) © ABE and RRC . Table 7.4. as set out in Table 7. Histogram A histogram is the best way of showing a grouped frequency distribution.

Note particularly how it illustrates the fact that the frequency falls off towards the higher age groups – any form of graph which did not reveal this fact would be misleading. When the lengths of the class intervals are all equal.) Look once again at the histogram of ages given in Figure 7. for the amended data would look like Figure 7. had given the last few groups as those in Table 7. then the heights of the rectangles represent the frequencies in the same way as do the areas (this is why the vertical scale has been marked in this diagram). Figure 7. for some reason or other. not its height. (Do not stop at this point if you have not quite grasped the idea. If however the lengths of the class intervals are not all equal. using heights instead of areas. Now let us imagine that the original table had not used equal class intervals but. A wrong form of histogram.3: Age distribution of employees in an office (amended extract) Age group (years) 40 to < 45 45 to < 50 50 to < 60 Number of employees 30 25 30 The last two groups have been lumped together as one. the frequency in each group is represented by a rectangle and – this is a very important point – it is the area of the rectangle.5: Histogram showing age distribution of employees in an office (amended data) Frequency 40 30 This column is the wrong size 20 10 0 15 20 25 30 35 40 45 50 60 Age (years) © ABE and RRC .3. which represents the frequency.5. then you must remember that the heights of the rectangles have to be adjusted to give the correct areas. Table 7.4.198 Graphical Representation of Information In a histogram. because it will become clearer as you read on.

that the last group in a frequency table is not completely specified. then nothing much is lost by leaving it off the histogram altogether. then you should try to judge from the general shape of the histogram how many class intervals to spread the last frequency over. For example. the height of the last group (50 to < 60) would be halved because the class interval is double all the other class intervals. In © ABE and RRC . The histogram in Figure 7.Graphical Representation of Information 199 Now. in order not to create a false impression of the extent of the distribution.6: Amended histogram showing age distribution of employees in an office (amended data) Frequency density per 5-year 40 interval 30 20 10 0 15 20 25 30 35 40 45 50 60 Age (years) Often it happens. less than about 1% or 2%). If the last group has a larger frequency than about 1% or 2%. The vertical axis has been labelled "Frequency density per 5-year interval" as five years is the "standard" interval on which we have based the heights of our rectangles. How would we draw this on a histogram? If the last (open-ended) group has a very small frequency compared with the total frequency (say. the last few groups in our age distribution table may have been specified as follows: Table 7. in published statistics. In the correct form of the histogram.4: Age distribution of employees in an office (further amended extract) Age group (years) 40 to < 45 45 to < 50 50 and over Number of employees 30 25 30 A distribution of this kind is called an "open-ended" distribution.6 gives the right impression of the falling off of frequency in the higher age groups. this clearly gives an entirely wrong impression of the distribution with respect to the higher age groups. Figure 7.

8 3.9 Frequency 2 3 4 5 4 2 Cumulative frequency 2 5 9 14 18 20 Table 7.5: Time taken for one employee to perform task Length of time (minutes) 3.000 Cumulative frequency 50.8 is for grouped data (as per Table 7.000 12.8 and see how the 168.600 5. the plotting would have been against the lower end of the group (or class interval).000 – © ABE and RRC .7 3.200 5.000 23.5 3.201 – 5.750.400 6. you would probably spread the last 30 people over two or three class intervals.200 Total Frequency 50.200 Graphical Representation of Information the example given.6 3.001 – 6. Study these two diagrams so that you are quite sure that you know how to draw them. but the "less than" form is the usual one. It can be drawn in either the "less than" or the "or more" form. Ogives for two of the distributions already considered are now given as examples: Figure 7. Whatever procedure is adopted. If we had been plotting a "or more" ogive.000 168. but it is often simpler to assume that an open-ended class has the same length as its neighbour. Look at the example Table 7.000 5.4 3. There is only one point which you might be tempted to overlook in the case of the grouped distribution – the points are plotted at the ends of the class intervals and not at the centre point. 5.001 – 5.6 and Figure 7.000 180.000 192.6).000 12.000 90.000 192.000 40.401 – 7.601 – 6. The Ogive This is the name given to the graph of the cumulative frequency.601 – 6.801 – 5.000 group (or class interval) and not against the midpoint.000 55.7 is for ungrouped data (as per Table 7. Table 7.000 point is plotted against the upper end of the 5.6: Annual income statistics Annual income to nearest £ 4. it is essential to state clearly what you have done and why.000 6.5) and Figure 7.000 145.

7: Ogive to show length of time taken by operator to complete a given task Cumulative 20 frequency (less than or equal to) 15 10 5 0 3.6 3.8 4 Length of time (mins) Figure 7.2 5.8 7.8 5.2 3.Graphical Representation of Information 201 Figure 7.2 Annual income (£000s) © ABE and RRC .6 6.4 6.8: Ogive to show annual income distribution Less than cumulative frequency (000s) 200 150 100 50 0 4.0 6.4 3.

4 6. Table 7.200 5.000 12. In each of Figures 7.601 – 6. we can compile the data from our example of annual income statistics as a "more than" cumulative frequency.001 – 5.801 – 5.7.000 23.201 – 5.000 47.000 24.0 6.9: Ogive to show annual income distribution More than cumulative frequency (000s) 200 150 100 50 0 4.000 6.000 12.200 Total Cumulative frequency (more than) 192.202 Graphical Representation of Information As an example of an "or more" ogive.000 142.000 55.9.000 192.401 – 7.8 7.000 12. © ABE and RRC .8 5.000 102.2 Annual income (£000s) Check that you see how the plotting has been made against the lower end of the group and notice how the ogive has a reversed shape.000 5.8 and 7.2 5.000 – Frequency 50.001 – 6.9 we have added a fictitious group of zero frequency at one end of the distribution.7: Annual income statistics Annual income to nearest £ 4. Figure 7. shown in Table 7.000 40.600 5.6 6.000 The ogive for this now appears as shown in Figure 7.400 6.

list the stems in numerical order (in our example. then listing the second digit (i. 57 and 59. the stems range from 5 to 8 as the lowest value is 50 and the highest value is 87) in a column. The leaf is usually the last digit of the number and the other digits to the left of the leaf form the stem. for the number 79. It is probably easiest to understand the stem and leaf diagram through an example. group the numbers with the same stems (e. Next. there is no loss of information). but is not a required step as ordering can be done later.) Finally. stem and leaf diagrams can also allow further analysis of the data as it is possible to obtain the original data points from the diagram (i. Then. As you will see. Similar to the histogram. each data value is split into a stem and a leaf. the stem is "7" and the leaf is "9"). order them now. you might want to arrange the data in numerical order. the "leaves") in a second column to the right of the appropriate "stem" (and again sorted in numerical order). In a stem and leaf diagram.Graphical Representation of Information 203 It is common practice to call the cumulative frequency graph a "cumulative frequency polygon" if the points are joined by straight lines. and a "cumulative frequency curve" if the points are joined by a smooth curve. But what is necessary is that you separate each number into a stem and a leaf.e. In more advanced statistical work the latter are used almost exclusively. always compile a "less than" cumulative frequency. Stem and Leaf Diagram A stem and leaf diagram is the final method of presenting frequency distributions graphically that we will study in this section. so let us construct a stem and leaf diagram using the following data of a student's exam marks: 79 61 69 65 83 80 57 64 73 79 82 50 67 76 77 68 62 70 87 71 65 61 62 80 80 59 70 65 82 77 Before you begin. frequency bar chart. If you have successfully followed these steps. it provides extra detail regarding individual values. and you should practise using relative frequencies whenever possible. the tens digit is the stem and the units digit is the leaf (e. 7 and 9. it provides a partial sorting of the original data to allow a visual identification of the distribution pattern of the data. histogram and ogive that we have already studied in that it has the advantage of preserving the original data. then you should have finished up with a stem and leaf diagram identical to the one shown in Figure 7.g.) Of course all of these diagrams may be drawn from the original figures or on the basis of relative frequencies. (NB. Unless you are told otherwise. in addition to providing a visual representation of the data. the "stems") in a column (and sorted in numerical order). which correspond to the original values of 50. It differs from the frequency dot diagram.e. for the stem value of 5 we have leaf values of 0. In a second column list the leaves to the right of the appropriate stem values.g. Thus. Since these are two digit numbers.10. The stem and leaf diagram is plotted by listing the first digits (i. frequency polygon.e. If your leaf values are not in increasing order. © ABE and RRC . 79 and 77 are in the same group as both have a stem of 7). Unlike the histogram. Like the histogram. a stem and leaf diagram is a way of grouping your original data into classes. add a title to your diagram.

There are two types and these are illustrated in the following examples. Their use is confined to the simplified presentation of statistical data for the general public.11: Number of cruises Years 1-3 (Source: Table 7. Figure 7. Pictograms consist of simple pictures which represent quantities.204 Graphical Representation of Information Figure 7.8) Year 1 200 cruises Year 2 250 cruises Year 3 300 cruises © ABE and RRC .000 180. as in Figure 7.000 Limited Form We could represent the number of cruises by showing pictures of ships of varying size. The data we will use is shown in Table 7. Table 7. of passengers carried 100.11. "ideograms".8: Cruises organised by a shipping line between Year 1 and Year 3 Year 1 2 3 Number of cruises 200 250 300 No.10: Stem and leaf diagram to show the distribution of exam marks Stem 5 6 7 8 0 7 9 1 1 2 2 4 5 5 5 7 8 9 0 0 1 3 6 7 7 9 9 0 0 0 2 2 3 7 Key: 50 means 50 marks Leaf B. PICTOGRAMS Pictograms are the simplest method of presenting information visually.8. or "picturegrams" – the words all refer to the same thing. These diagrams are variously called "pictograms".000 140.

It is difficult to judge what increase has taken place. These diagrams have no purpose other than generally presenting statistics in a simple way. The reader can become confused as to whether the quantity is represented by the length or height of the pictograms.000 passengers Each matchstick man is the same height and represents 20. their area on the paper. Figure 7.000 passengers.13 – it is difficult to represent a quantity of less than 10m barrels – so we are left wondering what "[" means. we could represent the numbers of passengers carried by using pictures of people as in Figure 7.13: Imports of crude oil Year 1: Year 2: Key: represents 10m barrels of oil © ABE and RRC . This is usually done by representing specific quantities of the variable with pictures or graphics of some sort – usually related to the subject of the variable. So. to avoid confusion it is better to try and give it a degree of accuracy in representing the numerical data. (Sometimes you will even find pictograms in which the sizes shown are actually wrong in relation to the real increases.12: Number of passengers carried Years 1-3 (Source: Table 7.Graphical Representation of Information 205 Although these diagrams show that the number of cruises has increased each year.12. or the volume of the object they represent.8) Year 1: Year 2: Year 3 Key:  20. They are not capable of providing real accurate detail. Figure 7. so there can be no confusion over size. Consider Figure 7.) Accurate Form If the purpose of the pictogram is to covey anything but the most minimal of information. they can give false impressions of the actual increases.

6 100 Construct the diagram by means of a pair of compasses and a protractor. 51% of 360  (c) 51  360  183. © ABE and RRC . Table 7.9) Industrial 843m therms Commercial 437m therms Domestic 1. They are popular in computer graphics. Convert the percentages into degrees. Like pictograms.383m therms Public 60m therms There are a number of simple steps to creating a pie chart.383 843 437 60 2.9. The data is taken from the Annual Abstract of Statistics as shown in Table 7.206 Graphical Representation of Information C.14: Example of a pie chart – gas sales in Great Britain (Source: Table 7.14. are used to show the manner in which various components add up to a total.723 % 51 31 16 2 100 * Central and local government uses. they are only used to display very simple information to non-expert readers. (a) (b) Tabulate the data and calculate the percentages. Don't overlook this point. Figure 7. known also as circular diagrams.g. because inaccurate and roughly drawn diagrams do not convey information effectively.9: Gas sales in Great Britain in one year (Source: Annual Abstract of Statistics) Uses Domestic Industrial Commercial Public* Total Million therms 1. e. including public lighting The data is illustrated as a pie chart or circular diagram in Figure 7. PIE CHARTS Pie charts. Suppose that we wish to illustrate the sales of gas in Great Britain in a certain year.

D. These bar charts are preferable to circular diagrams because:    they are easily read. for that reason. and that the components are drawn in the same order so as to facilitate comparison. is sometimes called a "component bar diagram" (see Figure 7. Component Bar Chart This first type of bar chart serves the same purpose as a circular diagram and. but only general ideas which are best studied by means of examples. and we will now illustrate a number of the types of bar chart in common use. You can supplement the examples in this study unit by looking at the commercial pages of newspapers and magazines. even when there are many components they are more easily drawn it is easier to compare several bars side by side than several circles. Ltd 16 14 12 10 8 6 4 2 0 Year 1 Year 2 Key: Wages Royalties Materials Note that the lengths of the components represent the amounts. BAR CHARTS We have already met one kind of bar chart in the course of our studies of frequency distributions. However the idea of a bar chart can be extended beyond the field of frequency distributions.15).15: Component bar chart showing cost of production of ZYX Co. Unless the whole diagram in each case represents exactly the same total amount. (e) The main use of a pie chart is to show the relationship each component part bears to the whole. We say "illustrate" because there are no rigid and fixed types. but this is not really to be recommended. namely the frequency bar chart. the frequencies of different values of the variate. don't use a diagram of this kind with more than four or five component parts.) You can insert the exact values of the data for each segment if more detail is required. since it is not possible to read this exactly from the diagram itself. by their length. (A key is illustrated in Figure 7. Figure 7. © ABE and RRC . If you have the choice. other diagrams such as bar charts (see next section) are much clearer. using a separate legend or key if necessary.Graphical Representation of Information 207 (d) Label the diagram clearly. Pie charts are sometimes used side by side to provide comparisons. The bars there were simply thick lines representing.15 below.

to distinguish them from those in which the bars are horizontal (see Figure 7.208 Graphical Representation of Information Bar charts with vertical bars are sometimes called "column charts".16). 50% might have been 1 million. i. Note how a loss is shown by drawing the bar on the other side of the zero line. as they each represent 100%. time.17. © ABE and RRC .17 is also an example of a multiple or compound bar chart as there is more than one bar for each category. Figure 7. Figure 7.16: Horizontal bar chart of visitors arriving in the UK in one year Others 4% European 51% Commonwealth 23% USA 22% This figure is also an example of a percentage component bar chart. you must be careful.e. Figure 7.e. whereas in another year it may have been nearer to 4 million and in another to 8 million.17 shows a categorical variable – i.15 we have a numerical variable. but they will not necessarily represent the same actual quantities – for example. If you compare several percentage component bar charts. Horizontal Bar Chart A typical case of presentation by a horizontal bar chart is shown in Figure 7. The example in Figure 7. the different branches form the different categories. whereas in Figure 7. namely. Each bar chart will be the same length. the information is expressed in percentages rather than in actual numbers of visitors.17: Horizontal bar chart for the So & So Company Ltd to show profits made by branches in Year 1 and Year 2 London A London B Key: Newcastle Year 1 Year 1 Bristol 5 0 5 10 15 20 25 Loss (£000s) Profit (£000s) Pie charts and bar charts are especially useful for "categorical" variables as well as for numerical variables.

For example. However. These three types of correlation are shown in Figure 7. This association is known as correlation.000). where both variables are increasing together. The scatter diagram presented in Figure 7. © ABE and RRC .18 confirms that there appears to be a positive relationship. If the scatter diagram does not show any relationship between the variables then no correlation exists. For example. they may indicate the existence and strength of an association between two variables. if one variable increases while the other decreases. its sales revenue increases (although sales revenue does seem to flatten out after advertising reaches £350. In contrast. we would expect the efficiency of the machines in this manufacturing company to decline as they get older. we would not expect any correlation between the cost of raw materials and employee output. the relationship shown in Figure 7. the relationship between the level of advertising expenditure and sales revenue for this manufacturing company becomes immediately clear. is described as positive correlation (both variables are positively correlated with each other). if we plot this data on a scatter diagram. Figure 7.210 Graphical Representation of Information However.18 Scatter diagram of advertising expenditure against sales revenue for a manufacturing company Sales revenue (£000) 250 200 150 100 50 0 0 10 20 30 40 50 60 Advertising expenditure (£000) While scatter diagrams are used to study possible relationships between two variables. Thus. suggesting that the efficiency and age of machinery are negatively correlated.19. it is important to stress that these diagrams cannot prove that one variable causes the other.18. the relationship is described as negative correlation. As the company spends more on advertising.

19: Scatter diagram examples showing different types of correlation Positive correlation Annual Income (£000s) 50 40 30 20 10 0 0 1 2 3 4                Years of post school education Negative correlation Machinery 250 efficiency (output per 200 day) 150 100 50 0 0 10 20 30 40                Age of machinery (months) No correlation Cost of raw 100 materials (£ per tonne) 80 60 40  20 0 0 20 40 60 80               Employee output (items per day) © ABE and RRC .Graphical Representation of Information 211 Figure 7.

Figure 7. Wherever necessary. guidelines should be inserted to facilitate reading.20. Diagrams should be accompanied by the original data. Avoid excessive detail.20: Sales of LMN Factory Sales (£000s) 70  60  50           J F M A M J J A S O N D Months © ABE and RRC . but take care not to distort the graph by overemphasising small variations. Try to include the origins of scales. and the scales of values clearly marked. or at least by a reference to the source of the data.212 Graphical Representation of Information F. Obeying this rule sometimes leads to rather a waste of paper space.       Graphs and charts must be given clear but brief titles. In such a case the graph could be "broken" as shown in Figure 7. GENERAL RULES FOR GRAPHICAL PRESENTATION There are a number of general rules which must be borne in mind when planning and using the graphical methods covered in this unit. as this defeats the object of diagrams. The axes of graphs must be clearly labelled.

01 to 36.01 to 56.01 to 66. of employees 6 8 12 18 25 30 24 14 Construct a stem and leaf diagram to illustrate the age distribution of customers purchasing clothing from a retail store over a one-hour period given the following data on customer ages: 8 38 58 13 40 61 16 41 63 25 44 67 26 47 75 29 49 78 30 51 82 32 54 86 37 55 95 © ABE and RRC . Illustrate the distribution by means of a frequency bar chart of the relative figures.01 to 51. giving columns containing the frequency. No.01 to 71. 1.00 61.00 3. The following figures show the number of faulty items made on successive days by a machine in an engineering workshop: 2 1 4 2 3 1 2 4 1 5 0 4 5 4 1 1 6 4 1 6 2 1 5 0 5 10 5 1 2 1 4 2 6 6 3 1 3 3 6 3 3 1 5 1 2 2 1 7 1 2 2 4 2 0 2 3 2 4 0 2 11 1 4 1 1 4 5 1 8 1 4 7 2 5 3 3 7 7 5 2 0 3 1 1 5 10 2 3 8 6 1 2 2 2 3 6 3 7 1 4 Tabulate these figures as a frequency distribution.00 56.00 51. Illustrate the following data by means of a histogram.01 to 41.00 46.01 to 61.00 41.00 66.01 to 46. the relative frequency. the cumulative frequency and the relative cumulative frequency.00 36. 2. a frequency polygon and an ogive: Weekly wage (£) 31.Graphical Representation of Information 213 Questions for Practice These questions are typical of those likely to be set in an examination.

16 Number of items sold (000) 41 38 34 29 26 30 25 22 19 15 Draw a scatter diagram of the data and comment on the relationship between price and the amount sold.250 5.400 6.27 6.214 Graphical Representation of Information 4.850 5.30 4.18 10.863.800 5.26 8.500 1.39 3.500 99.000 5.400 164.450 Sources of revenue in Ruritania for year ending 31 March 5.600 5.20 2.500 6.762. Home sales Commonwealth sales Foreign sales 6. Construct a circular diagram to illustrate the following figures: Source Income Tax Estate Duties Customs & Excise Miscellaneous Amount (£000) 1.500 The unit price of a commodity and its demand over a period of time are shown in the following table: Price (£ per item) 1. Illustrate by means of a bar chart the following data relating to the number of units sold of a particular kind of machine: Year 1 2 3 6.29 7. Now check your answers with those given at the end of the unit.24 9.700 5. © ABE and RRC .31 5.

61 0.00 Cumulative frequency 5 28 48 61 73 83 90 95 97 97 99 100 – Relative cumulative frequency 0.48 0.05 0 0 1 2 3 4 5 6 7 8 9 10 11 No.1 0. of faulty items per day © ABE and RRC .23 0.00 0.99 1.05 0.05 0. Number of faulty items 0 1 2 3 4 5 6 7 8 9 10 11 Total Number of days with these items 5 23 20 13 12 10 7 5 2 0 2 1 100 Relative frequency 0.2 0.13 0.Graphical Representation of Information 215 ANSWERS TO QUESTIONS FOR PRACTICE 1.12 0.20 0.95 0.97 0.07 0.97 0.00 – Frequency bar chart – faulty items produced by machine in workshop 0.28 0.83 0.01 1.90 0.02 0.10 0.15 0.25 Relative frequency 0.02 0.05 0.73 0.

of employees Combined histogram and frequency polygon    20    10   0  31 36 41 46 51 56 61 66 71  Weekly wage (£) Workings for ogive: Wage Group (£) 31.01 – 71.01 – 41.01 – 46.00 51.00 66.01 – 66.01 – 56.00 The ogive follows: Frequency 6 8 12 18 25 30 24 14 Cumulative frequency 6 14 26 44 69 99 123 137 © ABE and RRC .00 56.216 Graphical Representation of Information 2.00 61.00 46.01 – 61.00 36. 30 No.01 – 36.01 – 51.00 41.

Stem and leaf diagram to show the age distribution of retail customers for a clothing store over a one-hour period Stem 0 1 2 3 4 5 6 7 8 9 8 3 6 5 6 9 0 2 7 8 0 1 4 7 9 1 4 5 8 1 3 7 5 8 2 6 5 Leaf Key: 08 means 8 years old © ABE and RRC .Graphical Representation of Information 217 Ogive for wages distribution Cumulative frequency 140 120 100 80 60 40 20 0 31 36 41 46 51 56 61 66 71 Weekly wage (£) 3.

000 14. Sources of revenue in Ruritania Customs & excise Estate duties Miscellaneous Income tax 5.0% 172 15 163 9 359 Note: The angles do not add up to 360 exactly because of rounding errors.000 12.000 2. The amounts must first be turned into percentages and degrees.2% 45.000 8.6% 100. Bar chart of machine sales 18.000 4. You can.000 0 Year 1 Year 2 Year3 Key: Foreign sales Commonwealth sales Home sales © ABE and RRC . Income tax Estate duties Customs & excise Miscellaneous Total 47.9% 4.3% 2.218 Graphical Representation of Information 4.000 16. draw bar charts on a percentage basis. of course. Shading has been added for clarity. The bar chart is shown in below.000 10.000 6.

Graphical Representation of Information 219 6. In general. the scatter diagram shows that as price increases demand for the product (measured in terms of number of items sold) falls. This suggests that there is a negative correlation between the price of this particular item and product demand. The scatter diagram showing the relationship between product price and the number of items sold for a company follows. Sales (000s) 45 40 35 30 25 20 15 10 5 0 0 2 4 6 8 10 12 Price (£ per item) © ABE and RRC .

220 Graphical Representation of Information © ABE and RRC .

221 Study Unit 8 Measures of Location Contents Introduction A. C. Answers to Questions for Practice © ABE and RRC . The Arithmetic Mean The Mean of a Simple Frequency Distribution The Mean of a Grouped Frequency Distribution Simplified Calculation Calculating the Mean from a Stem and Leaf Diagram Characteristics of the Arithmetic Mean The Mode Mode of a Simple Frequency Distribution Mode of a Grouped Frequency Distribution Calculating the Mode from a Stem and Leaf Diagram Characteristics of the Mode The Median Median of a Simple Frequency Distribution Median of a Grouped Frequency Distribution Calculating the Median from a Stem and Leaf Diagram Page 222 223 224 225 226 230 231 232 232 232 234 235 236 236 237 238 243 B.

© ABE and RRC . eggs from Breed A are. in general. Distributions with different averages indicate that there is a different general level of the variate in the two groups. we commonly use such a figure when we talk about the "average" value of something or other. Having weighed a large number of eggs from each breed. In everyday life. heavier than those from Breed B. There are three such measures with which you need to be familiar:    the arithmetic mean the mode the median. A frequency distribution may be used to give us concise information about its variate. in reference to the two kinds of egg. but often we wish to compare two or more distributions. Drawing the frequency curves has enabled us to make an important general statement concerning the relative egg weights of the two breeds of poultry. make sure that you have understood them well before proceeding. In this case the distributions overlap and.1: Frequency polygon showing weight of eggs from two breeds of hen Frequency   Breed B       Breed A           Weight of eggs  Examining these distributions you will see that they look much alike except for one thing – they are located on different parts of the scale. We would like to calculate some figure which will serve to indicate the general level of the variate under discussion. that those from Breed A had a higher average weight than those from Breed B. but we would now like to take the matter further. The two frequency polygons might well look something like Figure 8.222 Measures of Location INTRODUCTION We have looked at frequency distributions in detail in Study Unit 6 and you should. which is a topic in which you would be interested if you worked for an egg marketing company. Consider for example the distribution of the weights of eggs from two different breeds of poultry. Remember that one of the objects of statistical analysis is to condense unwieldy data so as to make it more readily understood.1. by means of a quick revision. Figure 8. We might have said. we might have compiled frequency distributions and graphed the results. The single value which we use to describe the general level of the variate is called a "measure of location" or a "measure of central tendency". although some eggs from Breed A are of less weight than some eggs from Breed B.

and so decide on the appropriate measure of location to use in common situations. It is obtained by adding together all the values and then dividing the total by the number of values involved. the following set of values which are the heights. THE ARITHMETIC MEAN The arithmetic mean is what we normally think of as the "average" of a set of values. for example. Also. A. of seven men: Table 8. Notice that some values occur more than once. written Σ. median and mode. we indicate the sum of a number of values of x by the Greek letter sigma. egg weights. but we still add them all. so make sure you understand the meaning of the expression. graphical methods utilise the assumed mean method and the "class interval" method in calculating the arithmetic mean list the relative advantages and disadvantages of the arithmetic mean.Measures of Location 223 Objectives When you have completed this study unit you will be able to:     calculate the arithmetic mean and mode of grouped and ungrouped distributions. etc. At this point we must introduce a little algebra. For this reason in general discussion we will use (as you will recall from algebra) some general letter. in centimetres. whether it is heights. Take.1: Height of seven men Man A B C D E F G Total Height (cms) 187 160 163 180 180 168 187 1225 The arithmetic mean of these heights is 1225  7  175 cms. Thus. © ABE and RRC . including stem and leaf diagrams find the median of grouped and ungrouped data (including stem and leaf diagrams) by calculation and. wages. we may write: Σx  1225 We shall be using sigma notation extensively in this unit. where appropriate. We don't always want to specify what particular items we are discussing. in our example. usually x.

However if we had a much larger number of items the method would save a lot of additions. Example 2: Consider the following table and. the arithmetic can be eased somewhat by forming a frequency distribution. as practice. Consider the following example. Do this before moving on. you can see that Σf  n and that. The calculation of the arithmetic mean can then be described by formula: x Σx n or 1 x  Σx n The latter expression is customary in statistical work. of men at this height (f) 1 1 1 2 2 Σf  7 Product (fx) 160 163 168 360 374 Σ(fx)  1225 Indicating the frequency of each value by the letter f. © ABE and RRC . with only seven items it would not be necessary in practice to use this method. x . we have: x 1 (1225 )  175 cms. Example 1: Table 8.2: Height distribution data Height in cms (x) 160 163 168 180 187 Total No. the formula becomes: x Σ( fx ) Σf which is 1225/7  175 cms as before. 7 You will often find the arithmetic mean simply referred to as "the mean" when there is no chance of confusion with other means (which we are not concerned with here). Applying it to the example above. when the x's are not all the separate values but only the different ones. complete the (fx) column and calculate the value of the arithmetic mean. Of course. The Mean of a Simple Frequency Distribution When there are many items (that is when n is large).224 Measures of Location We indicate the arithmetic mean by the symbol x (called "x bar") and the number of items by the letter n.

In this case.3: Data for calculation of arithmetic mean Value of x 3 4 5 6 7 8 9 10 Total You should have obtained the following answers: the total numbers of items. The Mean of a Grouped Frequency Distribution Suppose now that you have a grouped frequency distribution.Measures of Location 225 Table 8. x  713  7. only the groups in which they lie (unless you have a stem and leaf diagram. It is most important that you do not get muddled about calculating arithmetic means. We do this by assuming that all the values in any group are equal to the midpoint (or mid-value) of that group. which we will discuss later). before going on to the next paragraph. but we can make an approximation sufficiently accurate for most statistical purposes. we do not know the actual individual values. The procedure is very similar to that for a simple frequency distribution (which is why we stressed the need for revision) and is shown in this example: © ABE and RRC . Revise it if necessary. you will remember. Σf  100 the total product.13 100 Number of items (f) 1 4 7 15 35 24 8 6 Product (fx) 3 16 35 Make sure that you understand everything we have covered so far. How then can we calculate the arithmetic mean? The answer is that we cannot calculate the exact value of x . Σ(fx)  713 the arithmetic mean.

226

Measures of Location

Table 8.4: Data for calculation of arithmetic mean of a grouped frequency distribution Group 0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 Total
x Σfx 1   1,660  33.2 Σf 50

Mid-value (x) 5 15 25 35 45 55 65

Frequency (f) 3 6 10 16 9 5 1 Σf  50

Product (fx) 15 90 250 560 405 275 65 Σ(fx)  1,660

Provided that Σf is not less than about 50 and the number of groups is not less than about 12, the arithmetic mean thus calculated is sufficiently accurate for all practical purposes. There is one pitfall to be avoided when using this method – if all the groups should not have the same class interval, be sure that you get the correct mid-values! Table 8.5 is part of a larger table with varying class intervals, to illustrate the point: Table 8.5: Data for grouped frequency distribution Group 0 to < 10 10 to < 20 20 to< 40 40 to < 60 60 to < 100 Mid-value (x) 5 15 30 50 80

You will remember that in discussing the drawing of histograms we had to deal with the case where the last group was not exactly specified. The same rules for drawing the histogram apply to the calculation of the arithmetic mean.

Simplified Calculation
It is possible to simplify the arithmetic involved in calculating an arithmetic mean still further by the following two devices:   working from an assumed mean in the middle of one convenient class working in class intervals instead of in the original units.

Let us consider the first device.

ABE and RRC

Measures of Location

227

(a)

Working from an assumed mean in the middle of one convenient class If you go back to our earlier examples, you will discover after some arithmetic that if you add up the differences in value between each reading and the true mean, then these differences add up to zero. Take first the height distribution (Table 8.1) considered at the start of this section: Table 8.6: Height distribution data, x  175 cms Man A B C D E F G Height, x (cms) 187 160 163 180 180 168 187 (x – x ) (cms) 12 15 12 5 5 7 12

Σ(x  x)  34  34  0

That is, Σ(x  x)  0 . Second, consider the grouped frequency distribution (Tables 8.3, 8.4) given earlier in this section: Table 8.7: Data for grouped frequency distribution, x  33.2 Group 0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 Total That is, Σ(x  x)  0 . If we take any value other than x and follow the same procedure, the sum of the differences (sometimes called deviations) will not be zero. In our first example, let us assume the mean to be 170 cms and label the assumed mean xo. The differences between each reading and this assumed value are: Mid-value (x) 5 15 25 35 45 55 65 Frequency (f) 3 6 10 16 9 5 1 Σf  50 (x – x )  28.2 18.2 8.2 1.8 11.8 21.8 31.8 f(x – x )  84.6 109.2 82 28.8 106.2 109 31.8

Σ(f(x – x ))  275.8  275.8  0

ABE and RRC

228

Measures of Location

Table 8.8: Differences between x and xo for height distribution data Man A B C D E F G Height, x (cms) 187 160 163 180 180 168 187 d = (x – xo) (cms) 17 10 7 10 10 2 17

Σ(x  xo)  54  19  35 That is Σ(x – xo)  35 cms or Σd  35 cms. We can make use of this property and use it as a short cut for finding x . We start by choosing some value of x as an assumed mean. We try to choose it near to where we think the true mean ( x ) will lie, and we always choose it as the midpoint of one of the groups when we are dealing with a grouped frequency distribution. In the example just given, the total deviation (d) does not equal zero, so 170 cannot be the true mean. As the total deviation is positive, we must have underestimated in our choice of xo, so the true mean is higher than 170. As there are seven readings, we need to adjust xo upwards by one seventh of the total deviation, i.e. by (35)  7  5. Therefore the true value of x is:
170  (35)  170  5  175 cms. 7

We know this to be the correct answer from our earlier work. Let us now illustrate the short cut method for the grouped frequency distribution. We shall take xo as 35 as this is the mid-value in the centre of the distribution. Table 8.9: Differences between x and xo for grouped frequency distribution data Group 0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 Total This time Σfd  90. Mid-value (x) 5 15 25 35 45 55 65 Frequency (f) 3 6 10 16 9 5 1 Σf  50 d = (x – xo) 30 20 10 0 10 20 30 f(x – xo) = fd 90 120 100 0 90 100 30

Σfd  310  220  90

ABE and RRC

Measures of Location

229

This means we must have overestimated xo, as the total deviation (Σfd) is negative. As 1 there are 50 readings altogether, the true mean must be 50 th of (90) lower than 35, i.e.
x  35  (90)  35  1.8  33.2 50

which is as we found previously. (b) Working in class intervals instead of in the original units This method of calculation can be used with a grouped frequency distribution to work in units of the class interval instead of in the original units. In the fourth column of Table 8.9, you can see that all the deviations are multiples of 10, so we could have worked in units of 10 throughout and then compensated for this at the end of the calculation. Let us repeat the calculation using this method. The result (with xo  35) is set out in Table 8.10. Note that here the symbol used for the length of the class interval is c, although you may also come across the symbol i used for this purpose. Table 8.10: Differences between x and xo for grouped data using class intervals Group Mid-value (x) 5 15 25 35 45 55 65 Frequency (f) 3 6 10 16 9 5 1 Σf  50 (x – xo) d=
x  xo c

fd

0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 Total

–30 –20 10 0 10 20 30
Σ

3 2 1 0 1 2 3
x  xo   22  31   9 c

9 12 10 0 9 10 3

Remembering that we need to compensate, by multiplying by 10, for working in class interval units in the table:
x  35  x  35  (9)  10 50 90  35  1.8  33.2 as before. 50

The general formula for x that applies for all grouped frequency distributions having equal class intervals can be written as:
x  xo  Σfd c n

where d 

x  xo c

ABE and RRC

230

Measures of Location

As we mentioned at an earlier stage, you have to be very careful if the class intervals are unequal, because you can only use one such interval as your working unit. Table 8.11 shows how to deal with this situation. Table 8.11: Differences between x and xo using unequal class intervals Group Mid-value (x) 5 15 25 35 45 60 d=
x  xo c

Frequency (f) 3 6 10 16 9 6 50

Product (fd) 9 12 10 0 9 15 24  (31)  7

0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 70 Total

3 2 1 0 1 2½

The assumed mean is 35 as before, and the working unit is a class interval of 10. Notice how d for the last group is worked out: the midpoint is 60, which is 2½ times 10 above the assumed mean. The required arithmetic mean is, therefore:
x  35  7  10 70  35   35  1.4  33.6 50 50

We have reached a slightly different figure from before because of the error introduced by the coarser grouping in the "50 to < 70" region. The method just described is of great importance. By using it correctly, you can often do the calculations for very complicated-looking distributions by using mental arithmetic and pencil and paper. With the advent of electronic calculators, the time saving on calculations of the arithmetic mean is not great, but the method is still preferable because:   The numbers involved are smaller and thus you are less likely to make slips in arithmetic. The method can be extended to enable us to find easily the standard deviation of a frequency distribution (which we shall examine in Study Unit 9).

Calculating the Mean from a Stem and Leaf Diagram
You will recall that a stem and leaf diagram presents frequency distributions of grouped data graphically. However because these diagrams preserve the original data, they can be used for further analysis to calculate the mean relatively straightforwardly, as if the data was ungrouped. Consider the following example of a stem and leaf diagram in Figure 8.2, which shows the distribution of marks awarded to an ice skater in a competition. Note that the stem values represent the units of a mark awarded and the leaf values represent the tenths of a mark awarded (i.e. a mark of 5.8 is expressed in the stem and leaf diagram as 5|8).

ABE and RRC

Measures of Location

231

Figure 8.2: Stem and leaf diagram to show the marks awarded to a competition skater Stem 0 1 2 3 4 5 5 9 0 1 1 3 0 5 8 Leaf 0

Note: 0|0  0.0 marks Although the stem and leaf diagram presents grouped data, the individual data is preserved so that the mean can be calculated using the simple formula:
x Σx n Σ(0.0  3.5  3.9  4.0  4.1  4.1  4.3  5.0  5.5  5.8) 10 40.2  4.0 10

Accordingly:
x x

Characteristics of the Arithmetic Mean
There are a number of characteristics of the arithmetic mean which you must be aware of:  It is not necessary to know the value of every item in a distribution in order to calculate the arithmetic mean. Only the total and the number of items are needed. For example, if you know the total wages bill and the number of employees, you can calculate the arithmetic mean wage without knowing the wages of each person. The arithmetic mean is fully representative because it is based on all, and not just some, of the items in the distribution. One or two extreme values can make the arithmetic mean somewhat unrepresentative by their influence on it. For example, if a millionaire came to live in a country village, the inclusion of his or her income in the arithmetic mean for the village might make the place seem very much better off than it really was! The arithmetic mean is reasonably easy to calculate and to understand. In more advanced statistical work, it has the advantage of being amenable to algebraic manipulation.

 

 

ABE and RRC

The average size of garment in demand might turn out to be. A common error is to say that the mode of the distribution in Table 8. you might ask.13? © ABE and RRC .52. The mode is used in those cases where it is essential for the measure of location to be an actually occurring value. is a value of the variate x.3724. the mode of the distribution obtained from the survey would be an actual value (perhaps size 8) and it would provide the answer to the problem. and do not fall into this trap! For comparison. let us say. like the mean.12 Accident distribution data Number of accidents per day (x) 0 1 2 3 4 Total Number of days with the stated number of accidents (f) 27 39 30 20 7 123 In this case the most frequently occurring value is 1 (it occurred 39 times) and so the mode of this distribution is 1. which we will discuss later). Mode of a Grouped Frequency Distribution When the data is given in the form of a grouped frequency distribution. Watch out. This is wrong. THE MODE The first alternative to the mean which we will discuss is the mode. Mode of a Simple Frequency Distribution Consider the following frequency distribution: Table 8. Note that the mode. is the mode of the following distribution shown in Table 8. it is not quite so easy to determine the mode (unless you have a stem and leaf diagram. This is the name given to the most frequently occurring value.12 is 39. not the frequency of that value. The mode is 1. This is not an actually occurring value and doesn't help us to answer our problem. An example is the case of a survey carried out by a clothing store to determine what size of garment to stock in the greatest quantity. calculate the arithmetic mean of the distribution – it works out at 1. What. However. 9.232 Measures of Location B.

Likewise. A number of procedures based on the frequencies in the groups adjacent to the modal group can be used. which can be chosen quite arbitrarily to suit our convenience. In practical work. but this is not true. You should note. You may be tempted to say that the mode is 75. The distribution could have been set out with class intervals of five instead of 10.Measures of Location 233 Table 8. we will get different modal groups if the grouping is by 15 or by 20 or by any other class interval. however. and so the midpoint of the modal group is not a good way of estimating the mode. that these procedures are only mathematical devices for finding the most likely position of the mode – it is not possible to calculate an exact and true value in a grouped distribution.13: Data for determining the mode of a grouped frequency distribution Group 0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 70 to < 80 80 to < 90 90 to < 100 100 to < 110 110 to < 120 Frequency 4 6 10 16 24 32 38 40 20 10 5 1 All we can really say is that "70 to < 80" is the modal group (the group with the largest frequency). this determination of the modal group is usually sufficient. but you may occasionally be asked for the mode to be determined from a grouped distribution. nor even a useful approximation in most cases. to illustrate the point): Table 8.14: Revised data for determining the mode of a grouped frequency distribution Group 60 to < 65 65 to < 70 70 to < 75 75 to < 80 80 to < 85 85 to < 90 Frequency 16 22 21 19 12 8 The modal group is now "65 to < 70". The reason is that the modal group depends on the method of grouping. and would then have appeared as follows (only the middle part is shown. © ABE and RRC . and we shall now describe one procedure.

This suggests to us that the mode may be some way towards the lower end of the modal group rather than at the centre. Draw that part of the histogram which covers the modal class and the adjacent classes on either side. In our example the mode is just less than 71. This method can be used when the distribution has equal class intervals.4. Draw in the diagonals AB and CD as shown in Figure 8. Let us consider our previous example of a stem and leaf diagram which shows the distribution of marks awarded to an ice skater in a competition (Figure 8. © ABE and RRC . The group below ("60 to < 70") has a frequency of 38. Remember that the stem values represent the units of a mark awarded and the leaf values represent the tenths of a mark awarded. previously as Figure 8. Now examine the groups on each side of the modal group. and the one above ("80 to < 90") has a frequency of 20.3. This means that these diagrams can be used to calculate the mode relatively straightforwardly as if the data was ungrouped. Where this line crosses the horizontal axis is the mode. as we have remarked the original data is preserved.234 Measures of Location We saw that the modal group of our original distribution was "70 to < 80". A graphical method for estimating the mode is shown in Figure 8. Figure 8.3: Using a histogram to determine the most likely value of the mode Class frequency 40 A C D 30 20 B Mode 60 70 80 90 Variate Calculating the Mode from a Stem and Leaf Diagram Although a stem and leaf diagram presents frequency distributions of grouped data graphically.2)). From the point of intersection draw a vertical line downwards.3.

13. In our example.e. 7. 7.  © ABE and RRC . 12 each number occurs only once so this distribution has no mode. in the set of numbers 7. 10. i. it is not based on all the items in the distribution. 7. namely 7 and 10. It is not unique. It is not affected by the occurrence of extreme values. there can be more than one mode. 9. 12. 6.1).Measures of Location 235 Figure 8. 8. It can only be determined approximately with grouped distributions. since no calculation is required. The mode may not exist. in the set of numbers. This set of numbers would be referred to as having a bimodal distribution. It is not amenable to the algebraic manipulation needed in advanced statistical work. Unlike the arithmetic mean. 10.0 marks As the mode is defined as the data value that occurs most often. but only on those near its own value. 11.4: Stem and leaf diagram to show the marks awarded to a competition skater Stem 0 1 2 3 4 5 5 9 0 1 1 3 0 5 8 Leaf 0 Note: 0|0  0.e. For example. 8. 10. Characteristics of the Mode Some of the characteristics of the mode are worth noting. For example. two data points of value 4. In ungrouped distributions the mode is an actually occurring value. 8. there are two '1' leaves on the '4' stem (i. so the mode is 4.1. all we need to do is look for the leaf (number) that occurs the most often on one stem of the diagram. there are two modes.        The mode is very easy to find with ungrouped distributions. 10. particularly as they compare with those of the arithmetic mean.

the median will be the 62nd item because that will be the middle item. 12 The two values in the middle are 8 and 9. it was easy to pick out the middle value because there was an odd number of values. 13 The total of these eleven numbers is 100 and so the arithmetic mean is therefore 100/11  9. as you have already learned. Our first rule for determining the median is. THE MEDIAN The desirable feature of any measure of location is that it should be near the middle of the distribution to which it refers. Median of a Simple Frequency Distribution Of course statistical data is rarely in such small groups and. as follows: Arrange all the values in order of magnitude and the median is then the middle value. 10. if a value is near the middle of the distribution. 12. because there are five values above and five values below 9. 8. Table 8.15 repeats the data from Table 8. How then do we find the median if our data is in the form of a distribution? Let us take the example of the frequency distribution of accidents already used in discussing the mode (Table 8.15: Accident distribution data Number of accidents per day (x) 0 1 2 3 4 Total Number of days with the stated number of accidents (f) 27 39 30 20 7 123 Cumulative frequency (F) 27 66 96 116 123 © ABE and RRC . 7. The value which divides the distribution into equal parts is called the median. by convention. Now. To see what the value of the 62nd item will be. and the other half to have smaller values. The total number of values is 123. But what if there is an even number? Then. Table 8. In the example just dealt with. take the following set of values: 6. Look at the following set of values: 6. 9. This suggests to us that a possible measure of location might be that value which is such that exactly half (50%) of the distribution has larger values and exactly half has lower values. For example. the median is taken to be the arithmetic mean of the two values in the middle. they must all be put separately into the list.13).236 Measures of Location C. and so when those values are arranged in order of magnitude. The median however is 9. 7. 10.12. 7. so the median is 8½. 8. while the mode is 10 because that is the number which occurs most often (three times). 9. let us draw up the distribution again. 7. 10. but adds a column setting out the cumulative frequency. 10. 8. 11. Note that all the values are to be used – even though some of them may be repeated. then we expect about half of the distribution to have larger values. we usually deal with frequency distributions. therefore. 8.091 (three decimal places). 10.

In our example the median would be read against 103 on the cumulative frequency scale (see Figure 8. the corresponding value of the variate. against a cumulative frequency of half the total frequency. if we were to list all the separate values in order. must lie in the "60 to < 70" group. read off the median. from the cumulative frequency column. It therefore follows that the 62nd item would be a 1 and that the median of this distribution is 1. the first 27 would all be 0's and from then up to the 66th would be 1's. from the cumulative frequency column. and because we do not know the exact values of the items in each group. Then. so we need to move 11/38 of the way into that group. a convenient way to find the median of a grouped distribution is to draw the ogive. Our median is therefore: 60  110  60  2. is very similar to the solution for an ungrouped distribution – we halve the total frequency and then find. then the median is always read off against 50%. Because a grouped frequency distribution nearly always has a large total frequency. that is 11/38 of 10 above 60.16: Data for determining the median of a grouped frequency distribution Group 0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 70 to < 80 80 to < 90 90 to < 100 100 to < 110 110 to < 120 Total Frequency (f) 4 6 10 16 24 32 38 40 20 10 5 1 206 Cumulative frequency (F) 4 10 20 36 60 92 130 170 190 200 205 206 The total frequency is 206 and therefore the median is the 103rd item which.5). it is not necessary to find the two middle items when the total frequency is even – just halve the total frequency and use the answer (whether it is a whole number or not) for the subsequent calculation. If the ogive is drawn with relative frequencies.89 38 The use of the cumulative frequency distribution will no doubt remind you of its graphical representation.Measures of Location 237 You can see from the last column that. Median of a Grouped Frequency Distribution The final problem connected with the median is how to find it when our data is in the form of a grouped distribution. there are 92 items before we get to that group and we need the 103 rd item. so we obviously need to move into that group by 11 items. Altogether in our "60 to < 70" group there are 38 items. The solution to the problem. But exactly where in the "60 to < 70" group? Well. In practice. the ogive.89  62. as you might expect. Table 8. © ABE and RRC .

Again. remember that the stem values represent the units of a mark awarded and the leaf values represent the tenths of a mark awarded. we can simply go through the data presented in our stem and leaf diagram and cross off pairs of high and low leaves. If we start with the © ABE and RRC .6: Stem and leaf diagram to show the marks awarded to a competition skater Stem 0 1 2 3 4 5 5 9 0 1 1 3 0 5 8 Leaf 0 Note: 0|0  0. but the individual data is preserved. Again.6). As the median is the middle value in the set.5: Reading the median from an ogive Cumulative frequency 220 200 180 160 140 120 100 80 60 40 20 0 0 20 40 60 80 100 120 103 Median  63 Value of the variate Calculating the Median from a Stem and Leaf Diagram Similarly. This means that the median can be calculated relatively straightforwardly. we can use a stem and leaf diagram to calculate the median value of a distribution of grouped data as the original data is preserved.238 Measures of Location Figure 8. the stem and leaf diagram presents grouped data. let us consider our previous example of a stem and leaf diagram which shows the distribution of marks awarded to an ice skater in a competition (Figure 8. Figure 8.0 marks As noted.

Thus median is of value when we have open-ended classes at the edges of the distribution. which you should compare with those of the mean and the mode. is thus 4.Measures of Location 239 rightmost leaf on the bottom stem and the leftmost leaf on the top stem. It is less affected by extreme values than the mean. The median is not very amenable to further algebraic manipulation. we are left with two middle values. for his height will be the median height. That value (or the average of the two values) is the thus median. Characteristic features of the median. The average of these values. and is readily understood as being the "halfway point". and hence the median.1. The millionaire in the country village might alter considerably the mean income of the village but he or she would have almost no effect at all on the median. namely the two '1' leaves on the '4' stem. as its calculation does not depend on the precise values of the variate in these classes.   © ABE and RRC . In our example. if we want to know the median height of a group of 21 men. whereas the value of the arithmetic mean does. are as follows:   It is fairly easily obtained in most cases. we do not have to measure the height of every single one. It is only necessary to stand the men in order of their heights and then only the middle one (number 11) need be measured. It can be obtained without actually having all the values. For example. we can work our way towards the middle of the values crossing off the pairs until we have only one (or two) leave(s) left that have not been crossed out.

Consumption of electricity of 100 householders (one week) Consumption (kilowatt hours) 0 – under 10 10 – under 20 20 – under 30 30 – under 40 40 – under 50 50 – under 60 60 – under 70 70 – under 80 80 or over Number of householders 5 8 20 29 20 11 6 1 0 2. the average) size of shareholding. The following table shows the consumption of electricity of 100 householders during a particular week. As part of a health study the heights of several dozen men in a particular location are measured as a sample. What sort of "average" would you calculate from the sample. it is found that the sample contains two brothers who are dwarfs. System B Calculate the mean rate of discharge for each system. Using the short-cut methods given in this study unit. from the grouped distribution in the following table. After this has been done.240 Measures of Location Questions for Practice 1. calculate the arithmetic mean (i. © ABE and RRC . and why? The rate of effluent discharge from two plants based on different systems was sampled at four-hourly intervals and the results are given below: System A 10 13 10 7 12 8 17 4 10 10 18 18 11 9 6 19 9 12 5 10 12 13 10 12 14 10 11 13 8 16 13 16 9 12 6 9 12 7 15 12 7 6 14 3 13 5 10 4 10 10 12 12 12 11 9 4 11 14 8 9 3.e. 4. Calculate the arithmetic mean consumption of the 100 householders.

The following distribution gives the weight in kilograms of 100 schoolboys. The number of cars asked for in a day has varied as follows: Cars wanted in a day 0 1 2 3 4 5 or more (a) (b) (c) What is the mean daily demand? What is the mean number supplied (i.02 0 6.e. Find the median weight. A garage keeps two cars for daily hire. hired out) per day? On what fraction of days is some demand turned away? Proportion of days 0. of schoolboys 2 15 37 31 11 4 100 © ABE and RRC .000 Number of persons with this size of holding 23 55 122 136 100 95 81 63 17 8 5.36 0.38 0. Weight in kg 45 – 49 50 – 54 55 – 59 60 – 64 65 – 69 70 – 74 No.Measures of Location 241 Distribution of shareholding Number of shares Less than 50 50 – 99 100 – 199 200 – 299 300 – 399 400 – 499 500 – 599 600 – 699 700 – 799 800 – 1.06 0.18 0.

The age distribution of employees working for a manufacturing company is given in the following stem and leaf diagram. of persons with these shares with these shares 80 142 110 75 52 38 300 – 349 350 – 399 400 – 449 450 – 499 500 – 549 550 – 599 28 15 9 7 3 2 8. The holdings of shares (nominal value £1) in DEF plc are given in the following table. Stem 1 2 3 4 Leaf 8 8 9 0 3 5 5 7 9 1 3 8 2 3 5 Note: 18  18 years old Based on the data presented in this diagram. of persons No. of shares held Under 50 50 – 99 100 – 149 150 – 199 200 – 249 250 – 299 No. © ABE and RRC .242 Measures of Location 7. No. Calculate (a) the arithmetic mean holding. of shares held No. Now check your answers with those given at the end of the unit. (b) the median holding and (c) the modal holding. calculate the: (a) (b) (c) mean median mode.

the arithmetic mean of the householders' consumption is 36.3 kilowatt hours. The mode would not be so affected.53 30 316  10. Consumption of electricity of 100 householders (one week) Consumption (kwH) 0 – under 10 10 – under 20 20 – under 30 30 – under 40 40 – under 50 50 – under 60 60 – under 70 70 – under 80 80 or over Mid-value (x) 5 15 25 35 45 55 65 75 Number of householders (f) 5 8 20 29 20 11 6 1 0 100 x fx 25 120 500 1. or might not exist if all the heights were different.3 Σf 100 i. 2. © ABE and RRC .e. but with such a small sample (a dozen or so) it would be erratic. but it is probably quicker just to add the numbers as they stand. Normally one thinks of the arithmetic mean as the best measure of average. In order to give the correct impression of the mean in the sample location.015 900 605 390 75 0 3.630   36.53 30 3.630 Σfx 3.Measures of Location 243 ANSWERS TO QUESTIONS FOR PRACTICE 1. You can group the data first if you wish. but in this case it would be too much affected by the two extremely low values to be representative. it would be best to use the median. System A: total number of readings  30 total of rates of discharge  316 mean rate of discharge  System B: total number of readings  30 total of rates of discharge  316 mean rate of discharge  316  10.

00 fx1 0 0. the most common class interval.0 74.0 3.5 189.18 0.0 4.7 2.00 Mean demand  Σfx1 1.5 -122.0 No.5 349.7 85.36 0.5 648.7  100 700  250  92. Therefore.08 0 1.5 449.3 1.6 -0.06 0.8 -96.02 0 1. and xo  250.8 1. 5.0 5.5 649.00 © ABE and RRC .7 99. of shares Less than 50 50 – 99 100 – 199 200 – 299 300 – 399 400 – 499 500 – 599 600 – 699 700 – 799 800 – 1.0 6.36 0.0 0.5 149.244 Measures of Location 4.18 0. Distribution of shareholding Mid-value (x) 25. the middle of the modal group.7  343 to the nearest whole number. Take c  100. the mean shareholding is 343 shares.5 900.5 249. (a) Demand (x1) 0 1 2 3 4 5 or more Frequency (as a proportion) (f) 0.000 Total x  250  d= fd -51.38 0. of persons with a holding of x (f) 23 55 122 136 100 95 81 63 17 8 700 x  xo c No.5 242.00   1 car per day Σf 1.38 0.0 648.0 1.5 749.5 549.0 2.0 52.6 251.

06  0.06  0.46  59. Weight in kg 45 – 49 50 – 54 55 – 59 60 – 64 65 – 69 70 – 74 No.00 fx2 0 0. Half the total frequency is 50 so the median lies in the "55 – 59 kg" group.38 0.00 Demand is turned away whenever three or more cars are requested in one day.02) Frequency (as a proportion) (f) 0.46  60 kg to the nearest kg.90 cars per day Σf 1. and in the (50  17)  33rd item within it.02  0. of schoolboys 2 15 37 31 11 4 100 Cumulative frequency 2 17 54 85 96 100 6. This proportion of days is 0.36 0.18  0. 33 5 37 © ABE and RRC . on eight days in every 100 demand is turned away.90   0. Median weight  55   55  4.90 Mean supply  (c) Σfx 2 0.Measures of Location 245 (b) Supply (x2) 0 1 2 (0. Therefore.38 0.52 0.26 1.08.

on a histogram.041 242  799 Cumulative frequency 80 222 332 407 459 497 525 540 549 556 559 561 4 3 2 1 0 1 2 3 4 5 6 7 (a) x  224.5 Distribution of the sizes of shareholdings in DEF plc Number of shares held Under 50 50 – 99 100 – 149 150 – 199 200 – 249 250 – 299 300 – 349 350 – 399 400 – 449 450 – 499 500 – 549 550 – 599 Total No.5  222) item within it. (c) The modal group is the "50 – 99" group and the estimated mode is given by drawing this group.5 124.5 74. Reading from this.5  (1. together with the surrounding class groups. You may have chosen a different xo.5 524.5 424.5th (280.5 274.) (b) Half of the total frequency is 280.5  71.5 574.B.5 224.5  58.5 d= fd 320 426 220 75 0 38 56 45 36 35 18 14 1.5 174. This is shown on the next page. Median  99.424  50)  224. xo  224. but the result will be the same.3  153 shares to the nearest whole number.5 and therefore the median lies in the "100 – 149" group and it is the 58.1 110  126 shares to the nearest whole number. Take c  50.246 Measures of Location 7.5 374.5  50  126.5 474. of persons with these shares 80 142 110 75 52 38 28 15 9 7 3 2 561 x  xo c Mid-value 24. © ABE and RRC .2  153. the estimated mode  83 shares. (N.5 324.5   799  50 561  224.

(a) x Σfx (18  18  19  20  23  25  25  27  29  31  33  38  42  43  45)  Σf 15  29. © ABE and RRC .e.1 i. the arithmetic mean age of manufacturing employees is 29.1 years. the data set is bimodal with the mode age of manufacturing employees being 18 years and 25 years. (b) (c) The median age of manufacturing employees is 27 years.e. two data points of value 18) and two 5 leaves on the 2 stem (i. there are two 8 leaves on the 1 stem (i. two data points of value 25).Measures of Location 247 Frequency (no of persons) 140 120 100 100 – 50 group 80 Under 50 group 50 Mode  83 100 150 Shareholding 8. In our example. Thus.e.

248 Measures of Location © ABE and RRC .

D. Range The Quartile Deviation. Deciles and Percentiles The Quartile Deviation Calculation of the Quartile Deviation Deciles and Percentiles Standard Deviation The Variance and Standard Deviation Alternative Formulae for Standard Deviation and Variance Standard Deviation of a Simple Frequency Distribution Standard Deviation of a Grouped Frequency Distribution Characteristics of the Standard Deviation The Coefficient of Variation Skewness Page 250 251 251 251 253 254 255 255 257 258 259 262 262 263 268 C. E. B. Answers to Questions for Practice © ABE and RRC .249 Study Unit 9 Measures of Dispersion Contents Introduction A.

the figures which go to make up a distribution may all be very close to the central value.5 Week 1 2 3 4 5 6 7 8 9 10 Total Mean output Although the two factories have the same mean output. We therefore need to supplement it with a measure of dispersion. Examine the figures in the following table: Table 9. We will discuss each of these in this unit. However. knowledge of this spread. there are several different measures of dispersion in use by statisticians. Factory A achieves its mean production with only very little variation from week to week. which we call "dispersion" or "variation". The measures in common use are:    range quartile deviation standard deviation. Each has its own particular merits and demerits.5 Factory B 136 92 110 36 102 57 108 81 156 117 995 99. whereas Factory B achieves the same mean by erratic ups and downs from week to week. This example shows that a mean (or other measure of location) does not by itself tell the whole story. is of great importance. the mean of 49 and 51 is 50. As was the case with measures of location. When applying statistical methods to practical problems.250 Measures of Dispersion INTRODUCTION In order to get an idea of the general level of values in a frequency distribution.1: Output from two factories over 10 weeks Weekly output Factory A 94 100 106 100 90 101 107 98 101 98 995 99. © ABE and RRC . we have studied the various measures of location that are available. but the individual values may be spread about the mean in vastly different ways. which will be discussed later. they are very different in their weekto-week consistency. but the mean of 0 and 100 is also 50. or they may be widely dispersed about it – for example. You can see that these two distributions may have the same mean.

RANGE The range is the simplest measure of dispersion: it is simply the difference between the largest and the smallest.17.e. For Factory B the range is 156  36  120. The larger range for Factory B shows that it performs less consistently than Factory A. For these reasons it is used a great deal in industrial quality control work. you must cast your mind back to the method of obtaining the median from the ogive.1. To understand it. In the example just given. using the appropriate formulae calculate the standard deviation using the "corrected/approximate variance" method calculate the coefficient of variation and explain its use calculate Pearson's first and second coefficient of skewness from the appropriate data explain the nature of skewness and describe the relationship between the mean. As a result. deciles and percentiles of grouped data calculate the standard deviation and variance of grouped and ungrouped distributions. the range is not much used in statistical analysis except in the case just mentioned. i. one or two extreme results can make it quite unrepresentative. quartile deviation and standard deviation as measures of dispersion. The values which divide the total frequency into quarters are called quartiles and they can also be found from the ogive. median and mode in a skewed distribution identify the advantages and disadvantages of the range. A. Consequently. as shown in Figure 9. the quartile deviation. Its disadvantage is that it is based on only two of the individual values. B. The advantage of the range as a measure of the dispersion of a distribution is that it is very easy to calculate and its meaning is easy to understand. we can see that the lowest weekly output for Factory A was 90 and the highest was 107.Measures of Dispersion 251 Objectives When you have completed this study unit you will be able to:         calculate the range of a distribution from given data find. THE QUARTILE DEVIATION. by both calculation and graphical methods. you remember. is the value which divides the total frequency into two halves. DECILES AND PERCENTILES The Quartile Deviation This measure of dispersion is sometimes called the "semi-interquartile range". The median. the range is therefore 107 90. and takes no account of all those in between. © ABE and RRC .

252 Measures of Dispersion Figure 9. 50% and 75% of the cumulative frequency. 100% corresponds to 206. the total frequency. the median and the upper quartile (Q3) respectively.5 2 29. giving the lower quartile (Q1).e. You will notice that we have added the relative cumulative frequency scale to the right of the graph.75 © ABE and RRC . Q1  46.5 2 Q3  Q1 2  14. It is then easy to read off the values of the variate corresponding to 25%. i.1: Identifying quartiles from an ogive Cumulative 220 frequency 200 180 160 140 120 100 80 60 Relative 100 cumulative frequency (%) 75 50 25 40 20 0 0 20 40 60 80 100 120 Value of the variate This is the same ogive that we drew in Study Unit 8 when finding the median of a grouped frequency distribution.5 median  63 (as found previously) Q3  76 The difference between the two quartiles is the interquartile range and half of the difference is the semi-interquartile range or quartile deviation: Quartile deviation    76  46.

i.5 and then read from the ogive the values of the variate corresponding to 51. and 4 75% of the total frequency. Remember that graphical methods are never quite as accurate as calculations. © ABE and RRC . As the total frequency is 206: the 25% point  the 75% point  206  51½ and 4 3  206  154½.e.2: Table 9.e.Measures of Dispersion 253 Alternatively. the left-hand scale).2: Data for finding the quartile deviation of a grouped frequency distribution Group Frequency (f) 4 6 10 16 24 32 38 40 20 10 5 1 206 Cumulative frequency (F) 4 10 20 36 60 92 130 170 190 200 205 206 0 to < 10 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 70 to < 80 80 to < 90 90 to < 100 100 to < 110 110 to < 120 Total The lower quartile is the 25% point in the total distribution and the upper quartile the 75% point. The end result is the same.5. Calculation of the Quartile Deviation The quartile deviation is not difficult to calculate in situations where the graphical method is not acceptable.5 on the cumulative frequency scale (i. and the table of values is reproduced for convenience as Table 9. We shall again use the same example. i. you can work out: 25% of the total frequency.e. 206  51. 4 We can make the calculations in exactly the same manner as we used for calculating the median.5 and 154. 154.

D2 . the upper quartile  70   70  6.2.254 Measures of Dispersion Looking at Table 9.. the 51½th item comes in the "40 to < 50" group and will be the (51½  36)  15½th item within it.125  46. Such partition values are deciles and percentiles. for example. the lower quartile  40   40  6. P5 is the value below which 5% of the data lies and P64 is the value below which 64% of the data lies. Using the same example as previously. So. The quartile deviation    76. Again we can say that. it is less representative than the standard deviation (which we discuss later). the second decile (D2) is the value below which 20% of the data lies.458  46. P99.. let us calculate as an illustration the third decile D3. then the quartile deviation should be used as the measure of dispersion.458 Similarly. In general. The quartile deviation is unaffected by an occasional extreme value.8335  14. However it is not based on the values of all the items in the distribution and. and the sixth decile (D6) is the value below which 60% of the data lies. Deciles and Percentiles It is sometimes convenient.. © ABE and RRC . The deciles are labelled D1. to this extent. The percentiles are labelled P1.667 2 Q3  Q1 2 24½  10 40 15½  10 24  14. to consider values similar to the quartiles. From their names you will probably have guessed that the deciles are the values which divide the total frequency into tenths and the percentiles are the values which divide the total frequency into hundredths. For instance.458 2 29.8 to 1 decimal place. D9. P2 . when a median is the appropriate measure of location. Obviously it is only meaningful to consider such values when we have a large total frequency. particularly when dealing with wages and employment statistics. the upper quartile will be the 154½th item which is in the "70 to < 80" group and is the (154½  130)  24½th item within it. So..125  76. but which divide the distribution more finely.125 Remember that the units of the quartiles and of the median are the same as those of the variate. The method follows exactly the same principles as the calculation of the median and quartiles.

and is the (61.8th item. squared: variance  Σ( x  x ) 2 . C. Except for the use of the range in statistical quality control and the use of the quartile deviation in wages statistics. n Consider the following example in Table 9.Measures of Dispersion 255 The total frequency is 206 and D3 is the value below which 30% of the data lies: 30% of 206 is 30  206  61.8th item lies in the "50 to < 60" group.1. © ABE and RRC . We could also have found this result graphically – check that you agree with the calculation by reading D3 from the graph in Figure 9.8  10 32  50. You will see that the calculation method enables us to give a more precise answer than is obtainable graphically. Therefore 30% of our data lies below 50.8th item within it. and then squaring them.8  60)  1. where n is the total frequency. STANDARD DEVIATION The most important of the measures of dispersion is the standard deviation. and so we need to know how to calculate the variance first.3. So: D3  50   50  18 32 1. The standard deviation is defined as the square root of the variance. The Variance and Standard Deviation We start by finding the deviations from the mean. A glance at the cumulative frequency column shows that the 61.8 100 We are therefore looking for the value of the 61. which removes the negative signs in a mathematically acceptable fashion. the standard deviation is used almost exclusively in statistical practice.6 to one decimal place. The variance is then defined as the mean of the deviations from the mean.6.

25 2.25 0.25 0.85  4.5 Deviation (x  x) 5.50  22.85 Week 1 2 3 4 5 6 7 8 9 10 Total Mean output The mean of the squared deviations is the variance: variance   228.5 7.25 2.5 1.25 2. The symbol usually used to denote standard deviation is "σ" (theta).50 10 Σ( x  x ) 2 n  22.5 0.25 56.5 0. is: SD  Σ( x  x ) 2 . © ABE and RRC .78. again where n is the total frequency.25 2.25 228.0 Variance Deviation squared ( x  x)2 30.25 42.5 18.5 1. then: SD  22.5 1. n In this example.5 1.3: Output from Factory A over 10 weeks Weekly output (x) 94 100 106 100 90 101 107 98 101 98 995 99. although you may sometimes find "s" or "sd" used. which is defined as the square root of the variance.0 18.25 90.85 The formula for the standard deviation.256 Measures of Dispersion Table 9.5 9.5 6.

where n  Σf Formula (e) Formula (d). is derived directly from the definition of the standard deviation. as shown in the following sections.Measures of Dispersion 257 Alternative Formulae for Standard Deviation and Variance The formula we have used to calculate the standard deviation is: σ Σ(x  x )2 n Formula (a) where: x  value of the observations x  mean of the observations n  number of observations. If you already know x but it is not an integer (as in the previous example).    If you already know x and it is a small integer. then you should use formula (c) – particularly if you are not asked to calculate x . as follows: σ Σx 2  x2 n Formula (b) or Σx 2  Σx  σ   n  n  2 Formula (c) The choice of the formula to use depends on the information that you already have and the information that you are asked to give. When you are dealing with data in the form of a simple or grouped frequency distribution. As the standard deviation is defined as the square root of the variance. then formula (b) is the best to use. like formula (a). therefore: variance  σ2  Σ( x  x ) 2 n By expanding the expression Σ(x  x )2 we can rewrite the formula for standard deviation in two alternative and sometimes more useful forms. © ABE and RRC . If you do not know x . you should aim to keep the arithmetic as simple as possible and the rounding errors as small as possible. Formula (e) is obtained by using the sigma notation as in formula (c) and this is the one to use in calculations. then the formula for calculating the standard deviation needs to be expressed slightly differently: σ Σf ( x  x )2 Σf Formula (d) or σ 1 (Σfx )2   Σfx 2   n n    . In any calculation. there is no reason why you should not use formula (a).

50 Variance  22.5 6.5 7.85 (22.25 Frequency (f) 1 1 2 2 2 1 1 10 fx f(x – x )2 90 94 196 200 202 106 107 995 99.50 0. This is shown in Table 9.5  10 22.5 90.5 Deviation squared ( x  x)2 90.25 56.25 30.4: Determining the standard deviation of a simple frequency distribution Weekly output (x) 90 94 98 100 101 106 107 Total Mean Deviation (x  x) 9.25 42.25 2.50 4.25 4.258 Measures of Dispersion Standard Deviation of a Simple Frequency Distribution If the data in Table 9.25 30.85 The standard deviation is then worked out in the same way as previously.50 42.4 where we set out the information for calculating the standard deviation by applying formula (d): Table 9. as follows: σ Σf ( x  x )2 Σf σ 228.5 1.5 1.25 2.85 is the variance) σ  4.78 © ABE and RRC . then only the different values would appear in the "x" column and we would have to remember to multiply each result by its frequency.25 0.25 56.5 5.25 228.3 had been given as a frequency distribution (as is often the case).5 0.

6.231 Calculating the standard deviation proceeds as follows: σ 1 (Σfx )2   Σfx 2   n n    1 995 2   99.449 99.836 19.231   10  10  fx 90 94 196 200 202 106 107 995 x2 8.201 11. the standard deviation is calculated as follows.402 11.231  99.208 20.5 10 22.Measures of Dispersion 259 However.5: Determining the standard deviation of a simple frequency distribution Weekly output (x) 90 94 98 100 101 106 107 Frequency (f) 1 1 2 2 2 1 1 10 Therefore: n  Σf  10 Σfx  995 Σfx2  99.78.100 8.449 fx2 8.836 9. we again assume that all the readings in a given group fall at the midpoint of the group. we could have applied formula (e) and reduced the calculation as follows: Table 9.000 20.604 10.236 11.025   99.100 8.231   10  10    1  990.000 10.236 11. Standard Deviation of a Grouped Frequency Distribution When we come to the problem of finding the standard deviation of a grouped frequency distribution.5  10 1  228. © ABE and RRC .231      1 99.85 (the variance)  4.002. Using formula (e) and applying it to the distribution shown in Table 9.

500   58.025 3.562. In that method we:    worked from an assumed mean worked in class intervals applied a correction to the assumed mean.250 Σfx2  58.150 15.260 Measures of Dispersion Table 9.67 30 228.025 4.250 x2 225 625 1.950 Calculating the standard deviation proceeds as follows: σ 1 (Σfx ) 2   Σfx 2   n n    1  1.950      1 58. Table 9.6: Determining the standard deviation of a grouped frequency distribution Class 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 70 to < 80 Total Therefore: n  Σf  30 Σfx  1. © ABE and RRC .89 (the variance)  15.950   30  30  Mid-value Frequency (x) (f) 15 25 35 45 55 65 75 2 5 8 6 5 3 1 30 fx 30 125 280 270 275 195 75 1.225 5.625 58.950  52.800 12.125 12.866.33 30 1  6.083.625 fx2 450 3. but we can extend the short-cut method which we used for finding the arithmetic mean of a distribution to find the standard deviation as well.7 applies this approach to working out the standard deviation.225 2.13 The arithmetic for this table is rather tedious even with a calculator.675 5.950   30  30    1  1.125 9.2502   58.

2889 The standard deviation is then the class interval times the square root of the corrected variance: SD  2. Remember the following points:    Work from an assumed mean at the midpoint of any convenient class. The correction is always subtracted from the approximate variance. As you are working in class intervals. In our case the correction is: (20/30)2  0.13 This may seem a little complicated. c  10 The standard deviation is calculated in four steps from this table. it is necessary to multiply by the class interval as the last step. © ABE and RRC . but if you work through the example a few times.7333  0. as follows: (a) The approximate variance is obtained from 1 82  82   2.4444. it will all fall into place.4444  2.2889  1.7: Determining the standard deviation of a grouped frequency distribution (short cut) Class 10 to < 20 20 to < 30 30 to < 40 40 to < 50 50 to < 60 60 to < 70 70 to < 80 Total Mid-value Frequency (x) (f) 15 25 35 45 55 65 75 2 5 8 6 5 3 1 30 d = x  xo c fd 4 5 0 6 10 9 4 29 9 20 d2 4 1 0 1 4 9 16 fd2 8 5 0 6 20 27 16 82 2 1 0 1 2 3 4 xo  35.Measures of Dispersion 261 Table 9. (c) (d) The corrected variance is thus: 2.7333 30 30 1  Σfd2 which in our case is equal to: n (b) 1  The correction is    Σfd  n  2 It is always subtracted from the approximate variance to get the corrected variance.513 class intervals  10  15.

We call the result the "coefficient of variation". we can express the standard deviation as a percentage of the mean in each case. most important. THE COEFFICIENT OF VARIATION Suppose that we are comparing the profits earned by two businesses. Now. If the intervals vary too much.000. while the other may be a small firm with average monthly profits of only £2. It gives slightly more emphasis to the larger deviations but does not ignore the smaller ones and. Characteristics of the Standard Deviation In spite of the apparently complicated method of calculation. the small firm is more erratic in its earnings than the large firm. but for the SD it has to be squared. Note that although a standard deviation has the same units as the variate. D. the general level of profits is very different in the two cases. Let us suppose that both standard deviations come to £500. £500 is a much more significant amount in relation to the small firm than it is in relation to the large firm. relatively speaking. when we try to compare a French business with an American business.262 Measures of Dispersion    The correction factor is the same as that used for the short-cut calculation of the mean. To avoid the trouble of converting the dollars to euros (or vice versa) we can calculate the coefficients of variation in each case and thus obtain comparable measures of dispersion. So although they have the same standard deviations.000. To overcome this difficulty.0% 2. we revert to the basic formula. But do not omit it until you have really grasped the principles involved. but what about the month-by-month variability? We can compare the variability of the two firms by calculating the two standard deviations. it would be unrealistic to say that the two businesses are equally consistent in their monthly profits. the coefficient of variation is a ratio and thus has no units. It is based on all of the individual items. Clearly. the standard deviation is the measure of dispersion used in all but the very simplest of statistical studies. Another application of the coefficient of variation comes when we try to compare distributions the data of which are in different units – for example. © ABE and RRC .000 This shows that.0% 50. it can be treated mathematically in more advanced statistics. The column for d2 may be omitted since fd2  fd multiplied by d.000 500  100  25. The formula for calculating the standard deviation in this way can be written as: Σfd 2  Σfd  SD  c   n  n  2  The assumed mean should be chosen from a group with the most common interval and c will be that interval. we get coefficients of variation (usually indicated in formulae by V or CV) as follows: for the large firm: V  for the small firm: V  500  100  1. One of them may be a fairly large business with average monthly profits of £50. Applying the idea to the figures which we have just quoted.

© ABE and RRC .2: Symmetrical distributions Frequency Variate When the items are not symmetrically dispersed on each side of the mean. The more skew the distribution. the mean. Two distributions may have the same mean and the same standard deviation but they may be differently skewed. we say that the distribution is symmetrical. shows two symmetrical distributions. If a distribution is symmetrical. the median comes first and then the mean – see Figure 9.3 and then look at the same one through from the other side of the paper! Figure 9. but these two are always in the same order: working outwards from the mode. But in a skew distribution.e. although the mode is still at the point where the curve is highest.Measures of Dispersion 263 E.4. A distribution which has a tail drawn out to the right is said to be positively skewed. This will be obvious if you look at one of the skew distributions in Figure 9.2. Figure 9. right in the middle. mode and median all occur at the same point. while one with a tail to the left is negatively skewed. i. SKEWNESS When the items in a distribution are dispersed equally on each side of the mean. we say that the distribution is skewed or asymmetric.3: Skewed distributions Frequency (b) (a) Positively skewed Negatively skewed Variate What does skewness tell us? It tells us that we are to expect a few unusually high values in a positively skewed distribution or a few unusually low values in a negatively skewed distribution. the mean and the median lie somewhere along the side of the "tail". Figure 9. the greater the distance from the mode to the mean and the median.

and so we can use the amount of this spread to measure the amount of skewness. except for those with very long tails. Examples of variates with positive skew distributions include size of incomes of a large group of workers. although values below 1 and above 1 are rare and indicate very skewed distributions. The most usual way of doing this is to calculate Pearson's first coefficient of skewness: Pearson's first coefficient of skewness  mean  mode SD However as we have seen the mode is not always easy to find.264 Measures of Dispersion Figure 9. © ABE and RRC . the more spread out are these three measures of location.4: Positions of the mean. When you do the calculation. remember to get the correct sign ( or ) when subtracting the mode or median from the mean. length of service in an organisation and age of a workforce. One such example is the age at death for the adult population in the UK. size of households. and so we can use an equivalent formula for Pearson's second coefficient of skewness: Pearson's second coefficient of skewness  3(mean  median) SD You need to be familiar with these formulae as the basis of calculating the skewness of a distribution (or its "coefficient of skewness". The value of the coefficient of skewness is between 3 and 3. mode and median in skew distributions Frequency Negatively skewed Positively skewed Median Median Mean Mode Mode Mean Variate For most distributions. You will get negative answers from negatively skewed distributions. and positive answers for positively skewed distributions. as it is also called). the following approximate relationship holds: mean  mode  3(mean  median) The more skewed the distribution. Negative skew distributions occur less frequently.

Measures of Dispersion 265 Questions for Practice 1.400 2.360 1.800 1.600 – 4.000 6.200 No.800 4.012 406 0 10.600 3.000 and the details are given below: Loan Amount (£000) 1 2 3 4 5 6 7 8 9 10 11 or more Number of customers 535 211 427 1.400 – 3.800 – 5.000 3.000 customers.400 – 6.400 5. (a) (b) Explain the advantages of using the coefficient of variation as a measure of dispersion.200 4. Explain carefully what is measured by the standard deviation. of borrowers 1 6 9 14 17 23 28 35 39 48 © ABE and RRC .600 6.000 Calculate: (i) (ii) (iii) 2. Calculate the arithmetic mean and standard deviation for the following data using the short-cut method.000 – 6.615 1. The coefficient of variation of the loan amounts.650 1.504 1.600 – 7.280 1. Each loan is a multiple of £1. The arithmetic mean of the loan amounts (£).800 – 2.200 – 4. Income of borrower £ Under 1. The standard deviation of the loan amounts (£). A small finance company has loaned money to 10.000 – 3.

25 cm on a single machine. Calculate for the two distributions the median values and explain any difference between these results and those obtained in (a). The results of a sample of 100 items produced by each operator are as follows: Accuracy of operators Dimension of item (cm) 3.000 5.000 Assets not over (£) 1.000 – Percentages Year 1 48. lower and upper quartile values for Year 1 and Year 11.000 3.2 3.000 10. Production is carried out on each of three shifts. each shift having a different operator – A.000 3.000.24 21 25 22 3.25 36 34 4 3. Write a brief report on the data. Derive by graphical methods the median.28 2 1 0 3. Estimate for each year the number of people whose assets are less than (i) £2.448 Year 11 18.27 6 6 0 3.000 100. It is decided to investigate the accuracy of the operators.000 10.500.9 6.5 10.23 10 7 49 3.2 18.8 25.9 4. B and C.5 0.6 19. and (ii) £25.21 1 0 3 3.000 5.3 17.4 0.000 25.3 29.000 25.266 Measures of Dispersion 3.9 21.29 1 0 0 Operator A B C © ABE and RRC .26 17 27 0 3. using your results where relevant. Items are produced to a target dimension of 3.140 Total number of owners (000s) You are required to: (a) (b) (c) (d) 4.000 100. Consider the following table: Percentages of owners with assets covered by estate duty statistics Assets over (£) – 1.1 11.22 6 0 22 3.4 1.

000 © ABE and RRC .250 9.000  £6.281 5. (a) The coefficient of variation is the standard deviation expressed as a percentage of the arithmetic mean. with profits expressed in different currencies.000 (b) fx 535 422 1.460 9. This is of use when comparing sets of data which have widely differing means.504 1.000 (i) x 60.650 1.960 12.920 9.280 6. say.268 Measures of Dispersion ANSWERS TO QUESTIONS FOR PRACTICE 1. (iii) coefficient of variation  2.615 1. We need to draw up the following table: Loan amount (£000) (x) 1 2 3 4 5 6 7 8 9 10 Number of customers (f) 535 211 427 1. This is particularly useful when dealing.496 51.108 4.024 8.108 6.360 1.650 0 1.280 1.259 to nearest £.376 3.375 3.65% 6.843 5.000 (x – x ) –5 –4 –3 –2 –1 0 1 2 3 4 (x – x )2 25 16 9 4 1 0 1 4 9 16 f(x – x )2 13.060 60.028 (ii) SD  Σf ( x  x ) 2  5. a situation which makes absolute comparisons difficult. It enables the relative variation of two sets of data to be compared.440 8.440 1.012 406 10.1028  2.259  100%  37.000 10.259 thousand pounds Σf  £2. The coefficient of variation has no units and thus the relative variation of sets of data expressed in different units can be compared.

1654  2.2727 class interval units  £1.000 6.200 – £1.300 6.500 * 2. of borrowers (f) 1 6 9 14 17 23 28 35 39 48 220 d = x  xo c fd 6 30 36 42 34 23 0 35 78 144 86 fd2 36 150 144 126 68 23 0 35 156 432 1.000 3.1654 SD  5. We take the deviation of each reading from the mean and square it to remove any negative signs in a mathematically acceptable fashion.200 Mid-value (x) 1.400 5.000 to < 6. It does this by taking every reading into account.1528  5.400 to < 6.400 to < 3.600 6. with xo  £5. We then add all the squared deviations.200 4.700 6.200 to < 4.900 4.170 6 5 4 3 2 1 0 1 2 3 * Assume the first class covers the range £1.800 to < 5.334.800 1. We need to draw up the following table.364 to the nearest £.363.Measures of Dispersion 269 2.800 to < 2.170  86  variance    in class interval units squared 220  220  2  5. The standard deviation measures the dispersion or spread of a set of data.100 + 234. find their mean and finally take the square root. x  £5.600 to < 4.600 3.100 2. © ABE and RRC .600 to 7.700 3.335 to the nearest £ 1.6  £1.000 to < 3.500 5.300 3.100  86  600 220  £5.800 4.400 2.100 5.800 so that its mid-value is £1.900 No.100 and c  £600 Income of borrower £ Under 1.55  £5.500.55  £5.3182 – 0.

000 5.000 3.4 From this.7 94.8 Year 11 18.8 43.1 95.000 Cumulative % frequency (less than) Year 1 48.8 77.0 99.000 10.9 98.270 Measures of Dispersion 3 (a) We need first to draw up the following table Assets over (£) 1. we can draw ogives for both years: Ogive to show cumulative percentage of owners with assets covered by estate duty statistics Cumulative percentage frequency 100 Year 1 90 80 Year 11 70 60 50 40 30 20 10 0 0 5 10 15 20 25 Assets (£000s) © ABE and RRC .000 100.000 25.3 99.3 77.9 55.8 88.

000 Reading off the graph at cumulative percentage frequencies of 25 and 75 we obtain the lower and upper quartile figures: Year 1: lower quartile  £500.000   £3.000 29.5  £1.25  £1.25 Year 11: The 50% mark is in the £3.000 – £5.000 – £3. © ABE and RRC . (b) Year 1: The 50% mark is in the £1.100 Year 11: median  £4. This is to be expected as graphical methods are approximate and are limited by the size of the graph paper.500. Thus.Measures of Dispersion 271 Reading off the graph at a cumulative percentage frequency of 50 we obtain the median: Year 1: median  £1.9 6.21 These figures are more precise than the results in (a).5 1.500.21  £4. (c) Reading from the ogive at a value on the horizontal scale corresponding to £2. 98.3  £2.000 from Year 1 to Year 11.5  19.9  £2.448.025.115.134. upper quartile  £9.025.000  12.000  7. upper quartile  £2.7  £2.800 Year 11: lower quartile  £1.3% of owners had assets less than £25. (d) The total number of owners with assets covered by estate duty statistics rose by 692.500.000.000 was 18.000 11.000   £1. 70% of owners had assets less than £2.500 was: Year 1: 70  18. Therefore the number of those with assets less than £25.183.000  1.500.000  115. 100 Year 11: From the table of values in (a).9  £3.000  50  48. In Year 11 the corresponding percentage was 95%.000 group: median  £3.000  50  43.1  £2.384 in Year 1 and 18.913. we find that in Year 1.000 29.177. we note that in Year 1.000 11.140. the number with assets less than £2. In Year 11 the corresponding figure was 37½%.600 100 37.000 group: median  £1.000 in Year 11.500.