-1-

©2018

ODUNLAMI, A. A.

-2-
 © ODUNLAMI, A. A.

ISBN:- 978-978-50642-6-1

All rights reserved

No part of this book may be reprinted, by any

process whatsoever, without written permission of
the copyright owner.

All correspondence in respect of this publication

must be made with the author through the address
below:

ODUNLAMI, A. A.

S. 52, ODEREMI ABATAN STREET,

ALAGBADO, LAGOS.

Published by

It’s by God’s Grace Publishing Coy.

Ilesa, Osun State.

08033035630, 09031568040, 08056459520

-3-
 ACKNOWLEDGEMENTS

I wish to express my gratitude to God almighty for the

opportunity granted me to complete this book.

Words are not enough to express gratitude to my amiable Head of

Department of Educational Management Venerable L. A. Oni. Moreso
let me use this medium to re-affirm my vow of undying love and affection
to the lecturers in the Department. Mr. Ijila, S. O; Mr. Ogunniyi, J. O;
Mrs. Opeloye and also Dean of Education, Mr. Kolawole, B. K.

I cannot but appreciate other lecturers in the College whose

mentorship role has brought me to where I am. You are indeed a tutor,
facilitator, initiator and mentor whose knowledge impartation cannot be
quantifiable. I say a big thanks to you all.

My indefatigable appreciation to the following Professors in the

Faculty of Education, Ekiti State University, Ado-Ekiti, Ekiti State:
Omirin, M. S; Kolawole, E. B; Ajayi, I. A; Komolafe, C. O; Ayodele, J. B.
whose role model and knowledge acquisition has been the brain behind the
publication of the text. Also to other lecturers’ within and outside the
faculty whose names were not mentioned. I say thanks

To Ambrose, Emmanuel and Precious for type-setting the Manu-

scripts of this text, I say thank you, and Corper Gift for editing, I equally
appreciate your effort.

To my lovely siblings, I say kudos to you all. To my wife, and

children whose immeasurable efforts play a predominant role in the
completion of this book, I say a big thanks to you.

A million thanks to you all.

-4-
 FOREWORD

As I went through the pages of this book, I could

easily conceive a straight forward, but rigorous
introduction to the subject of Educational Statistics and
Models with details for beginner to understand the basic
concepts and techniques of the subject.

The simplicity of the writing style, the day to day

practiced examples and the graded exercises are the
enticing features of the text.

The author is an experienced teacher and a scholar.

He is currently my doctoral students in the Department of
Educational Management, Faculty of Education.

I have no doubt in my mind that this text would be

of significant interest to students in Secondary schools,
Colleges of Education, Polytechnics and Universities.

Obviously, researcher and all users of statistical

techniques will find the book handy.

Professor J. B. Ayodele.
Dean, Faculty of Education,

Ekiti State University, Ado-Ekiti, Ekiti State.

-5-
 TABLE OF CONTENTS

Title page

Acknowledgements

Foreword

CHAPTER ONE:-------------------------------------------- 6

Meaning of statistics

CHAPTER TWO:------------------------------------------- 18

Importance of statistics

CHAPTER THREE:----------------------------------------- 21

Sampling techniques

CHAPTER FOUR:-------------------------------------------26

Central tendency

CHAPTER FIVE:--------------------------------------------42

Quantiles

CHAPTER SIX:---------------------------------------------54

Chi-square

CHAPTER SEVEN:----------------------------------------- 72

School enrollment

CHAPTER EIGHT:----------------------------------------- 86

CHAPTER NINE:------------------------------------------ 92

CHAPTER TEN:-------------------------------------------- 97

INDEX:---------------------------------------------------------- 103
-6-
 CHAPTER ONE

MEANING OF STATISTICS

Numerical facts are called data and the study of

data is called statistics.

Statistics deal with scientific and analytical method

of collecting, organizing, analyzing, and presenting data in
such a way that some meaning and conclusion could be
made out of something that appears to be jungle of data.

Statistics can be defined as the study that involves

in making unintelligible intelligible in the sense that, it is
from the unintelligible mass of figure we derive intelligible
decision which enables us to cope better with life.

Odunlami (2018), define statistics as a systematic

and scientific process for gathering, organizing, analyzing,
and interpreting numerical data before a meaningful
conclusion is drawn.

Types of statistics

Statistics can be broadly classified into two

categories viz descriptive and inferential statistics.

Descriptive statistics refer to the type of statistics

which deals with collecting, organizing, summarizing, and
describing quantitative data. For example, suppose a
mathematics teacher finds the average score of his class.
The average score is a descriptive statistics since (average
score) describes the performance of that class but does
not make any generalization about other classes. Example
of descriptive statistics are graph chart (pie) chart
columnar chart, bar histogram, Pictograph table etc.
-7-
 Other example of descriptive statistic is central
tendency (mode mean median) correlation coefficient
(degree of relationship) kurtosis, skewness etc.

Inferential statistics deals with the method by which

inferences are made to a larger sample on the basis of the
observation made of the smaller sample e.g. suppose a
mathematics teacher decide to use the average score of
one class to estimate the average score of other two or
more classes of the same mathematics course, the process
of this estimation is problem of inferential. In short any
procedure of making generalization that goes beyond the
original data is called inferential statistics.

Inferential statistical provide a way to test the

significance of result obtained when data are collected
example of inferential statistical tools are student t-test,
analysis of variance (ANOVA), analysis of covariance
(ANCOVA), correlational analysis etc.

Data classification

Data can be categorized into the following parts.

- Classification by arrangement

- Classification by source

- Classification by measurement

- Classification by precision

- Classification by number of variance

Classification by arrangement include raw data and array

-8-
 Raw data is the set of unorganized, unarranged and
unprocessed information. Raw data can also be defined
as any piece of information obtained before it is arranged,
analyzed or processed. It is called raw data because it
has not been processed by any statistical method.

Example

2 3 5 6 2 4 5 6 7 8

4 6 8 1 0 4 2 7 6 4

5 6 1 10 2 9 8 6 4 9

Array data are data arranged either in ascending or

descending order of magnitude.

Example

0 1 1 2 2 2 2 3 4 4

4 4 4 5 5 5 6 6 6 6

6 6 7 7 8 8 8 9 9 10

Advantages of array data

- The smallest and highest scores in the data can be

easily seen.

- We can easily obtain the range of the data

- We can see at a glance whether any score appear

more than once in the data

- We can easily divide the score into classes

- We can easily see the trend of the data

-9-
 - We can easily calculate the mode, median and
mean of the data

- Classification by source: Source includes primary

data and secondary data.

Primary data: The term primary data refers to the type of

data originated by the investigator or researcher for the
purpose of problem at hand i.e. data collected by the
researcher himself for a given purpose is called primary
data suppose a researcher collect a data for the purpose
of finding out the relationship between school certificate
and degree certificate of a given set of students, the set of
grades of both school certificate and degree certificate
collected by himself and used for the research work are
primary data.

Finally, primary data are the data collected and used

specially for the purpose for which such data have been
collected.

Secondary data: secondary data are the ones which are

not originated by the investigator or researcher himself
but which he obtain from someone’s record i.e. data taken
from other data are secondary in the above instance. if
the researcher does collect the information himself, he
collect the grade from the examination / record office of
the student, then the data collected in this case is called
secondary data.

Primary Vs Secondary Data

The term primary and secondary data are relative in the

sense that data that are primary to one person may be
secondary to the other. For example data collected during
- 10 -
the grade II examination by institute are primary to
National Teacher Institute Department (NTI) but, to a
person who uses those data for further research, it is
termed as secondary data.

- Classification by measurement: It can be classified

into qualitative and quantitative data.

Quantitative data: This data are recorded on a naturally

occurring numerical scale. The following are example of
quantitative data

A) The weight of 100 students in a class

B) The ages of a set of teacher

c) The height of certain trees in a forest reserve

d) The scores of a sample of 100 testees.

e) The volume of water in a swimming pool

f) The number of civil servant in a state

Qualitative data: Qualitative data are measurement that

cannot be measured on a natural scale; they can only be
classified into one of the following categories. Example of
qualitative data is (i) beauty (ii) attractiveness (iii) the
political parties in a particular country (v) the species of a
plant.

- Classification by precision: variable is any quantity

that may take more than one value within a given
context. In other words, a values, called the domain
of the variable.

- 11 -
 Suppose a class teacher collects the ages of the
students in his class, then the set of ages is a variable
since their ages vary from one member to another.

A constant is any quantity which does not change

as value within a given context, examples of constant are
∏ = 22∕7’ e= 2.7182.

In a given equation y = 4x + 3,

Y and x are variables while 3 is a constant.

Data can also be classified by preciseness, as discrete

data and continuous data.

Discrete data can be described precisely as one way of

obtaining data by counting. A discrete data is the quantity
that assumes an integral value. Discrete data are always
expressed in whole numbers. Discrete data are those ones
that can be counted. Example is:

(i) The number of lecture theatres in a university.

(ii) The number of theatres could be none, 1,10, 20

and so on but, it could neither be 11∕2 nor23/4

(iii) The number of houses in a given town which can

assume any of the values 0,1,2,3,…. But cannot be 1.5 or
1001/2the number of houses is a discrete data.

(iv) The number of motor cars in a car assembly can only

take integral value such as 0,1,2,3,4,…10...15..100, thus
the number of cars in an assembly is a discrete data.
(v) the number of children in a family it could be
0,1,2,3….10 and so on. Discrete data can also be

- 12 -
obtained from situation where counting is not involve.
Example.

Shoe size of a set of people 10, 42, 6, 31/28,4

Shirt size of a set of men 13 ,15,1/2,14 ,61/2, 20

Bed size 31/2, 5, 5,1/2 6 1/2

A particular characteristics of discrete data is the fact that

possible data values progress in definite step like bed size
are measured as 21/2,31/24,5,7,12and so on or there are
1,2,3,4,10……..cars in a motor park (not 11/2,31/2, 10,27)

Continuous data is a data which can take an infinite

number of values between any two points on the scale. In
otherwords, they cannot be measured precisely their
values can only be approximated. Continuous data is the
type of data that does progress from point to the next
without a break they involve numerical measurement
such as weight, height, volume, pressure, temperature,
age at times etc. Continuous data can be expressed as
decimal or fraction or whole numbers.

Worked example

State which of the following represent continuous

and discrete data

A number of children within a family.

B the life span of a fish.

C height of a set of student

D weight of iron sheets

- 13 -
E number of towns in a country.

Solution

(a) Discrete (b) continuous (c) continuous

(d) Continuous (e) discrete

Give the domain of each of the following variable and

state whether the variable is discrete or continuous

Solution

Number N of children in a class.

Domain – any integer from o to the number N is in the

class

(b)number V of litres of petrol in a can.

Solution

Domain – any number value starting from o to the

capacity of the can

Variable –continuous

(C) state X in the old western region of Nigeria

Solution

Domain – Oyo, Ondo, Ogun, Lagos , Edo, Ekiti, Delta, and

Osun variable – discrete

(D) radius r of a circle start from zero assuming a point to

be a circle)to any value of the radius r.

(E)number of K of cars in assembly

- 14 -
Solution

Domain – K takes any values 01,2,3,…,…….

Variable – discrete

EXERCISE

Express each of the following in numerical form (domain)

and state whether it is discrete or continuous

1. Number of lecturers in mathematic department in a

college.

2. Number of headquarters in Isokan local

government area in Ondo state Nigeria

3. Normal set of teeth of a man.

4. Life span of human beings

5. Number of pages of book

6. Duration of examination

7. Number of state in Nigeria

8. Volume of sphere

9. Ages of animals

10. Number of rivers all over the world

Observation is the value of a variable for a member

of a population

Parameter is a characteristic of a population which

helps to summaries information about the population with
- 15 -
regard to the variable under study. Some of the common
parameter is measures of location and measure of
dispersion.

A statistic is a characteristic of a sample

Collecting data once you decide on the type of data

appropriate (quantitative or qualitative) for the problem at
hand there is a need to collect data you can obtain data in
four different ways

1 data from a designed experiment

2 data collected observationally

3 data from a publish source

4 data from a survey.

Data from a designed experiment:- This collection involves

conducting a designed experiment in which the researcher
exert strict control over unit (object or thing or people). In
the study, experimental design is made up of treatment
/experimental group and control group.

Data collected observationally: in observational study,

the researcher observes the experimental unit in their
natural setting and records the variable(s) of interest.

Data from a published source: This is done by

collecting data from a published source(s) such as books,
journals or newspapers, sporting, news, statistics,
abstract, annual abstract of statistic, monthly digit of
statistics, financial statistics, economic trends and regional
trends.

- 16 -
Data from survey: with a survey, the researcher
samples a group of people, asks one or more question and
records the responses.

Methods of collecting data: The method of collection

statistical data may be classified as primary method and
secondary method.

The primary methods include:

- Direct personal interview

- Indirect personal interview

- Information from correspondent

- Questionnaire to be filled by the enumerator

Direct personnel interview this is the method whereby the

researcher or his agent collect the data through personal
contact this is the best method and preferable method
since he (researcher) has personal contact with every
member of the population used for the research this
method reduces the chances of incorrect data being
recorded even though this method is the most reliable but
most expensive and time consuming. It also involves a lot
of experience on the part of a researcher to interact well
with his clients in asking personally for his required
information.

Indirect personnel interview:- This is whereby the

researcher ask his agent to interview his sample on his
behalf

- 17 -
The disadvantage of interviewing is that inaccurate or
false data may be given to the interviewer the reason may
be (1) forgetfulness (2) misunderstanding of concepts.

Questionnaire:- This is the quickest and easiest method of

gathering from large and widely scattered members of the
population. Questionnaire may be personally delivered to
each members of the population under investigation or
otherwise the questionnaire should be accomplished with
self-addressed envelope with a stamp fixed to it because
the respondent may not return it.

EXERCISE

Distinguish between

1. Primary and Secondary data

2. Descriptive and Inferential statistics

3. Observation and Parameters

4. Interview and Questionnaire

5. Give three examples each of raw and array data.

- 18 -
 CHAPTER TWO

IMPORTANCE OF STATISTICS

Before any statistical work could be done data must

be collected, the collection of data is a very important part
of statistics any mistake, error and bias which arise in
collection of data will automatically affect our conclusion
and decision. The following are some of the importance of
statistics:

- A knowledge of statistics is an essential aspect of

the training all students of educational planning /
educational management must be acquainted with. In
Nigeria, Federal and State government have been
devoting significant share of their budget to cater for the
expansion in the education sector. The role of statistics in
ensuring efficiency in the educational system towards the
effective pursuance of educational goals is of utmost
importance

- Statistics enable educational managers to predict

intelligently how much of a thing do we have under a pre-
determine conditions.

- It also enables educational managers to draw

general conclusions and inference about the phenomenon
they are dealing with and the extends to which such
conclusion can be generalized.

- Besides, quantifiable (quantitative) data are

required on the input and output of the educational
system while non-quantifiable (qualitative) data are
required in the process of the system.

- 19 -
- Students input: - To access educational goals and
development, statistics are needed to take stock of
student’s population in schools at different levels of
educational system and the flow of students to determine
their progressive and the efficiency of the system.

- Staff input: - Teaching and non-teaching staff are

needed in adequate number (quantity) and staff (quality)
in the educational system. Educational managers must
know what is in stock and what is required.

- Physical and materials resources input: - physical

facilities in terms of land, building, equipment, machines
and consumables materials are needed in the educational
system. Statistics are needed to determine this for their
procurement and for use adequately.

- Financial resources input: - Adequate financing of

education demands that there should be adequate facts
and figures about sources of income and aspect of
educational expenditures.

INPUT----------------PROCESS--------------------OUTPUT

Input: - Students input, staff input: - How many students

are we taking care of, do we need teachers, how many do
we need.

Process: statistics are required for formulating policy and

standard operating control and monitoring services in the
educational system.

Output: Statistics are needed to access the quantity and

quality of the product of the educational system.

- 20 -
Employer’s opinion about the product in different worlds of
life is also statistics about the product.

Problems of data collection in Nigeria

Despite the enormous benefit of the knowledge of

statistics to educational management and manager. It has
been very difficult (if not impossible collecting adequate
data for planning in Nigeria).

- Data collection at any level is very expensive

- There are problems emanating from government

officials; executives or people in high income
groups are not easily accessible or approachable
under the pretense of official secret or burecracy.
This will no doubt hinder accurate and correct data
which will in-turn affect the expected outcome.

- In Nigeria the high rate of obsolete of data

(outdated) how accurate is the data needed for
national planning or historical perspective of Nigeria
settings. How many of our museum or document in
most Archie’s are correct.

- Level of illiteracy is another impediment to data

collection in the country. At times it is very difficult
in getting information from illiterate people even
the ones release may be incorrect

- At times researchers may seat in his office or at

home and give report to a problem without
recovered to data collection

- 21 -
 CHAPTER THREE

SAMPLING TECHNIQUES

In most cases it is very difficult to have contact with

every member of the population so it is best to choose a
good respective sample of the population. The method of
selecting part of the population is called sample

What is Population?

This can be described as the entire element of

subjects in an area of study. The elements may be in-
animates or animates, provided they fall within the area of
study.

Sample: This is an integral part of a subject of the

population. population

Sample

Characteristics of a good sample

- A good sample must be representatives of the

population i.e. all segments in the population must be
represented.

- A good sample must take mathematics analysis

easy i.e. it must not be lopsided e.g. if you want to pick a
sample of 100, we must not make it 90 male and 10
female. If we do that such sample is already lopsided.

- A good sample must be reasonably large i.e. the

larger the population, the larger the sample.
- 22 -
- A good sample must not be biased. In otherwords
selection must be objective.

- A good sample must be accessible.

Sampling Techniques

In selecting our sample for the population one or

more of the following sampling techniques must be
considered.

- Simple Random Sampling Techniques:- This is a

method of choosing our sample from the population
without being biased, whereby every member of
the population has equal right of been chosen /
picked i.e. there is no partiality and the selection is
purely objectives.

This can be done in 3 ways

(a) By balloting

(b) By use of a computer

(c) Use of random number (Raffle draw)

This method has major disadvantages of being lopsided.

- Stratified Random Sampling Techniques:- To solve

the problem of being loop sided or a section not
been touched at all. This is a techniques whereby
the population is divided into different
heterogeneous group that are available and from
each heterogeneous group, simple random
sampling techniques is used (e.g. separate Islam,
Christian, Pagan etc.)
- 23 -
 Disadvantage:- There could be more elements in
one group or the other e.g. in a class of 20. There
are 18 Christians and 2 Muslim. This is lopsided
already.

- Proportional Stratified Random Sampling

Techniques:- The population is divided into
different heterogeneous of group in which the
sample to be choosing would depend on the
number of subjects in that sub-group. It would be
proportional e.g. ratio A to B – C. You say total
ratio = no picked from each group.

- Multi-Stage Sampling Techniques:- This is a

technique whereby choosing our subject will involve
more than one stage e.g. we may select some
states from the Federation and from those selected
states. We may also select some local government
and from those local government selected some
schools may be selected. It takes more than one
stage before you reach your subject.

- Cluster Sampling Techniques:- This is similar to

stratified random sampling techniques but the
subjects are homogenous i.e. they have the same
characteristics. The first thing we do is to divide
the population into a number of strata (sub-groups)
e.g. all the students here are B.Ed students. I may
divide you into group since I am teaching the same
group. Once I ask one student where do we stop? I
do not need to doubt him because he falls within
the same group I am teaching while in stratified
they do not have anything in common. If they are

- 24 -
 20 sub-group. You must visit all the groups because
they are heterogeneous groups.

- Snow-balling Sampling Techniques:- This is a

technique where you start with a small group or
even an individual who is a volunteer and he would
be bringing other members. If a revival is organized
and the first day there are 2 to 3 people and
wonders happen the next day. The population
increases on daily basis. If a student is caught for
examination mal-practice and the student caught
say he is not the only one. It is called snow-baling.
Take a stick of matches and strike it, it can burn the
whole house. It starts from a very small point
before expanding gradually.

- Systematic Sampling Techniques:- This is a

technique whereby the sample is selected at a
regular and constant interval. It could be a multiple
of N. For instance if your no is multiple of 5 e.g. 5,
10, 15, 20, 25, 30 etc. It must be at regular
interval.

- Quota Sampling Techniques:- Here, every segment

is represented either qualify or not qualify. The
New P. D. P. chairman if not for quota, he would
not have been choosing.

- Purposeful / qualify Sampling Techniques:- This is a

type of techniques whereby any subject selected
must have a reason and be for a particular purpose.
If you put anybody there, you must be able to give
reasons why you have choosing such a person. For
instance you want to go and pray for a classmate

- 25 -
 who is in hospital and he is a Muslim. You must pick
one person who is a Muslim among those that are
going for such visitation.

EXERCISE

1a Define data.

1b List and discuss various classification of data.

1c. Mention and discuss method of collecting data.

2a. Explain how population and sample differ.

2b.Mention and discuss various method of sampling

techniques.

- 26 -
 CHAPTER FOUR

CENTRAL TENDENCY

An average or measure of central tendency is a

typical or representative of a set of data, since such
representative score tends to lie centrally within the array
of data.

Types of Central Tendency

The commonest types of average are (a) mode (b)

median (c) mean

Find the range, mean, median and mode of the following

ungrouped data.

1,3,5,3,3,6,7,8,1,3,9

Range = Highest – Lowest number

Since 9 is the highest number and 1 is the lowest number.

Therefore, range = 9 – 1 = 8

To find mean which is otherwise called average

Addition of all the number divided by the number of

appearance.

1+3+5+3+3+6+7+8+1+3+9

11

Mean =49/11 = 4.45

Properties of Mean

- 27 -
 1. The value of the mean is determined by every score
in the series.

2. It is greatly affected by extreme values.

3. The sum of the deviations about the mean is zero.

4. The sum of the squares of deviations from the

mean is less than those computed from any other
score.

5. The mean reflects every score in the distribution.

6. The mean is shifting towards where ever there is a

change.

Median= to find the median of any number, it must be re-

arrange either through ascending or descending order.

Properties of Median

1. Median is a positional average. It is not influenced

by the size of items but by the position of the
items.

2. If the median is less than the mean, the distribution

is skewed towards right (positively skewed). But if
the median is greater than the mean, the
distribution is skewed towards the left (negatively
skewed, if the mean =median = mode, then the
distribution is symmetrical.

3. The sum of the absolute values of the deviations is

at least from the deviation are measured from the
median.

1,1,3,3,3/,3,/5,6,7,8,9
- 28 -
Median = 3 since 3 is the half way of 11 and when it
involves even number the median would be two e.g.
2,2,3,/5,6,/7,8,9. The median is 5 and 6 divided by 2.

Mode on the otherhand would be the number that occurs

most

Mode = 3 since three occur 4 times.

Importance of Mode

1. The mode is the mode descriptive average since it

signifies the most typical value in the given set, and
indicates the precise value of an important part of
it.

2. The mode is not affected by the extreme; hence, it

will be a more representative average of many
purposes.
GROUPED DATA

Class Interval F X FX

11-15 2 13 26

16-20 1 18 18

21-25 4 23 92

26-30 2 28 56

31-35 1 33 33

10 225

X ∑FX/∑F =225/10 Mean = 225

- 29 -
Let assume mean be preferable to the one around the middle.
Anytime you see d Assume mean must be there.

Class Interval F X d Fd

11-15 2 13 -5 -10

16-20 1 18 0 0

21-25 4 23 5 20

26-30 2 28 10 20

31-35 1 33 15 15

10 45

Let assume mean = 18

Let: A.M. + ∑Fd/∑F

18 + 45/10

18+4.5

= 22.5

ALTER MEAN OR ALTERNATIVE MEAN

Class Interval F X X/i FXi

11-15 2 13 6.5 13

16-20 1 18 9 9

21-25 4 23 11.5 46

26-30 2 28 14 28

31-35 1 33 16.5 16.5

10 112.5

- 30 -
I = 2 x =i

Mean = what you would have use to divide use it to multiply.

X i∑FX/∑F

= 2 (112.5)/10

X = 22.5

Let’s change our alter to 5

Class Interval F X X/i FXi

11-15 2 13 2.6 5.2

16-20 1 18 3.6 3.6

21-25 4 23 4.6 18.4

26-30 2 28 5.6 11.2

31-35 1 33 6.6 6.6

10 45

I = 5 x =i

X i∑FX/∑F

= 5 (45)/10

X = 22.5

MEDIAN

Class Interval F FX

11-15 2 2

16-20 1 3

- 31 -
21-25 4 7

26-30 2 9

31-35 1 10

10

Size of median is obtained by N/2

Median: It means middle or half way

Where L1 =Lower boundary of the median class

N =Total frequency

C =median class size

Fw =median class frequency

Cfb =cumulative frequency of all the classes

lower than the median class.
Med. L1 + (N/2 –Cfb) Class interval

Fw

20.5 + (10/2 -2) x 5

20.5 + (2/4) x 5

20.5 + 2.5

Median = 23

ALTER OR ALTERNATIVELY WAY OF GETTING MEDIAN

Class Interval F CF

11-15 2 2
- 32 -
16-20 1 3

21-25 4 7

26-30 2 9

31-35 1 10

10

3 and 7 which one is greater 7. That is why we pick 7 for

Median.

Mean = N/2 = 10/2 = 5

Inferimo, lower boundaries, lower limit, superimo, higher limit,

higher boundaries.

ALTER ALTERNATIVE WAY OF FINDING MEDIAN

Median = L1+ L2 - L1(M – Cfb)

Fw

20.5 + 25.5 – 20.5 ( 5 – 3)

20.5 + 5/4 (2)

= 20.5 + 2.5 = 23

MODE

Class Interval F CF

11-15 2 2

16-20 1 3

21-25 4 7
- 33 -
26-30 2 9

31-35 1 10

10

Mode = L1 = (D1) x i

(D1 + D2)

= 20.5 + (4-1) x 5

(4-1) + (4-2)

20.5 + (3) x 5

(3+2)

20.5+ (3/5) x 5

20.5 + 3

23.5

ALTER OR ALTERNATIVE WAY OF GETTING MODE

Class Interval F CF
11-15 2 2
16-20 1 3
21-25 4 7
26-30 2 9
31-35 1 10
10

- 34 -
Mode = L1 +F1 – F0
(2F1 – F0 - F2) x i
= 20.5 + (4 – 1)
2(4) – 1 – 2
20.5 +(3)

(8-3)
20.5 + (3/5)5
20.5+ 3
= 23.5

Where

L2 =upper boundaries of the modal class.

L1 =lower boundaries of the modal class.

C =class size of the modal L2 – L1.

F0 =frequency of the immediate lower class to the modal

class.

F1=frequency of the modal class.

F2=frequency of the immediate higher class is the modal

class.

Estimate the mode of the distribution

Class interval 0-4 5-9 10-14 15-19 20-24 20-24 25-29 30-34

Frequency 1 2 3 5 4 2 2 1

- 35 -
Solution

Class interval frequency

0-4 1

5-9 2

10-14 3 f0

15-19 5 f3

20-24 4 f2

25-29 2

30-34 1

Modal class = 15-19

L1 = 14.5 (2) L2 = 19.5 (3) C=L2-L1=19.5-14.5 = 5 (4) f0 = 3

(5) f1 = 5 (6) f2 = 4
Mode = L1 + (f1 – f0 ) C
F1 - f2 + f1 - f0
= 14.5+ (5-3)5
(5-3)+(5-4)
= 14.5 + 10
2+1
= 14.5 + 10/3 =14.5 + 3.333 = 17.833

Skewness

Skewness is the degree of asymmetry i.e. the existent to

which a distribution deviate from the normal.

- 36 -
Types of Skewness

There are two types of skewness positive skewness and the

negative skewness. A distribution is said to be positively skewed
when the tail of the distribution is towards the right. For
example if you are given any distribution in which the mode is
smaller. You have positive skewness and the answer is bad.

+ ve

Mode 20 mean 60 median 70

It shows that the performance is poor.

Positive skewness occurs whereby a test is too difficult. The

scores may look like the figure on the right with little difference
among the poorer students. While Negative skewness is a type
of skewness whereby the tail of the distribution is towards the
left.

-ve

mean 30 median 40 mode 70

If the answer for mode is bigger than mean, median. The

answer is very good. It shows that the performance is good.

Negatively skewness occurs when a test is too easy. The

score may look like figure on the left with no appreciative
difference among the best students.

Two levels of Skewness

Sk = Mean – Mode ---------------------(1)

SD
- 37 -
Sk = 3 ( mean – median) -----------------(2)

Complete (1) the coefficient of skewness for each of the

following and give literary interpretation to your answer.

Mode = 70, Mean =80 standard deviation = 20

Median = 35, Mean =35, standard deviation = 10

Mode = 17, Mean =12, Median =15

Median =13, Mean = 20, Mode = 15

Solution

Coefficient of skewness = mean – mode

= 80 -70 = 10

20 20 =0.5

Coefficient of skewness = 3 (mean – median)

= 3(35 – 45) = 3(-10) = -3

10 10

Standard Deviation

Class interval F x Fx (x-x)2 F(x-x)2

11-15 2 13 26 90.25 180.5
16-20 1 18 18 20.25 20.25
21-25 4 23 92 0.25 1
26-30 2 28 56 30.25 60.25
31-35 1 33 33 110.25 110.25
10 225 372.5
x = Σfx
Σf
X = 225

- 38 -
 10
X =22.5
(x - x) =
13 – 225
-9.52
90.25
Variance = Σf (x - x)2
Σf
= 372.5
10
Variance = 372.5
S.D = √Σf (x-x)2
Σf
= 372.5
10
= 37.25
S.D = 6.1032
S.D = 6.1
EXAMPLE
1. In a moderately asymmetrical distribution the mode and
mean are 25.6 and 36.5 respectively. What is the
median?

2. In a moderately skewed distribution, the mean and

median re respectively 21 and 20.6. Calculate the mean.

3. If the mean and median of an asymmetrical distribution

are 23.5 cm and 24 cm respectively. Estimate the mode.

(1)Given mode = 25.6 mean = 36.5, median?

Mean – Median = 1/3 (mean – mode)

36.5 – median = 1/3 (36.5-25.6)

- 39 -
 = 1/3 (10.9) = 3.63

Median = 36.5 -3.63 = 32.87

(2) Mode = 21, median = 20.6 and mean?

Mean = ½ (3 median-mode)

=1/2 (2 x 20.6) – 21

½ (61.8 -21)

½ (40.8)
=20.4

(3) Mean = 23.5, median = 24.0, mode?

Mode = 3 median – 2 mean

3 (24,0)- 2(23.5)

72 – 47 = 25cm

Kurtosis

Kurtosis of a distribution is the peakness or otherwise of a

distribution.

There are three main type of kurtosis vis – lepokurtic and

platkurtic.

Leptokurtic – highly peaked (lepo-thin)

Mesokurtics – moderately peaked

Platykurtic – flattened (plat – flat)

- 40 -
 Lep

Meso Plat

Lep F

T
Score value

In the fig. above represents two curves with similar central

tendencies but different kurtosis Curve T is more peaked that F
and the change in height of T is more rapid than that of F as
the score value increases. Distribution T is more leptokurtic
(thinner) than B. On the otherhand, F is said to be more
platykurtic (flatter) than T. Percentile coefficient of kurtosis =
½ (Q3 – Q1)

P90-P10

Where Q1 = Lower quartile

Q3 = Upper quartile

P90 = 90th percentile

p10 = 10th percentile

- 41 -
EXERCISE

Class Interval F

10000 - 11999 12

12000 - 13999 14

14000 – 15999 24

16000 – 17999 15

18000 – 19999 13

20000 - 21999 7

22000 – 23999 6

24000 – 25999 4

26000 – 27999 3

28000 – 29999 2

Find the mean, assume mean, median, alter median, mode,

alter mode, standard deviation and skewness.

- 42 -
 CHAPTER FIVE

QUANTILES

Quantiles are a point value which divides the set of

observation into two groups with known proportions in
each group. Examples of quantiles are deciles, percentiles
etc.

Quartiles are the three point values (Q1, Q2, Q3)

which divide an array data into four equal parts.

Q1 known as the first quartile is a point value which divides

the distribution into ratio 1:3 (25% or ¼) of the items are
below and 75% = ¾ of the items are above).

Q2 known as the median is a point value that divides the

distribution based on the frequency into two equal parts.

Q3 known as the upper quartile is a point value that

divides the distribution based on the frequency into ratio
3:1 (75% = ¾ of the items are below the point and
25%= ¼ are above it).

Example 1

Compute the first middle and 3rd quartiles of 2, 3, 5, 8, 9,

8, 15, 10, 7

Solution

Array 2 3 5 7 8 8 9 10 15

Rank 1st 2nd 3nd 4th 5th 6th 7th 8th 9th

- 43 -
1st quartile = lower quartile = (N+1)th = (9+1)th

(4) (4)

=2.5th = 3 + 5 = 4

2nd quartile = median =(N+1)th =(9+1)th

(2) (2)

= 5th =8

3rdquartile = upper quartile = ¾(N+1)th = ¾(10)

= 5th = 8

2rd quartile = upper qua tile = ¾ (N + 1)th =3/4(10)

= 7.5 th = 9 + 10 = 9.5

Example

Compute the lower quartile, median and the upper quartile

of the following distribution.

Score 11 12 13 14 15 16 17 18

Frequency 1 2 3 4 4 3 2 1

- 44 -
Solution

Score frequency Cumulative frequency Ranks

11 1 1 1

12 2 3 2-3

13 3 6 4-6

14 4 10 7-10

15 4 14 11-14

16 3 17 15-17

17 2 19 18-19

18 1 20 20

N=20

Lower quartile = Q1 = 1/4(N+1)th score

= 1/4 (20+1) = 51/4th score = 13

Median = Q2 = 1/2(N+1)th score

= 1/2 (20+1) = 101/2th score = 14.5

3
Upper quartile = Q3 = /4 (N+1)th score

=3/4(21) = 153/4th score = 16

Quartiles for grouped data

Q1 = L1 + N - CFQ C

4 fQ1
- 45 -
Where I= 1, 2, 3

L1 = Lower boundary of the lower quartile class.

CFQi = Cumulative frequency up to the lower class next to

the Q class

FQi = frequency of the Q1 class.

C = class size of the Q1 class.

Example

Find Q1, Q3 semi-interquartile, inter-quartile, Decile 7 and

77th percentile.

Class interval 10-20, 20-30, 30-40, 40-50, 50-60, 60-70, 70-80, 80-90, 90-100

Frequency 2 4 7 10 6 5 3 2 1

Class interval frequency CF Ranks

10-20 2 2 1st -2nd

20-30 4 6cfQ 3nd- 6th

30-40 7 13 7th- 13th Q1=10th

40-50 10 23 14th -23th Q2 = 29th

50-60 6 29 24th-29th

60-70 5 34 30th-34th Q3 =30th

70-80 3 37 35th-37th

80-90 2 39 38th-39th

90-100 1 40 40th
- 46 -
Lower quartile
Q1 = L1 +(N/4 – Cfb) x 10

Fw

30.5 +(40/4 – 6) x 10

30.5 + (10 – 6) x 10

30.5 + (4) x 10

30.5 + 5.71 = 36.2 q1

Q3 = L1 +(3N/4 – Cfb) x 10

Fw

60.5 + (3x40 – 29) x 10

60.5 + (30 – 29) x 10

60.5 + (1) x 10

60.5 + 10/5

60.5 + 2 = 62.5 q3

- 47 -
Inter-quartile range = Q3 – Q1 = 62.5 – 36.2 Interquartile =
26.3

Semi-interquartile range = Q3 – Q1 = 62.5 – 36.2 26.3

2 2 2

Semi-interquartile range =13.15

D7 = L1 +(7N/10 – Cfb) x 10

Fw

L1 + (7 x 40/10 – Cfb) x 10

50.5 + (28 – 23) x 10

50.5 + (5) x 10

50.5 + 50/6

50.5 + 8.333

D7 = 58.8

P77 = L1 +(77N/100 – Cfb) x 10

Fw

L1 + (7 x 40/100 – Cfb) x 10

60.5 + (34 – 29) x 10

- 48 -
60.5 + (5) x 10

60.5+ 50/5

60.5 + 10

P77=70.5

(1) Inter-quartile range is the positive difference

between the upper and lower quartiles.

(2) semi inter-quartile range or quartile deviation is the

half the difference between the lower and upper quartiles.

EXERCISE

The frequency distribution of marks in an examination was

as follows:

Class interval 1-5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-51

Frequency 2 3 5 10 12 13 9 6 3 1

Find q1, q3, inter-quartile range, semi-interquartile range,

D9 and 85th percentile.

Deciles

Deciles are the nine point value which divides the array set
of item into ten equal parts. The deciles are D1,D2,……..D9

D1 is a point value which divides the distribution into ratio

1:9, 10% of

the distribution lies below it and 90% lie above the point
D1
- 49 -
N.B.

Q2 =D5 = median

Percentiles are the ninety point value which divide the

distribution into 100 equal parts.(p1,p2…….p99).

Quartiles are the four point values which divide the

distribution to 5 equal parts.

Qctiles are the seven point values which divide the

distribution into eight equal parts.

Quartiles formula for ungrouped data

It should be noted that the location is based on the

distribution frequency

Q1 = (N+1)th score

4 =Lower quartile

D2 = (N+1)th score

2 =median
Q3= (N+1)th score =upper quartile

D1 =1 (N+1)th score

10

D7 = 7 (N +1)th score

10

D51 =51 (N+1)th 14.7th =14.41

100 = 14 + (0.41) = 14.41

- 50 -
Example

Find the first deciles, 7th deciles, 21st percentile 3rd octile
of the following

distribution.
12,13,14,16,18,17,19,11,23,20,22,24,25,26,28,27,29,30,15.

Solution

Mark 11 12 13 14 15 16 17 18 19 20 21 22 23

Rank 1th, 2th, 3th, 4th, 5th, 6th, 7th, 8th,9th ,10th ,11th ,12th ,13th

Mark 24 25 26 27 28 29 30

Rank 14th 15th 16th 17th 18th 19th 20th

D1 = 1 (N+1)th = 21th

10 =12 + 1 (13 – 12)

10 =12.1

D7 = 7 (N+1)th =14.7th

10 = 24 + 7 (25 - 24)

10 =24.7

P21 = 21 (N+1)th = 14.7th =14.41

100 = 14 + (0.41) 1 +14.41

Example

The below was the age distribution of United Kingdom

population in hundred thousand at June 30, 1962.

- 51 -
10-14 119

15-29 99

30-44 108

45-59 102

60-74 62

75-89 21

Calculate the D1,D3,P3,5th Octile 3rd Quintile.

Solution

Class Frequency Cf Rank

interval
10-14 119 119 1 – 119
15-29 99 218 120 -218
30-44 108 326 219 – 326
45-59 102 428 327 – 428
60-75 64 492 429 – 492
75-89 21 513 493 – 513
D1=1st Decile, N = 513, N = 513/10 =
51.3

D1 class =10-14, fD1 = 119, Cfb=O

L1=0.5,C=15
D1 =L1 + (N/10 –cfD1)C

fD3

= -0.5 + (51.3-0)15

119
- 52 -
 = -0.5+6.47

=5.97

N = 513, D3 = 3N = 3(513) = 153.9

10 10

D3 class = 15-29

L1 = 14.5, 3N = 143.9, cfD3 = 119, C =

10

fD3 = 99

D3 = L1 + (3N/10-cfD3)C
fD3

=1 4.5 + (153.9-119)15
99

=14.5 + (34.9)15
99

=145 +5.29 =19.79

3rd percentile

N = 513, 3Nth = 3 x 513

100 100

P3 class = 0 -14, L1 = -0.5,cfP3 = 0, C = 15

P3 =0.5 + (15.39)15

119

= 0.5 + 1.94 = 2.44

- 53 -
5th Octile

N = 513, 5Nth = 5(513)th

8 8

=320.625th

5th Octile class = 320.625th = 30 – 44

L1 = 29.5, 5N = 320.625. cfQ3 = 218, C = 15

5th Octile = L1 + (5N/8 –cfD3)15

180

= 29.5 + (320.625 -218)15

180

= 29.5 + 14.25 = 43.75

3rd Quintile

N = 513, 3Nth = 3 x 512th

5 5

=307.8th

3rd Quintile Class 307.8th = 30-44

3rd Quintile = L1 + (3N/5 – CFQ)15

180

= 29.5 + (307.8-218)15

180

=29.5 + 12.47 + 41.97

- 54 -
 CHAPTER SIX

CHI-SQUARE

It is a degree of relationship which involve frequency

counts

If calculate or critical value is greater than the table value.

It shows that the result is significant which contradiction
to the null-hypothesis is. The null-hypothesis would be
rejected.

What is hypothesis?

Hypothesis can be defined as a tentative statement or a

congestional statement connecting 2 or more variables.
Some call it a guess, or a wise guess. There are 2 types

(1) Null hypothesis

(2) Alternative hypothesis or directional hypothesis

The null hypothesis is denoted by Ho it is an hypothesis

of equality i.e. it is not partial e.g. there is No significant
difference between the performance of male and female
student in mathematics.

The Alternate hypothesis is denoted by Hi which

says, there is a significant difference between the
performance of male and female students.

- 55 -
 I drink beverage, if you don’t see 7up then you can
bring coke.

While alternate hypothesis If I want ponded yam

and you are bringing rice. Then you are a fool.

Language of Hypothesis

The language of hypothesis would help us to know

the data and the statically tools to use, e.g. is it chi-
square, regression e.t.c. And it would enable us to know
the type of conclusion to draw.

When we are looking for relationship between 2

variables, we can either use chi-square or correlation but
when it involves frequency count. It must be chi-square
e.g. sex and smoking habit.

But if X2 chi-square calculated is less than X2 square table.

It shows that the result is not significant which is in
agreement with your null hypothesis will be accepted.

Xc2 = if Xc2 calculate value is greater than XtC it shows that

the result is significant result which is against null
hypothesis: reject null hypothesis

The change there is that we don’t recognize negative. We

discard it and accept positive no

e.g tc = -3.4

tt = 1.5
- 56 -
Disregard -3.4 and take 3.4.

Smoking Habit

Sex Smoking Non-smoking Total

M 3(3.3) 2(1.7) 5

F 7(6.7) 3(3.3) 10

Total 10 5 15

RW x CT = 5 x 10 = 50
GT 15 15
Fe = RT x GT = 5 x 5 = 25
GT 15 15

10 x 10 = 100
15 15

Calculate

Xc2 = ∑ (fo - fe)

Fe
= (3-3.3)2 + (2-1.7)2 + (7-6)2 + (3-3.3)
3.3 1.7 6.7 3.3

= (0.3)2 + (0.3)2 + (0.3)2 + (-0.3)2

3.3 1.7 6.7 3.3

= 0.09 + 0.09 + 0.09 + 0.09

3.3 1.7 6.7 3.3
= 0.027 + 0.053 + 0.013 + 0.027

Xc2 = 0.12 ans

Degree of freedom = raw-1 and column -1 (r - 1) x (c - 1)

- 57 -
Some limitations of chi-square
1. We can only use chi-square neither in frequency
data nor for scale score.
2. No expected value in each cell must be less than
5 (in this case we use yate’s correlation).
3. The sum of the expected frequency must be
equal to the sum of the observed frequencies.
4. The frequencies must be independent (mutually
exclusive).
5. When df is 1, we apply the correlation continuity.
EXERCISE

Assume that we want to find out whether an

anxiety and social interaction are related.
Anxiety
No Low Medium High Total
Interaction 3 4 8 10 25
No 9 11 8 3 31
interaction
Total 12 15 16 13 56
(1) state the hypothesis
(2) Compute the chi-square the increase in
anxiety leads to increase in social interaction.
That is, there significant different between
anxiety and social interaction?

- 58 -
Student T-test

Student’s t-distribution is the statistic in calculating the

probability associated with Ho. The t is a statistic generally
applicable to a normally distributed Radom variable where
the mean is known (or as we shall see, assumed to be
known) and the population variance is estimated from a
sample.

X Y x2 y2

5 3 25 9

1 4 1 16

2 3 4 9

4 2 16 4

1 1

12 13 46 39

∑x=12 ∑y=13

∑x2 = 46 ∑y2 = 39

X = 12/4 = 3, Y = 13/5 = 2.5

tC = x–y .
∑x + ∑y
2 2
1 + 1
N1 + N 2 - 2 N 1 N2

- 59 -
Where ∑x2 = ∑x2 – (∑x)2
2
= 46 - 12 N1
4
= 46 – 144
4
∑x2 = 46 - 36
= 10

∑y2 = ∑y2 – (∑y)2

N2
= 39 - 132
5
= 39 – 169
5
= 39 – 33. 8
∑y2 = 5.2

tc = 3 - 2.6 .
10 +5.2 1 + 1
4 + 5.2 4 5
= 0.4 .
15.2 9
7 20
= 0.4 .
(2.17)(0.45)
= 0.4
0.9765
= 0.4 .
0.98818
tc = 0.4048
tc = 0.41

- 60 -
When table value is greater than calculated value.
It shows the result is significant.
You accept null hypothesis. If table value is less reject
the null hypothesis.
Summation Notation (∑)
7 = 2+3+4+5+6+7 = 27
∑r
2
7 = 2n+3n + ------+7n
∑rn
r =2

5 = 13+24+35
∑ ∑ K
3 n
= 1 + 16 + 2 +3
n=3 k=1 = 260 ans.

EXERCISE
10 (K2+5)
∑7

6 f1 x f1 t
∑
i=1

6 (x+y)k
∑
K=4

When dealing with difference then you either use t

test, analysis of variance (ANOVA) F- ratio, Analysis of co-
variance (ANCOVA) etc. if the difference is between 2
variables. It must be t-test but when it is between 2 or
more variables then it must be ANOVA is more preferable.

- 61 -
 However, it should be noted that ANOVA can also
handle 2 variables but t-test cannot handle more than 2.
e.g. any Lecturer teaching post-graduate students should
not be lower than Snr-lecturer. But if you have any
Lecturer lower than the prescribed cadre can only teach
Pre-degree, undergraduates. A Professor can teach Pre-
degree, under-graduates, and Post-graduates students.
But, if you are doing experimental work that involves pre-
test and post-test. Then it must be ANCOVA, with the pre-
test acting as co-varies. If you are dealing with prediction
then it must be regression analysis.

Outlines

1. The statistics and its distribution.

2. Establishing confidence interval.

3. Test of difference of mean.

a. Uncorrelated
Simple
b. correlated

4. Test of difference of variance correlated.

Notations and their meanings
M = population mean
σ2 = variance

- 62 -
M = group mean
S.σ = standard deviation
Sm = standard deviation of the simple mean =
standard Error M–Վ is a constant
Σ
t = m-u varies. This was developed by W.S.
Gusset. (Who used the Sm pseudonym “student”) t
is tabulated for various degrees of freedom, usually
from 1 to 30. For degrees of freedom larger than
30.S2 (the sample variance) is a sufficiently reliable
estimate of σ2 (population variance) so that the
distribution of t is almost identical to that of Z
normal distribution.
Sm = σ = standard error
√n
t = M-u x (+√n)
σ
tsm = (M - u) (+√n )

u = M + tsm

Example 1
Suppose M = 32, Sm = 5, n = 7, calculate the
population mean.
Solution
u = m + tsm
t = 2.447 at 6 df with 95% confident interval
= 32 + (2.447) 5 = (32 – 12,235) or (32 +
12.235)
- 63 -
 = 19.77 or 44.24 = (19.765, 44.235) = (19.37,
44.24)

F-ratio

X1 x2 x3 x12 x22 x32

2 3 1 4 9 1
4 1 3 16 1 9
1 4 2 1 16 4
5 2 4 25 4 16
∑x112 ∑x210 ∑x310 ∑x146 ∑x230 ∑x330
4Xt2 = total of x =32
4Xt2 = 46 + 30 + 30 = 106

Step I = sum of squares total.

sst = ∑xt2 - (∑xt)2
N
N = N1 =4
N2 =4
=4+4+4+4=12
N3 =4 N =N1+N2+N3
b = between 106 – 32
t = total 12
106 - 1024
12
106 – 15.33
SSt = 20.67 = 20.7

Step II = Sum of squares between groups

SSb = (∑x1)2 + (∑x2)2 + (∑x3)2 – (∑x)2
n1 n2 n3 N
122 + 102 + 102 – (32)2
4 4 4 12
144 + 100 + 100
4 4 4

- 64 -
 36 + 25 + 25 – 1024
12
86 – 85.33
SSb = 0.67

Step III:- Sum of squares within groups

SSw + SSb = SSt
SSw + 0.67 = 29.67
SSw = 20.67 – 0.67
SSw = 20

SSw = ∑ X12 – (∑X1)2 + ∑X22 – (∑x2)2 + ∑x32 – (∑x3)2

n1 n2 n3
= 46 - 122 + 30 - 102 + 30 – 102
4 4 4
= 46 -144 + 30 -100 + 30 -100
4 4 4
= 10 +5+5
= 20
Step4:- Mean squares of between groups
msb = SSb
dfb degree of freedom = 3 – 1 = 2
msb = 0.67
2
msb=0.335
Step5 = Mean square of within groups
msb= ssw
d.fw = no of case - groups
= 12 – 3 = 9
msw = 20
9
msb = 2.222

Step 6:- Between / within

msb = 0.335
msw 2.222
- 65 -
F - calculated = 0.15

Source of variance
Source of variance ssb df msw fc ft
Between groups ssb 0.67 2 0.335 0.15
4.26
Within groups ssw 20 9 2.222
Total 20.67 11
You accept the null-hypothesis since f- table value is
higher than f –calculated value.
0.05 – mean the degrees at which you say are
not perfect. But you can still make mistake.

Regression Analysis
Y= a + bx
Dependent intersect independent
Y = a x bx -------- (1)
Multiply by x
Xy = ax + bx2 ------- (2)
Sum (1) x (2)
∑y = an + b∑x ------- (3)
∑xy = a∑ + a∑x2 ------- (4)
Re – arrange
an + b∑x = ∑y
a∑x + b∑x2 =∑x
Using crammer’s method
D=n a
∑xb = n∑x2- – (∑x)2
∑x ∑x2

Da = ∑y ∑x = ∑y ∑x2 - ∑x ∑xy
∑xY ∑x2

:- a = Da =∑x2 ∑x ∑xy
D n∑x2 - (∑x)2
D = n ∑y = n∑xy - ∑x∑y
b

∑x ∑xy
- 66 -
b = Db = n∑xy - ∑x ∑y
D n ∑x2 – (∑x)2
Substitutes in equation (1)
Y = a + bx

Correlation

Derived method

x y x=x – x y=y–y xy x 2 - y2

2 3 -1 1 -1 1 1

4 1 -1 -1 -1 1 1

3 2 0 0 0 0 0

5 1 2 -1 -2 4 1
1 3 -2 1 -2 4 1
15 10 -6 10 4

X = 15 10
5 5

X=3 x=2 rxy = ∑xy

√Ex2. Ey2
Maximum = -6
√10 x 4
You can get is 1
= -6
√40
= -6
6.325
= - 0.949
Rxy = -0.95
- 67 -
Correlation spearman Brown’s method
X y Rx Ry
2 3 4 1.5
4 1 2 4.5
3 2 3 3
5 1 1 4.5
1 3 5 1.5
The highest figure takes 1st position and like that
not when you have 2 numbers and you have only 1
chair
1+2 = 3
2 2
= 1.5
And the other forfeits the 2nd position so we
talk about 3 now.
= 1, 2, 3, 4+5
= 4+5 = 9
2 2
= 4.5
You now say d = deviation
1.5 – 4 = 2.52
= 6.25
4.5 – 2 = 2.52
= 6.25
6 = is constant
d2 rxy = 1 - 6∑d2
6.28 N(N2-1)
6.25
0 = 1-6 (37)
12.25 5 (52 - 1)
12.25 = 1 – 222
37 5(24)
- 68 -
 1- 222
120
= 1 – 1.85
Rxy = -0.85 ans
Pearson product Moment correlation method
X y xy x2 y2
2 3 6 4 9
4 1 4 16 1
3 2 6 9 4
5 1 5 25 1
1 3 3 1 9
15 10 24 55 24
rxy = N∑xy-∑x∑y
√N∑N2–(∑x)2–N(∑y2)-(∑y)2
= 5 (24) – (15) (10)
√5(55)-152) (24)- 102)
= 120 – 150
√ (275 – 225) (124 – 100)
= -30
√(50)(20)
= - 0.238
- 30
√1000
= -30
31.62
= -0.948
rxy = -0.95 ans
EXERCISE

1. You have written a computer programme to

generate random number in the range 0 to 9. The
programme was successfully run to produce 500

- 69 -
 digits and the following distribution of the digit
resulted.
Digit 0 1 2 3 4 5 6 7 8 9
Observed 40 36 28 62 58 60 34 70 40 72
freq.

Are you satisfied that your method of generating random

number is satisfactory?
2. Two horse, X and Y were tested according to the
time (in seconds) taken to run a particular track
with the following results.
Horse X 33 29 30 27 28 33 29 and 30
Horse Y 26 29 30 25 26 and 28
Analyses the above data and report whether or not you
can decimate between the running times of the horses.
3. In a math’s test, the average score for 100 boys
taking the test was 72 with a standard deviation of
10, for 120 girls the average score was 60 and the
standard 12. Test the hypothesis that boys are
better in Math’s than girls are.
4. The following are the JAMB scores and POST-JAMB
examination of 10 JAMB candidates in mathematics.
S/N JAMB POST
1 60 58
2 72 69
3 50 56
4 80 78
5 60 59
6 20 23
7 68 65

- 70 -
8 75 72
9 48 50
10 40 43
Test whether there is a significant difference in the
performance of these ten candidates in JAMB and post
JAMB
5. In an aptitude test was conducted for
Administrative and clerical officers, the result is as
follows:
Mean sample standard sample
Deviation size
Administrative Officers 62 3 5
Clerical Officers 56 4 10
Is there any evidence of significant difference in the
means of the two groups.
6. In an aptitude test, was conducted for
administrative and clerical officers, the result is as
follows:
Mean Sample standard sample
Variation size
Administrative Officers 62 3 5
Clerical Officers 56 4 10
Is there any evidence of significant in the means of the
two groups.
7. Solve the following using derived correlation
method.
X 15 10 20 14 11 9 10 7 6 1
y 12 13 10 15 10 15 13 9 7 8

- 71 -
8. Solve the following using Spearman Brown’s
correlation method
X 15 10 20 14 11 9 10 7 6 1
Y 2 4 3 7 14 13 12 10 20 11
9. Solve the following using Pearson product moment
correlation
X 12 13 10 15 10 15 13 9 7 8
Y 2 4 3 7 14 13 12 10 20 11
10. Solve the following using student t-test
X 5 1 2 4
Y 3 4 3 2 1
11. Solve the following using F -ratio
X 2 4 1 5
Y 3 1 4 2
Let say that there is a survey in mathematic
criterion approach non-reference approach is used.
Hypothesis: the approach does not produce any
difference.
Criterion Non- Total
Ref.
01 20 10 30
02 4 9 13
Total 24 19 43

- 72 -
 CHAPTER SEVEN

ANALYSIS OF SCHOOL ENROLMENT.

In diagnosis school environment. There things that

are of vital importance.

- Enrolment trend.

- Enrolment ratio.

- Enrolment rate.

Enrolment trend: It enables us to know absolute

increase or decrease in enrolment as well as growth
rate of enrolment. This can be seen from two angles.
Absolute increase over a given period of time or
through the growth rate of enrolment over a given
period of time. However growth rate of enrolment is
found to be more useful in calculating the enrolment
trend than the other method:

2001 / 2002 session: Table A

Schools Boys Girls Total

A 1820 1600 3420

B 1680 1220 2900

C 875 - 875

D 1320 1480 2800

- 73 -
 Total 5695 4300 9995

2002 / 2003 session Table B

Schools Boys Girls Total

A 2040 1820 3860

B 1850 1325 3175

C 1250 - 1250

D 1680 1840 3480

Total 6780 4988 11,765

2002 / 2003 - 2040

2001 / 2002 - 1820
220
Increase = Enrolment 1 – Enrolment 2
= 2040 - 1820 = 220

(2) Growth rate of enrolment

G = Et + 1 – Et x 100
et 1

GT = 2040 – 1820 x 100

1820 1

= 220 X 100
1820 1

- 74 -
 = 12.087%

= 12.09%

1820 – 1600 X 100

160 1

= 13.75 = 13.8%

Enrolment ratio: Is define as the ratio between the

number of pupil’s enroll of a given age of at a given level
of education and the size of the population in that given
age. The enrolment ratio can be calculated for a given sex
weather boy or girl for private public schools for different
region in the country.
There are 3 ways of measuring enrolment ratio.

1. Overall enrolment ratio : This is otherwise

known as general enrolment ratio or crude
enrolment ratio and it is the ratio between total
school age pupils enroll in the education sector
and total population of age group within the
society

Oer = E (t) x 100

Pt 1

Overall enrolment ratio is adequate for distant study of

enrolment development in a country. But it is weak
because it does not inculcate the no of pupils or students
enroll at each level of educational system. It is
- 75 -
● Et = total school enrolment at all levels

● Oer = overall environment

● Pt = total population of school age in year t

Therefor crude for example if enrolment figure at all levels

of educational system in Nigeria is 212, 168, 240, as total
school age 392, 407, 322 pupils

Oer = 212, 168 240 x 100

392 407 322

= 54.068

= 54.07

Only 54-07% are enrolling in school.

b. level specific enrolment ratio: this is also known as level

enrolment ratio. It is the most commonly used indicator of
development is two types.

1. Gross level enrolment ratio

2. Net level enrolment ratio

1. Gross level: It relates total student enrolment a

specified educational level (regardless of those enroll) to
the students’ population that most enroll for that school
level. It is denoted by gross enrolment.

- 76 -
GLER = Emt x 100
Pgt 1
Emt enrolment at school level weather primary secondary
and tertiary.

Pgt = population of those that supposed to be in

secondary school.

2. Net level enrolment ratio: This includes the

enrolment in school at a particular year and the
number of student who is of the same age group
and the level. Discard under age and ignore over
age 11 – 16 are to be in secondary schools.

c. Enrolment rate: the term rate and ratio are binomials

size of one number to rate is a special ratio often used in
the analysis of floors. It indicates the relative frequency of
the occurency of a population.

Thus we talk of birth, death, promotion/ transition rates

and so no white enrolment rate are vital policy variable
that can affect that determination of future enrolment.
Enrolment ratio is indicator of countries educational
growth enrolment rates are very essential ingredient in the
flow model of enrolment projection.

The non-schooling gap.

It is the difference between the estimated population of

appropriate age group and the number enroll in the

- 77 -
education sector carries ponding to that group. It
represents the number of student who should be
benefiting from education but are not actually there.

PROMOTION RATE.

This is also known as progression rate. It is the

ratio of the total number of students that are promoted to
the next higher level in a given academic year.

To the total number of students that enroll in the

same class in the previous academic session. This is
symbolically describe as thus pg + 1 to power + equal no
of student promoted to next class g + 1 class year +
pupils one.

Grade class I

Year 2000.

Egt = total number of student that are enroll in year t =

2000

Pgt + 1 = total no of student enroll in the former class g

in the previous year t.
Pt+1
g + t = class 2 in years 1

Repetition rate: This is define as the ratio of number of

student that are repeating a specific class in a given
academic year to the total number of students enroll in
- 78 -
the same class levels in the previous academic years. It is
express as:

Rgt = Rgt + 1 x 100

Egt 1

Rgt + 1 = No of student that are repeating the same class

level in year t + 1 = Egt = enrollment in year t class one
in year 2000

4. Dropout rate: It is the ratio of student that are not

promoted or those repeated from a specific class level in
a given year to the total number of student enroll in the
same class level in the same academic session.

A child that is first in a class still dropout in a school what

could be the factor

 Religious crisis

 Death.

 Physical disabilities

 Financial problem

 Parent separation (devours)

 Family problem

 Mobility of labour

 Sickness.
- 79 -
Egt - Pgt = (P t+1g+1 x Rgt = Rgt+1 x 1) x 100
Et

Egt = dgt x 100 (2)

Egt 1

6. Wastage rate: this is the proportion of total

enrollment accounted for by stagnation (repetition and
dropout) express as a percentage.

Add dropout + repetition

= Wastage rate:

Remove promotion from 1

Wt = 1-pg+1 x 100
Egt 1
Promotion = 78% = 0.78

Repetition = 12% = 0.12

Dropout = 10% = 0.10

Assuming that in Ekiti state in 2000/11 session

The number that got promoted to J.S.S. (2) 216/ 200

The total student enrollment was 282 400 student. The

number that repeated J.S.S. 1 in 2011/12 56,120 and the
number that dropout (withdraw) from the system 10,080.
Calculate the Promotion, Repetition, Dropout and wastage
rate.
- 80 -
(7) Transition rate: it is the rate at which student
move from one level of education in a new academic
session: e.g. primary - secondary - tertiary. It should be
noted that while students are transiting from one level of
education to another. The following might happen

1. Some may be repeating the final class

2. Some might have withdrawn from the level of education

without completing it.

3. Some might have completed the circle and move to the

next higher level

4. Some might even complete the system and join the

labour force.

Mathematically transition rate is given as:

Tti = E i+1t+1 x 100

Eti 1

i = represent final class in the lower level of education

t+1 = represent first class in the next higher level of

education.

Eti = represent enrollment in the first class in the lower

level of education

Ei+1t+1 = represent no of new entrants in the first class of

the next higher level of education in year t + 1.
- 81 -
How many year in 2000 and how many transited to J.S.S.
1 in 2001.

Repeater 56.120 x 100

282,400 1
=19,8725
Dropout 10,080 x 100
282,400 1
=3.5694
Wastage = 19.8 + 3.5
= 23.5
Promotion 226,280 x 100
282,400 1
= 80.127
Promotion Rate.
Pgt = Pt+1g+1 x 100
Egt 1
216,200 x 100
282,400 1
= 76.58%
Repetition rate.
Rgt+1. 56.120 x 100
Egt 282,400 1
= 19.87%
Dropout rate.
Egt – (pgt = pgt+1 + Rgt = Rgt+1) x 100
1
100 – (216,200+ 56,120) x 100
282,400 1
272,320 x 100
282,400 1
=96.43
100 – 96.43 =3.57%

- 82 -
Edgt = dgt X 100
Egt 1
10,080 x 100
282,400 1
=19,8
Wastage rate.
Wt (1 – pg+1t + 1) x 100
Egt 1
Wastage = repetition + dropout
(19.87 + 3.57)%
23.44%
100 – Promotion
100 – 76.56%
= 23.44%
Cohort Analysis
It shows the extent to which educational system is able to
use his import in the production of set of students in a
particular level to the time they left that level of education.
Cohort analysis is used to determine the internal efficiency
of an educational system. When the educational system is
able to see the pupils (student) through the system in the
shortest possible period then the system is efficient. Given
the information below prepare a cohort analysis showing
the movement of pupils from J.S.S.1 to J.S.S 3 and
calculate the wastage ratio of the school system.
Entrant =1000 student
Promotion rate = 80%
Repetition rate = 15%
Dropout rate = 5%
- You don’t admit student again.
- Students are to repeat twice before total
withdrawer from the system.

- 83 -
 Promotion.

Repetition.

Dropout.
J.S.S.1 J.S.S.2 J. S.S.3 Output
2001 1000
800
640 512
2002 150
240
288 231
2003 23
54
86 69
2004

2005
Inputs 1,173 1094 1014

86 x 80 = 68.8
100
1000 x 80 = 800 promotion = 69
100 812
1000 x15 = 150
100
800 x 80 = 640
100

- 84 -
15 x 100 =23
100
Student input – years (input)
1173 + 1094 + 1014 = 3281
Input = 3281
Wastage ratio = Actual
Input – output ratio
Idea input – output ratio
Input – output = input
Output
Actual input – output ratio =
Actual input
Actual output
= 3281
812
Actual input = 4.40 – output ratio
Ideal ratio.
J.S.S.1 J.S.S.2 J.S.S.3
2001 1000
1000
1000

1000 + 1000 + 1000

Input = 3000
Output = 1000 = 3
Wastage ratio = 4.04
3.00
Wastage = 1.35
1 < Wr < 2
Wastage ratio is greater than or equal to 1 or less than or
equal to 2.
The closer the wastage ratio to 1 the move efficient
is it. The fat away from 1 the less efficient the system is
- 85 -
e.g. 1.77 is fending toward to 2 which mean they would
use times two of the resources to train the children it is
less efficient. While 1.35 is fairly efficient.
If the wastage ratio is equal to 1 that means the
education system is perfectly efficient which is not
attainable anywhere in the world.
Problem of wastage
1. Poor attitude of leaners toward learning e.g
technological problem.
2. Death of personnel core subject teachers e.g
physics, English, mathematics, or lack of teachers.
3. Library problem : lack of adequate text – books
4. Inadequate supervision both internal and external
5. Desire to make quick money
6. Parental influence- parents encouraging their
children in examination mal-practice by buying
questions for them.
7. Inadequate funding
8. Poor instructional materials
9. Incessant strike action e.g. Waec is a yearly exam if
there is a strike that lasted for 6 months. Obviously,
it would have the performance of the students.
Co-efficiency of efficiency equal to wastage ratio
multiply by 100
1
= 1 x 100
1.35 1
= 100
1.35
Co-efficient of = 74.07%
Efficient is internally efficient.

- 86 -
 CHAPTER EIGHT
ENROLMENT PROJECTION
A projection is a conditionally statement about the
future. It is the elaboration of the effect in the future of
making a set of assumption about trend in the parameters
characterizing the educational system.
A projection does not necessary offers the most
probable (in some person’s judgment) outcome rather its
main function is to demonstrate to the decision maker the
result which follows from carrying some of the parameters
(or from leaning them unchanged). Depending on the
desirability of the decision maker of me projected outcome
he may intervene with policy changes to affect the
underline trends.
One major way of looking at enrolment projection
in educational management is compounding and
discounting.
Compounding and discounting are technics used in
comparing the size of variable at different pout in times.
Compounding deals with finding the future worth of
present resources and other variable such as population,
enrolment e.t.c. growing by geometrical progression.
Discounting is concern with the calculation of present
worth of a future amount.
It is the opposite of compounding as it looks from
the future to the present. Compounding is very useful in
education planning as it helps to solve problems relating
- 87 -
targets and projection of enrolment, teacher’s demand and
supply, cost, finance e.t.c.
For this purpose educational planners must have
the mastery of this technique.
Assuming a school has an enrolment of 1,000 in
year 2000. And the school is expected to grow at the rate
of 6% for the next 10 years.
Years Increase Enrolment in Enrolment in
during year t yrs t – 1 year t
2000 - - 1000
2001 1000x0.06 1000 1060
2002 1060x0.06 64 1060 1124
2003 1124x0.06 67 1124 1191
2004 1191x0.06 71.46 1121 1262
2005 1262x0.06 76 1262 1338
2006 1517x0.06 80 1517 1418
2007 1001x0.06 85 1001 1503
2008 1001x0.06 96 1925 1593
2009 100x0.06 101 1936 1689
2010 100x0.06 205 2052 1790
1791
To 2 s.f. If the last is up to five then approximate.
The computation of the figures as shown in the
table above is time watching, time consuming and as well
as it is difficult to calculate. Since we know that the annual
growth rate is 6% and the time interval is 10 years, the
enrolment for each year can be obtained directly by
multiplying the enrolment of the preceding year by a
factor which is always the same, 1 + rate of increase = 1
+ 0.06

- 88 -
= 6% 6 . = 0.06
1000
This is the characteristics of a geometric progression to
this end. The information presented in table 1 can be
calculated more directly as presented in table 2.
Year Enrolment (t-1) Calculate Enrolment
2000 - - 1000
2001 1000 1000x1.06 1060
2002 1060 1060x0.06 1124
2003 1124 1124x0.16 119
2004 1191 1191x0.06 1262
2005 1262 1262x0.06 1338
2006 1338 1338x0.06 1419
2007 1419 1419x0.06 1504
2008 1504 1504x0.06 1594
2009 1594 1594x0.06 1690
2010 1690 1690x0.06 1791
In the example given above the experiment of 1.06 between
year 2000 and 2010 is 10. Therefore in mathematical terms,
this can be stated as 1 + r 10 rate of increase (1 + 0.06)10 = 6%
= 0.06
It thus, implies that to obtained enrolment in 2010. The
enrolment of 2000 multiply by (1.06)10 to this end the for
mincers emerge
En = Eo x (1 x r)n
Where En = enrolment in the final year 2010.
Eo = enrolment in the initial year = 2000
r = rate or increase 6%
n = no of years 10
The interval section 2000 and 2010 = 10 years

- 89 -
Computer method
Inserting the figure from our example
En = 1000 x (1.06)10
1000 x 1.790847
En = 1791 (1.06)10
out of the four variables in the above equation.
En, Eo, r, n. we only know 3. The fourth can be worked out
using the logarithms table.
En = Eo x (1 + r)n ------- (1)
Log En = n log (1 + r) ------- (2)
Log Eo
Log En = log Eo + n log (1 + r) ------- (3)
Using equation 3. We have log En = log 1000+ 10 log (1 +
0.06) log 100 (3)
Log En = log 1000 + 3 + 10 x 0.0252
Antilog = 3.253
=1,7910
=1791
Supposing we know that at the initial stage that enrolment in
year 2000 was 1,000 and in 2010, it was 1,791 and ask 1,791
and we are ask to find rate of growth.
Log En = n log (1+r)
Log Eo
n log (1+r) = log En
log En
10 log (1+r) = log 1791
Log 1000
Divide both side by 10
3 log 1006 = log 1,791
Log 1000
Log (1+r) = 1.791
10
Log (1+r) = 0.2531
10
Anti–log 0.2531

- 90 -
 Log (1+r) = 1.060
1+r = 1.060
r = 1.060 – 1
r = 0.060 x 100
r = 6%
Supposing we want to know how long it will take the enrolment
of 1,000 to reach 1,791 if it grows at the rate of 6% per
annum.
Log En = n log (1+r) ----- (2)
Log Eo
n log (1+r) log En
Log En
10 log (1+r) 1791
1000
10 log (1+0.06) 1.791
n log (1+0.06) log 1.791
n log 1.06 = log 1.791
n = (0.02531) = 0.2531
n = 0.2531
0.02531
n = 10
Discounting factor
It is the reciprocal of compounding factor
1
(1+r)n
Eo = En x 1 .
(1+r)n
It wants to look at project want of future amount.
Eo = 1791 x 1 .
(1+0.06)10
1791 x 1 .
(1.06)10
1791 x 1 .
1.791
1791 x 0.558

- 91 -
 = 1000
Q4 = 2002 / 2003
902, 200
2,200 – over age.
4800 – under age.
1,322,000
(a) 902, 200 – (2200 + 4800)
902,200 – 7000
=895,200
895,200 x 100
1322000 1
= 67.72%
(b) 902,200 x 100
1322000 1
=67.72%
(16B) 2005 / 2006
J.S.S. 2 = 326,463
Promoted to J.S.S 3 247, 403
Repeated of J.S.S 2
2006 / 2007 = 69.060
(1) Repetition = 69060 x 100
326463 1
= 12.15%
Dropout = promotion + repetition
= total enrolment
Compounding
En = Eo x (1+r)n
En = 1000 x (1+0.05)10
En = 10,000 x (1.05)10
10,000 x 1.63889
=16,300

- 92 -
 CHAPTER NINE
MODEL
A model can come in many shapes, sizes, and
styles. It is important to emphasize that a model is not the
real world but merely a human construct to help us better
understand real world systems. In general all models have
an information input, an information processor, and an
output of expected results. Simplifying assumptions must
be made;

 boundary conditions or initial conditions must be

identified;
 the range of applicability of the model should be
understood.

Descriptive Models

A descriptive model describes logical relationships, such as

the system's whole-part relationship that defines its parts
tree, the interconnection between its parts, the functions
that its components perform, or the test cases that are
used to verify the system requirements. Typical descriptive
models may include those that describe the functional or
physical architecture of a system, or the three dimensional
geometric representation of a system.

Analytical Models

An analytical model describes mathematical relationships,

such as differential equations that support quantifiable
analysis about the system parameters. Analytical models
can be further classified into dynamic and static models.
Dynamic models describe the time-varying state of a

- 93 -
system, whereas static models perform computations that
do not represent the time-varying state of a system. A
dynamic model may represent the performance of a
system, such as the aircraft position, velocity, acceleration,
and fuel consumption over time. A static model may
represent the mass properties estimate or reliability
prediction of a system or component.

Hybrid Descriptive and Analytical Models

A particular model may include descriptive and analytical

aspects as described above, but models may favor one
aspect or the other. The logical relationships of a
descriptive model can also be analyzed, and inferences
can be made to reason about the system. Nevertheless,
logical analysis provides different insights than a
quantitative analysis of system parameters.

Domain-specific Models

Both descriptive and analytical models can be further

classified according to the domain that they represent.
The following classifications are partially derived from the
presentation on OWL, Ontologies and SysML Profiles:
Knowledge Representation and Modeling (Web Ontology
Language (OWL) & Systems Modeling Language (SysML))
(Jenkins 2010):

 properties of the system, such as performance,

reliability, mass properties, power, structural, or
thermal models;
 design and technology implementations, such as
electrical, mechanical, and software design models;

- 94 -
  subsystems and products, such as communications,
fault management, or power distribution models;
and
 system applications, such as information systems,
automotive systems, aerospace systems, or medical
device models.

The model classification, terminology and approach is

often adapted to a particular application domain. For
example, when modeling organization or business, the
behavioral model may be referred to as workflow or
process model, and the performance modeling may refer
to the cost and schedule performance associated with the
organization or business process.

A single model may include multiple domain categories

from the above list. For example, a reliability, thermal,
and/or power model may be defined for an electrical
design of a communications subsystem for an aerospace
system, such as an aircraft or satellite.

What is an Interactive Lecture Demonstration?

Interactive Lecture Demonstrations introduce a carefully

scripted activity, creating a "time for telling" in a
traditional lecture format. Because the activity causes
students to confront their prior understanding of a core
concept, students are ready to learn in a follow-up lecture.
Interactive Lecture Demonstrations use three steps in
which students:

1. Predict the outcome of the demonstration.

Individually, and then with a partner, students

- 95 -
 explain to each other which of a set of possible
outcomes is most likely to occur.
2. Experience the demonstration. Working in small
groups, students conduct an experiment, take a
survey, or work with data to determine whether
their initial beliefs were confirmed (or not).
3. Reflect on the outcome. Students think about why
they held their initial belief and in what ways the
demonstration confirmed or contradicted this belief.
After comparing these thoughts with other
students, students individually prepare a written
product on what was learned.

Why Use Interactive Lecture Demonstrations

Research shows that students acquire significantly greater

understanding of course material when traditional lectures
are combined with interactive demonstrations. Each step
in Interactive Demonstrations--Predict, Experience,
Reflect--contributes to student learning. Prediction links
new learning to prior understanding. The experience
engages the student with compelling evidence. Reflection
helps students identify and consolidate that they have
learned.

How to Use Interactive Lecture Demonstrations in

Class

Effective interactive lecture demonstrations require that

instructors:

 Identify a core concept that students will learn.

- 96 -
  Chose a demonstration that will illustrate the core
concept, ideally with an outcome different from
student expectations.
 Prepare written materials so that students can
easily follow the prediction, experience and
reflection steps.

In simplest terms, a mathematical model is an abstraction

or simplification that allows us to summarize (describe) a
system. Once you have a mathematical model you have a
list of inputs and a list of outputs and some sort of definite
algorithm that tells you what the outputs will be given the
inputs. Once we agree on that definition of what a
mathematical model IS then we can talk about your
question and its answer or answers.

Some of the benefits of building and using mathematical

models:

 Ability to predict system behavior

 A clear idea of the important inputs and outputs
 Ability to analyze anomalous behavior by comparing
it to the model-predicted behavior

Some of the disadvantages of using mathematical models:

 The model may eliminate important predictive

power by being too simple
 The model may be capable in certain
circumstances, but not in others and the assumed
conditions may not be obvious or understood by
later users

- 97 -
 CHAPTER TEN

EFFICIENCY IN EDUCATION

Educators often feel ambivalent about the pursuit of

efficiency in education. On the one hand, there is a basic
belief that efficiency is good and worthy goal; on the other
hand, there is sense of worry that efforts to improve
efficiency will ultimately undermine what lies at the heart
of high-quality education. Part of the difficulty stems from
a misunderstanding about the meaning of efficiency as
well as from the legacy of past, sometimes misguided,
efforts to improve the efficiency of educational systems. It
is therefore useful to begin with a basic discussion of the
efficiency concept.

The notion of efficiency applies to a remarkably

large number of fields, including education. It is a
disarmingly simple idea that presupposes a transformation
of some kind. One can think in terms of what was in hand
before the transformation, what was in hand after the
transformation, and one can also think about the
transformation process itself. The before elements are
commonly referred to as ingredients, inputs, or resources
while the after elements are called results, outputs, or
outcomes. The transformation process is sometimes less
obvious and can become confused with ingredients. For
- 98 -
example, in an educational setting, a teacher can be
thought of as an ingredient while teaching is an important
part of the actual transformation process.

The concept of efficiency is often connected to a

moral imperative to obtain more desired results from
fewer resources. Efficiency needs to be thought of as a
matter of degree. Efficiency is not a "yes/no" kind of
phenomenon. It is instead better thought of in relative or
comparative terms. One operation may be more efficient
than another. This said, the more efficient of the two
operations could become even more efficient. The quest
for greater efficiency is never over, and this sense of a
perennially unfinished agenda is one source of the
generalized sense of anxiety that tends to surround the
efficiency concept.

The Choice of Outcomes

If the goal is to obtain more desired results from fewer

resources, then it is important to be clear about what is
being sought. Society might have a very efficient system
because a large amount of outcome is being obtained
relative to the resources being spent or invested, but if the
outcomes are out of sync with what is truly desired, there
is a real sense in which the system is not very efficient. Of

- 99 -
course, this invites important questions about who gets to
decide what counts as a desirable outcome, and in
education there are longstanding and ongoing debates
over what the educational system ought to be
accomplishing.

In addition to reaching agreement about the mix of

outcomes to pursue, there are important measurement
issues to consider. An interest in efficiency is frequently
accompanied by an interest in measuring magnitudes. If
one is seeking more out of less, one frequently wants to
know "how much more," and the result has been a boom
in the efforts by educational psychologists and others to
develop valid and reliable measures of the learning gains
of students. Critics of efficiency analysis in education
worry that ease of measurement can unduly influence the
selection of the outcomes that the system will be
structured to achieve. In other words, the worry is that
the drive for efficiency will lead, perhaps inadvertently,
toward the use of educational outcomes that are chosen
more because they are easy to measure than because of
their intrinsic long-term value for either individual students
or the larger society. Standardized tests of various kinds
have been relied upon as measures of the outcomes of
schooling and have been criticized on these grounds.

- 100 -
Sometimes there is interest in the economic consequences
of schooling, and this interest has prompted analysts to
use earnings as a measure of schooling outcomes. A rich
literature has developed in the economics of education
where efforts have been made to estimate the economic
rate of return to different levels and types of schooling.
This is a challenging area of research because earnings
are influenced by many factors and it is difficult to isolate
the effects of schooling. The goal of this research is to
capture the value added by schooling activities.

The relevance of the value-added concept is not limited to

economists' studies of rates of return. Even in cases where
the focus is on learning outcomes as measured by tests or
other psychometric instruments, there are questions to
answer about the effects of schooling activities relative to
the effects of other potentially quite significant influences
on gains in students' capabilities. Serious studies of the
efficiency of educational systems measure educational
outcomes in value-added terms.

Measurement issues also arise from the collective nature

of schooling. The results gained from schooling
experiences are likely to vary among individual students
and this prompts questions about how best to examine the
result for the group in contrast to an individual student. Is

- 101 -
one primarily interested in, say, the average performance
level, or is there a parallel and perhaps even more
important concern with what is happening to the level of
variation that exists across all of the students within the
unit, be it a classroom, grade level within a school, a
school, a district, a state, or a nation? The early research
on educational efficiency in the 1960s placed a heavy
emphasis on average test score results for relatively large
units like school districts. More recent work demonstrates
greater interest in measures of inequality among students.
The standards-driven reform movement includes a
considerable amount of rhetoric about all students
reaching high standards; the analysis of efficiency
presupposes an ability to move beyond the easy rhetoric
to make clear decisions about how uniform performance
expectations are for students.

In addition, there is an important distinction to maintain

between the levels at which a system operates and the
rate at which inputs are being transformed into outcomes.
One can "get the outputs right" so that the desired items
are being taught/learned in the correct proportion to one
another. In such a case, gains in the understanding of
mathematics are occurring in the correct proportion to,
say, gains in language capabilities. But this says nothing
about the absolute level at which the system is operating.
- 102 -
The naive view might be that the system should operate
at 100 percent of its capacity, but this overlooks the fact
that scarce resources are needed to operate at this level
and that education is not the only worthy use of these
precious resources. Policy-makers must make often
difficult trade-off decisions about the level at which the
educational system will operate relative to the level of
other competing social services. The early twenty-first
century is witnessing a considerable amount of debate
over the proper level at which to set the educational
system, often as part of an effort to define what counts as
an "adequate" education.

With respect to outcomes, the goal is to reach agreement

about (1) the relative mix of performance outcomes to
realize; (2) the degree of uniformity of performance across
students; and (3) the level of capacity at which the system
should operate. In addition, there needs to be an ability to
measure what is being accomplished.

- 103 -
 INDEX
A L

advantages of array 9 lower quartile 50

alter 38 M

addition 29 meaning of statistics 6

C model 97

central tendency 29 mean 65

cohort analysis 82 median 37

D mode 52

data classification 8 N

discounting factor 92 numeric facts 6

E O

enrolment projection 87 overall enrolment ratio 73

efficiency in education 103 outline 64

F P

frequency count 101 primary vs secondary data 10

G parents separation 79

group of data 32 promotion rate 80

gross level 76 Q

H qualitative data 11

Hybrid description 98 quintile 48

I R

Importance of statistics 20 repetition rate 81

Q S

Quantiles 48 Sampling techniques 24

Skewness 40

W T

what is interactive 100 types of statistics 6

- 104 -
 APPENDIX TABLE 2A

Areas in table combined for student’s t distribution.*

EXAMPLE: To find the value of t which corresponds to and area of.

10 in both tails of the distribution combined, when there are 19
degree of freedom, look under the 10 column and proceed down to
the 19 degree of freedom row, the appropriate f value is 1.729.
Area in both tails combined

Degrees of freedom .10 .05 .025 .01

1. 6.314 12.706 31.821 63.657
2. 2.920 4.303 6.965 9.925
3. 2.353 3.183 4.541 5.841
4. 2.132 2.776 3.747 4.604
5. 2.015 2.571 3.365 4.032
6. 1.943 2.447 3.143 3.707
7. 1.895 2.365 2.998 3.449
8. 1.860 2.306 2.896 3.355
9. 1.833 2.262 2.821 3.250
10. 1.712 2.228 2.764 3.169
11. 1.796 2.201 2.718 3.106
12. 1.782 2.179 2.681 3.055
13. 1.771 2.160 2.650 3.012
14. 1.761 2.145 2.624 2.977
15. 1.753 2.131 2.602 2.947
16. 1.746 2.120 2.583 2.921
17.. 1.740 2.110 2.567 2.898
18. 1.734 2.101 2.552 2.878
19. 1.729 2.093 2.539 2.861
20. 1.725 2.086 2.528 2.845
21. 1.721 2.080 2.518 2.831
22. 1.717 2.074 2.508 2.819
23. 1.714 2.069 2.500 2.807
24. 1.711 2.064 2.492 2.797
25. 1.708 2.060 2.485 2.787
26. 1.706 2.056 2.479 2.779
27. 1.703 2.052 2.473 2.771
28. 1.701 2.048 2.467 2.763
29. 1.699 2.045 2.462 2.756
30. 1.697 2.042 2.457 2.750
40. 1.684 2.021 2.423 2.604
60. 1.671 2.000 2.390 2.660
120. 1.658 1.980 2.358 2.617
Normal distribution 1.645 1.960 2.326 2.576
*Taken from Table III of Fisher and Yates, Statistical Table for Biological, Agricultural
and Medical Research, published by Longman Group Ltd, London (previous published
Oliver & Boyd, Edinburgh) and by permission of the authors and publishers.

- 105 -
 APPENDIX 2B
TABLE 2 Critical values of t*

Level of significance for one tailed test

10. .05 .25 .01 .005 .0005

Level of significant for two-tailed test

Degrees of .20 .10 .5 .02 .01 .011

freedom
1. 3.078 6.314 12.706 31.821 63.657 636.619
2. 1.886 2.920 4.303 6.965 9.925 31.598
3. 1.638 2.353 3.183 4.541 5.841 12.941
4. 1.533 2.132 2.776 3.747 4.604 8.610
5. 1.476 2.015 2.571 3.365 4.032 6.859
6. 1.440 1.943 2.447 3.143 3.707 5.959
7. 1.415 1.865 2.365 2.998 3.449 5.405
8. 1.397 1.860 2.306 2.896 3.355 5.041
9. 1.383 1.833 2.262 2.821 3.250 4.781
10. 1.372 1.812 2.228 2.764 3.169 4.587
11. 1.363 1.796 2.201 2.718 3.106 4.437
12. 1.356 1.782 2.179 2.681 3.055 4.318
13. 1.350 1.771 2.160 2.650 3.012 4.221
14. 1.345 1.761 2.145 2.624 2.977 4.140
15. 1.341 1.753 2.131 2.602 2.947 4.073
16. 1.337 1.746 2.120 2.583 2.921 4.015
17.. 1.333 1.740 2.110 2.567 2.898 3.965
18. 1.330 1.734 2.101 2.552 2.878 3.922
19. 1.328 1.729 2.093 2.539 2.861 3.883
20. 1.325 1.725 2.086 2.528 2.845 3.850
21. 1.323 1.721 2.080 2.518 2.831 3.819
22. 1.321 1.717 2.074 2.508 2.819 3.792
23. 1.319 1.714 2.069 2.500 2.807 3.767
24. 1.318 1.711 2.064 2.492 2.797 3.745
25. 1.316 1.708 2.060 2.485 2.787 3.725
26. 1.315 1.706 2.056 2.479 2.779 3.707
27. 1.314 1.703 2.052 2.473 2.771 3.690
28. 1.313 1.701 2.048 2.467 2.763 3.674
29. 1.311 1.699 2.045 2.462 2.756 3.659
30. 1.310 1.697 2.042 2.457 2.750 3.646
40. 1.303 1.684 2.021 2.423 2.704 3.551
60. 1.296 1.671 2.000 2.390 2.660 3.460
120. 1.289 1.658 1.980 2.358 2.617 3.373
A 1.282 1.645 1.960 2.326 2.576 3.291

- 106 -
 APPENDIX 3
TABLE 3 Critical values of person r*
Level of significance for one-tailed test
df .05 .025 .01 .005
Level of significance for two-tailed test
(=N-2; N= number of pairs ) .10 .05 .02 .01

1. .988 .997 .9995 .9995

2. .900 .950 .980 .990
3. .805 .878 .934 .959
4. .729 .811 .882 .917
5. .669 .754 .833 .974
6. .522 .707 .789 .834
7. .532 .666 .750 .898
8. .549 .632 .716 .765
9. .521 .602 .685 .735
10. .497 .576 .658 .708
11. .476 .553 .634 .784
12. .458 .532 .612 .661
13. .441 .514 .592 .641
14. .426 .497 .574 .623
15. .412 .482 .558 .606
16. .400 .468 .542 .690
17.. .389 .456 .528 .575
18. .378 .444 .516 .516
19. .369 .433 .503 .549
20. .360 .423 .492 .537
21. .352 .413 .482 .526
22. .344 .404 .472 .515
23. .337 .396 .462 .505
24. .330 .388 .453 .496
25. .323 .381 .445 .487
26. .317 .374 .437 .479
27. .311 .367 .430 .471
28. .306 .361 .423 .463
29. .301 .355 .416 .456
30. .296 .349 .409 .449
35. .275 .325 .381 .418
40. .257 .304 .358 .393
45. .243 .288 .338 .372
50. .231 .273 .322 .354
60 .211 .250 .265 .325
70 .165 .232 .274 .302
80 .183 .217 .256 .283
90 .173 .205 .242 .267
100 .164 .195 .230 .254
From R.A. Fisher and F. Yates, “Statistical Table for biological, agricultural and Medical
Research,” 6th ed. Oliver and Boyd, Edinburg, 1963. Reproduced by permission of
authors and publisher.

- 107 -
 APPENDIX 4

TABLE 4 Critical values of chi-square*

Level of significance

df .20 .10 .05 .02 .01 .001

1. 1.64 2.71 3.84 5.41 6.63 10.83

2. 3.22 4.61 5.99 7.82 9.21 13.82
3. 4.64 6.25 7.82 9.84 11.34 16.27
4. 5.99 7.78 9.49 11.67 13.28 18.46
5. 7.29 9.24 11.07 13.39 15.09 20.52
6. 8.56 10.64 12.59 15.03 16.81 22.46
7. 9.80 12.02 14.07 16.62 18.48 24.32
8. 14.03 13.36 15.51 18.17 20.09 26.12
9. 12.24 14.68 16.92 19.68 21.67 27.88
10. 13.44 15.99 18.31 21.16 23.21 29.59
11. 14.63 17.28 19.68 22.62 24.72 31.26
12. 15.81 18.55 21.03 24.05 26.22 32.96
13. 16.98 19.81 22.36 25.47 27.69 34.53
14. 18.15 21.06 23.68 26.87 29.14 36.12
15. 19.31 22.31 25.00 28.26 30.58 37.70
16. 20.46 23.54 26.30 29.63 32.00 39.25
17.. 21.62 24.77 27.59 31.00 33.41 40.79
18. 22.76 25.99 28.87 32.35 34.81 42.31
19. 23.90 27.20 30.14 33.69 36.19 43.82
20. 25.04 28.41 31.41 35.02 37.57 45.32
21. 26.17 29.62 32.67 36.34 38.93 46.80
22. 27.30 30.81 33.92 37.66 40.29 48.27
23. 28.43 32.01 35.17 38.97 41.64 49.78
24. 39.55 33.20 36.42 40.27 42.98 51.18
25. 30.98 34.38 37.65 41.57 44.31 52.62
26. 31.80 35.56 38.89 42.68 45.64 54.05
27. 32.91 36.74 40.11 44.14 46.96 55.48
28. 34.03 37.92 41.34 45.42 48.28 56.89
29. 35.14 39.09 42.56 46.69 49.59 58.30
30. 36.25 40.26 43.77 47.96 50.89 59.70
*From R.A, Fisher and F Yates, “Statistical Table for Biological, Agricultural and medical
Research,” 6th ed. Oliver and Boyd, Edinburg, 1963. Reproduced by permission of
authors and publisher.

*For df greater than 30, the value obtained from the expression may be used as t ratio
2X - 2df -1

- 108 -
 APPENDIX 5

TABLE 5 Distribution of F

P = 0.5
n1
n2 1 2 3 4 5 6 8 12 14 a
1 161. 199. 215. 224. 230. 234.0 238. 243. 249. 254.
4 5 7 6 2 9 9 0 3
2 18.5 19.0 19.1 19.2 19.3 119.3 19.3 19.4 19.4 19.5
1 0 6 5 0 3 7 1 5 0
3 10.1 9.55 9.28 9.12 9.01 8.94 8.84 8.74 8.64 8.53
3
4 7.71 6.94 6.59 6.39 6.26 6.16 6.04 6.91 6.77 5.63
5 6.61 5.79 5.41 4.53 5.05 4.95 4.82 4.68 4.53 4.36
6 5.99 5.14 4.76 4.12 4.39 4.28 4.15 4.00 4.84 3.67
7 5.59 4.74 4.35 3.84 3.67 3.87 3.73 3.57 3.41 3.23
8 5.32 4.46 4.07 4.35 3.69 3.58 3.44 3.28 3.12 3.93
9 5.12 4.26 3.86 3.63 3.48 3.37 3.23 3.07 3.90 2.71
10 4.96 4.10 3.71 3.48 3.33 3.22 3.07 3.91 2.74 2.54
11 4.84 3.98 3.59 3.36 3.20 3.09 3.95 2.79 2.61 2.40
12 4.75 3.88 3.49 3.26 3.11 2.00 2.85 2.69 2.50 2.30
13 4.67 3.80 3.41 3.18 3.02 2.92 2.77 2.60 2.42 2.31
14 4.60 3.74 3.34 3.11 2.96 2.85 2.70 2.83 2.35 2.13
15 4.54 3.68 3.29 3.06 2.90 2.79 2.94 2.48 2.29 2.07
16 4.49 3.63 3.24 3.01 2.85 2.74 2.59 2.42 2.24 2.01
17 4.45 3.59 3.20 2.96 2.81 2.70 2.55 2.38 2.19 1.96
18 4.41 3.55 3.16 2.93 2.77 2.66 2.51 2.34 2.15 1.92
19 4.38 3.52 3.13 2.90 2.74 2.63 2.48 2.31 2.11 1.88
20 4.35 3.49 3.10 2.87 2.71 2.60 2.45 2.28 2.06 1.84
21 4.32 3.47 3.07 2.84 2.66 2.57 2.42 2.25 2.05 1.81
22 4.30 3.44 3.05 2.82 2.66 2.55 2.40 2.23 2.03 1.78
23 4.28 3.42 3.03 2.80 2.64 2.53 2.38 2.20 2.00 1.76
24 4.26 3.40 3.01 2.78 2.62 2.51 2.39 2.18 1.98 1.73
25 4.24 3.38 2.99 2.76 2.60 2.49 2.34 2.16 1.96 1.71
26 4.22 3.37 2.98 2.74 2.59 2.47 2.32 2.15 1.95 1.69
27 4.21 3.35 2.95 2.73 2.57 2.46 2.30 2.13 1.93 1.67
28 4.20 3.34 2.95 2.71 2.56 2.44 2.29 2.12 1.91 1.65
29 4.18 3.33 2.93 2.70 2.54 2.43 2.28 2.10 1.90 1.64
30 4.17 3.32 2.92 2.69 2.53 2.42 2.27 2.09 1.89 1.62
40 4.08 3.23 2.84 2.61 2.45 2.34 2.18 2.00 1.79 1.51
60 4.00 3.15 2.76 2.52 2.37 2.25 2.10 1.92 1.70 1.39
12 3.92 3.07 2.68 2.45 2.29 2.17 2.02 1.63 1.61 1.25
0
A 3.84 3.99 2.60 2.73 2.21 2.09 2.94 1.75 1.52 1.00

Value of n1 and n2 represent the degrees of freedom

associated with the larger and smaller estimates of
variance respectively.

- 109 -