COLLEGE OF COMPUTING AND INFORMATICS

DEPARTEMENT OF: - STATISTICS

RESEARCH ON: - SCHOOL RELATED FACTORS AND SCHOOL VARIATION ON
ACADEMIC PERFORMANCE OF STUDENTS: IN CASE OF BATE, GADA AND
MODEL PRIMARY SCHOOL ON GRADE EIGHT STUDENTS.

PREPARED BY: - BEZABEH(MILKESSA) MEKONNEN

ID. NO 2462/0 6

ADVISOR: DANEIL BIFTU

MAY, 2016

HARAMAYA, ETHIOPIA
i
ABSTRACT

This research contains five chapters and each chapter has their own subtitle. This
research was conducted on the school related factors and school variation on
academic performance of students in case of Model primary school, Bate and Gada
primary school on grade eight students. Education is a base for the development of
a country. It promotes social, economic and cultural development. To conduct this
research, we were used primary data that was collected by distributing to the
respondent and secondary data from school. From different method of statistical
analysis, this research is done by using descriptive statistics and inferential
statistics (one-way variance (ANOVA) and multiple regression models). Based on
the results and discussion, library service, parent’s involvement on economy,
availability of textbook, distance of home from school and performance and
advantage of more time spent on study is the main school related factors. This
means the academic performance of students affected by those explanatory
variables.

1
 CHAPTER ONE
1. INTRODUCTION
1.1 Back Ground of the Study
Education is a base for the development of a country. It promotes social, economic and cultural
development. To this regar says that without education it is hardly possible using or utilizing
environmental resource for further development. Thus, it is a purposeful and deliberative human
action towards the requisition at knowledge, skill and attitude through settings goals or
objectives; it was to be evaluated periodically. Education was seen as the necessary instrument
immediately and essentially for the multi-ethnic society (Olaniyi B.1993). Throughout the world,
people are looking to education to pave the way for a more just social order, because education
instills in the young crucial humanitarian values such as equity, tolerance and peace. Progress in
education was taken to be essential for sustainable development, environmental protection,
improvement in material and child health and participation in democratic social and political
process. Education is also currently becoming the single most important contributor to national
economic growth. Access to good-quality schooling is thus, at central importance to national
development (Harb N. 2006)

Schooling was perceived, as an important determines schoolings were considered as an

instrument for affections both the national economy and the individual income and earnings.
Hence, schooling might be considered as an important tool for economic growth at the nation
along with functioning of democratic forms, which justify for an important component at public
investment (Amitava R. 2010). Measuring at academic performance at students is challenging
since student performance is product at socio-economic, psychological and environmental
factors. For the last 20 years, education in Ethiopia is growing as a profitable industry with prime
objective of maximizing profit by delivering high quality education that produces well-educated,
skilled, mannered students according to needs and requirements of the dynamically growing
market. With the introduction of rapid expansion of the education system in Ethiopia,

2
fundamental change in approach to the problems of teaching and learning is a necessity for
improving quality (Syed T and Raza N. 2006)

That is, why the scope of research is always there to find out what are the school related factors
and school variation on academic performance of students this study focuses on investigating on
academic performance of student on Bate, Gada and Model primary school on grade Eight (8)
students. Between this three school there may be variation or not. For check their variation the
research will must be necessary.

1.2 statement of the problem

 What are the factors motivate in school related factors and school variation on
academic performance of the student?
 Which variables more affect school related factors and school variation on
academic performance of student?
1.3. Objectives of the Study

1.3.1 General Objective

The main objective of the study will be investigating the school related factors and school
variation on academic performance of the students.
1.3.2 Specific Objective
- To identify the main factors to motivate in school related factors and school variation on
academic performance of the students.
- To identify by which variables the variation exists, between those three schools.
- To give direction for the raised problems based on the output of the study

1.4 Significance of the Study

The purpose of this study are, expected to have a practical uses in teaching and learning
process in the school and to identify by which variables are the variation exist between
those three school. It helps in the administrative office in being aware of the variables
that has positive and negative effects. Policy makers and other concerned man use the
result of this study in implementing educational policy. It may also help the text book
procedure to prepare the materials by considering the suggested solutions. It may also

3
 serve as initiating idea who is interested to make further studies on the same area and
problems

CHAPTER TWO

2. LITERATURE REVIEW
Most of research review supports the hypothesis that the students achievement depends on socio-
economic, psychological and environmental factors. The findings of the research stated that
student performance is affected by different factors in different school. Such as learning ability
because, new paradigm about learning ability assumes that all students can and should learn
his/her level. However, it should not be considered as constraint. Since there are other factors
like race, gender, sex that can affect student’s performance (Hansen, 2000).

2.1 Library Service

It is an intellectual service centers of a good university to potent, to hand the instrument

educations, to adaptive progressive method of student’s participation and to enrich, (Koch,
1985). Library is an essential channel by which the wisdom of the past can be brought to aid or
to satisfy the needs of the present. Students need reading materials to enrich their knowledge on
the subject matter. In addition to this, students need reference materials to understand more what
they learn. This means that the absence of reference material can challenge them (Glob, 1994)

2.2 Time for Study

How students spend their time in and out of school receives considerable attention in discussions
about children’s school performance. We know that 91% of children’s time from birth to age 18
is spent outside of school (Usdan, 1990) Once students start school, 70% of their waking hours
are spent outside of school (Clark, 1990). The aim of time study is to establish a time for a
qualified worker to perform specified work under conditions and at a defined rate of working.

2.3 Parental Involvement

The literature review for the effect of family on children’s learning was updated in July 2006. To
accomplish this task, recent manor research syntheses in family, school partnership and
family/parent involvement, which are listed in the additional readings, were used to identify

4
studies; almost without exception the original study was obtained and read. Also, a hand search
of journals within the past six years I education and educationally-related fields was conducted.
To the extent that these resources focused on teachers and community, information was added to
these areas. Illustrative examples are teacher outreach to parents /families and key practices
schools use to engage families from diverse backgrounds. Parent bolster student’s performance
when they provide a positive educational experience through strong, consistent values about the
importance of education; demonstrate a willingness to help student and intervene at school; and
become involved. Continuity across home and school about students learning is an important
protective factor. Consensus and consistent in the partnership between home and school
strengthens a positive, pro-education message. The relationship between home and school risk
factor for student’s learning (Christenson and Sheridan, 2001).

2.4 Number of student in a class

The number of students in a class is also a factor that affects the suitability of teaching method
which of course greatly influences the effectiveness of learning. Classes of manageable number
of students will encourage the teacher to have class activities and give a chance to students to
participate. On the contrary, in a large number of students in class besides disciplinary and
suffocation problem the teacher may be forced to resort to lecture method. (Taddesse 1997)
referring to the use of different methods in large and small number of students in a class.

2.5 Text Books

The significance of text books to teachers and students is beyond question. “the text is to the
teacher what the spade is to the gardener, the chisel and saw and driver to the wood worker, the
type writer to the typist” A teacher man not know what to teach and why to teach unless he/she is
guided by a text book. “A text book can set a teacher on the right track to see the principle more
clearly and to understand the purpose and procedure of his/her method more completely and
precisely.” (Taddesse, 1999:3)

2.6 School Administration

The teaching and learning process in school has facilitated and governed by the administrative
state of the school. The school administrators (UNESCO, 19997) enhance the network between
teachers, students, families, educational bureaus and other governmental and non-governmental

5
organization. Each teacher (Directors supports, were generally organized as, an essential element
in fostering teacher’s development. They are usually responsible for the allocation of resources
and can mobilize support from within the school, from the community, and from higher
authorities. They also have the line management responsibilities for staff and therefore in a good
position to produce encouragement and leader ship (UNESCO, 19997)

2.7 Class Room Interactions

Teacher activities and the way they approach their students influence class room interaction. If
there is good interaction in the class room the students will be interested to learn and participate,
follow to the teachers and explanation and then they express their views, answer questions and
carry tasks, activities (Globa, 1994)

6
 CHAPTER THREE

3. METHOLODOLOGY
3.1 Study Area
The study will conduct in Bate, Gada and Model School which located 525km East Hararghe
zone which is far 525km from Addis Abebe the capital city of Ethiopia. Model School was
established around 1952 according to E.C and Bate School was established around 1950 both are
started teaching and learning activities in different year according to E.C.

3.2. Study Population

A cross sectional study that included all target population will conduct on this study to
investigate the school related factor and school variation on academic performance of students.
This study will be conducted on two different schools on grade eight students.

3.3. Study variables

There are two variables in this study.

Dependent Variable is:-Performance of students

Independent Variables are:-

- Library service
- Parental Involvement
- Rule and Regulation of the school
- Availability of text books
- Relationship between teachers and students
- Number of Students in a class
- School location
- Time for study

3.4 Sample Size Determination

The target population of this study is students in Bate,Gada and Model School. Since it is
difficult to investigate the whole population, the sample will take from the population by using

7
 sampling technique. In this study the target population is heterogeneous that means students in
Bate School ,Gada and Model School are heterogeneous according to their learning process.

Therefore in this study the samples size of the study will be select by sampling techniques of the
stratified random sampling. Stratified random sampling is a sampling technique which used
when target population is heterogeneous. To determine sample size the simplest way will be
conduct a pilot survey since there is no previous result and difficult to guess the distribution of
the population. Pilot survey is uses to determine sample size and checks the appropriate usage of
question through the main variable. For this proposal the questionnaires should be proper to all
Bate, Model and School grade eight students. The sample size determination formula that is
adopted in this study will be:
3 2
d
n0 = ∑ W h S 2h for, V = Z 2 , d=marginerror
h=1
0.05
V 2

3
W yh
h=
Nh , and y h= ∑
N
i=1 n
n0
If <5% in this case n 0= n otherwise calculated by using the following formula:
N
n0
n=
n0 Then sample size for each stratum are obtained by using
1+
N
Nh
n h= ∗n Where, Sh- standard deviation of stratum
N
W h - Weight of stratum
yh - mean of stratum
V - Specified variance
Z α =1.96 at 95% α level of significance
2

ℇ - Absolute error which is 4 % or 0.04 to maximize

sample size.

In this study the population is classify as the following stratum;

8
 Stratum 1Bate School: Grade 8th students

Stratum 2Model School: Grade 8th students

Stratum3 Gada school: Grade 8th students

Based on sample survey the determines sample size of the three strata will be calculate

as;

Stratum 1 (N1)

Stratum 2 (N2)

Stratum 3(N3)

Proportional allocation:-

Proportional allocation is used when the size of the sample from a given stratum is

proportional to the size of stratum. Therefore by apply equal proportional allocation for each

stratum to select the sample from each stratum will found by using the formula:

Nh * n
nh 
N

Where Nh =total number of population in each stratum

9
 nh=sample size for each stratum

In this research uses Model school primary school as stratum one, Bate as stratum two
and Gada primary school as stratum three.
By using pilot survey I take a sample ten from student from Model primary school, five
from Bate school and five from Gada primary school. Using pilot survey the following
data is conducted.
School name yh Sh Nh Wh
Model school 78.6 10.09 90 0.3
Bate school 79.8 8.98 60 0.2
Gada school 81.9 10.11 150 0.5

161.454
d=0.04∗26.83=2.683 ^
y st =26.83 V =¿ ¿,1.874 n 0= =86
1.874
n0
Since =15 % this is greater than 5% so n can be calculated as follows.
N
86
n= =67
86 To find the sample size for each stratum
1+
300
90
n1 = ∗67=20.1 ≈ 20
300
60
n2 = ∗67=13.4 ≈13
300
150
n3 = ∗67=33.5 ≈ 34
300
Therefore sample size n=13+20+34=67

3.5. The Method of Data Collection

There are two major sources of data collections. Namely primary source of data collection and
secondary source of data collection. Therefore in case of this study the source of data will be use

10
primary source of data collection which will be collect by using self-structured questionnaire.
So, the tentative way of information gathering will distribute as questionnaire to students in Bate
Gada and Model School to collect data

The most crucial methods of data collection in this study will be primary data through
questionnaires by contribute for students.

3.6. Method of Data Analysis

After using different method for collecting data the next step will be the analysis of data, editing,
summarizing and interpreting the appropriate data. The available is data transformed into reliable
and useful information with the help of statistical analysis. i.e. by using SPSS and Minitab
software.

3.6.1. Descriptive Statistics

Descriptive statistics is a kind of statistic which describes the data using tables, graphs and
summary measure from the large original version. Such a data is not very helpful in drawing
conclusions or making decision. For this study the researcher used from descriptive statistics
(frequency table and histogram charts) to describe the response of respondent.

3.6.2. Inferential Statistics

Inferential statistics

Inference is the objective of statistics, especially in decision making and prediction plays
important role in individual life. Therefore, the objective of statistics is to make inference about a
population based on the information of the sample. Since, population is characterized by
numerical descriptive measures called parameters. Statistical inference is concerned with making
inferences about population parameter based on estimating the sample data.

3.7 MULTIPLE LINEAR REGRATION

The general objective of regression analysis is to describe the model relationship between one
dependent variable more than one independent variables. In this study there are more than two
independent variables and one dependent variable.

Assumption of Multiple Linear Regressions Model

11
 1. The relationship between y and x repressors is linear
2. The error term has zero mean
3. The variance of the error term was constant
4. The independent variable and the error term was independent
5. The errors are normally distributed
6. The errors are uncorrelated.
7. 3.7.1 Model Specification

The model should be properly specified (including all relevant variables, and excluding
irrelevant variables)

Yi = βo +β1X1 + β2X2 + β3X3………………..βiXi+ε Montgomery etal, (2008).

Where ,Y i=theresponse(dependent )variable

Xi = independent variables

ε 1= error term of the model

βo, β1, β2, β3, βi are parameters of the model

3.8 Parameter Estimation of the Models

The most popular method of estimating the multiple linear regression parameters (β) is least
square method .The model parameters are estimated by using least square estimation method.

This is given as, β = (x’x)-1x’y

Where, Y = nx1 vector of observation

X= nxp invertible matrix

β = px1 vector of regression coefficients that is estimated from the data

yi=βo + β1x1 +β2x2 +β3x3 +…………………………………+βnxn +ε

3.8.1 Individual t-test of the Model

This indicates the individual test of parameter by using t-test.

Steps:

Step1 identify the hypothesis

12
 Ho: βj=0

H1 β j 0

Step2 critical value i.e. t α/2 (n-k)

^
βj
Step3 calculate test statistics i.e. t cal =
SE ( ^β)

Step4 decision i.e. if |t cal| >t tab then we reject the null hypothesis

Step5 give conclusion based on the above decision.

P-value and Hypothesis Testing

The lower P-value indicates that there is evidence that the null hypothesis is not significance that
means rejected Ho. The smallest P-value shows there is strong evidence to reject null hypothesis.
Reject Ho at α level of significance when P-value is less α value and fail to reject Ho when P-
value is greater than α value.

3.8.2 Assumptions of the model

The major assumptions that will be, made in the study of linear regression analysis will be as
follows:

 The relationship between x and y is linear

 The error term ℇ has constant variance.
 The error term ℇ has zero mean E(ℇ)=0
 The regression model is correctly specified.
 The error term are normally distributed i.e ℇi ~N(o,s2)
 There is n auto-correlation in disturbance E(ℇi ℇj)=0

3.8.3 Parameter Estimation of the model

The model parameter will be, estimated by using least square estimation method. Thus, the least
square estimator of β is β = (X’K)-1 X’Y Where y=nx1 matrix of observation X=nxp invertible
matrix β j=px1 vectors of regression coefficient that was estimated from the data.

The model parameter will be, estimated by using least square estimation method. Thus, the least
square estimator of β is β= (X’K)-X’Y Where y=nx1 matrix of observation

13
 X=nxp invertible matrix β j=px1 vectors of regression coefficient that was estimated from the
data.

3.8.4 The overall tests of model

The overall goodness of the model will be, checked by performing the hypothesis in the

following ways. That is Ho: β1=β2=… βk=0 Vs

H1: βk ‡ 0 (at least one parameter different from zero)

The test statistics is F-distribution and its computation based on the following table.

Source of SS DF MS F Sig.
variation

Regression SSR p-1 MSR MSR/MSE

Error SSE n-p MSE

Total SST n-1

Then we reject Ho if alpha value is greater that the corresponding p-value and do not reject
otherwise.

3.8.5 Model Adequacy Checking

This model will be, used to check whether the major or standard assumptions of multiple linear
regression analysis are meeting. This will be through the following. Residual analysis it is a
measure of variability y that left unexplained by the regression model. Thus, and departures from
the assumptions on the errors should show up in the residuals.

It analysis effective through plotting: the plotting are

- Normal probability plot

- Plot of residuals against fitted value.
Test of Multicollinearity

Multicollinearty test is the correlation between the repressors (explanatory) variables. It was
measured by variance inflation factor (VIF) .When VIF>10 there is multicolliniarity. But VIF<

14
 10 there was no multicolliniarity and it fulfills the assumption multiple linear regression which
the repressors are not correlated.

Note that: VI ( ^β ¿ =1/ (1-R2) (Author,1953).

Where,

 Coefficient of multiple determination (R2)

The linear regression model of deforestation by using Coefficient of multiple determinations (R2)
and Adjusted R2

 Linearity: - It is assumed that the relationships between dependent and independent variable
were linear.
 Residual Analysis:-it is conducted after estimating regression equation to assess the model
assumption through graphical techniques. Its analysis is through plotting. These plotting are:-

 Normal Probability Plot: - it is about the normal assumption of the multiple regressions by
plotting residuals against the fitted values.

 Assumption normally

 It is assumed in multiple regression that the residuals (observed minus fitted values) are
distributed normally (i.e., follow the normal distribution). Again, even though most tests
(specifically the F-test) are quite robust with regard to violations of this assumption, it is always
a good idea, before drawing final conclusions, to review the distributions of the major variables
of interest. You can produce histograms for the residuals as well as normal probability plots, in
order to inspect the distribution of the residual values.

 Histogram with Normal Curves: - it checks about the normality assumptions.

3.9 The Overall Tests of the Model

Statistical hypothesis is statement about the nature of population. It is often stated in terms of
population parameters. To test the overall model always to use the ANOVA table.

15
ANOVA table

Table 3.1

Source of variation Df SS MS F

Regression k-1 SSR MSR MSR/MSE

Residual n-k SSE MSE

Total n-1 SST MST

SSR = ^β 'x'y - n y 2

SSE = y'y - ^β 'x'y

SST = y'y - n y 2

Where,

df – degree freedom

SSR - sum of square of regression

SS – sum of square

SSE - sum of square error

SST -sum of square total

MSE -mean square error

16
MSR - mean square regression

The Overall goodness of the model is checked by performing the hypothesis in the following
ways;

1. Ho: β1=β2=β3 = β4=0 (there is no significance relationship between different predictors and
dependent variable)

H1: β1≠ β2≠ β3≠ β4≠ 0(there is significance relationship between different predictors and
dependent)

2. Specify level of significance (α)

3 Test of statistics

Fcal=MSR/MSE

4. The critical value Fα (K-1) , (n-k)

5 .Decision rule

Reject Ho if Fcal> Fα(K-1 , n-k)

Do not accept Ho if Fcal< Fα(K-1 , n-k)

After fitting the linear model of the given data set, an assessment is made of the adequacy of fit.
The strength of the linear relationship between y and the set of prediction x 1, x2,..xkcan be
accessed through the examination of the scatter plot y versus ŷ and the coefficient of multiple
determinations. The R2(coefficient of multiple determination) may be interpreted as the
proportion of the total variability in the response variable y that can be counted for by the set of
predictor variable x1,x2…xk. In multiple regressions, R2 measures the relationship between one
variable y and set of predictors x1, x2…xk.

If intercept is included and the number of explanatory is more than one, R 2 also referred to as the
coefficient of correlation, in such cases, the coefficient of determination ranges from 0 to 1

Formula: R2=1- SSerr/SStot

Where, SSE=sum square error, SST= sum square total

Adjusted R2:-The use of an adjusted R2 is an attempt to take account of the phenomenon of the
R2 automatically and spuriously increasing when extra explanatory variables are added to the

17
model. It is a modification of R2 that adjusts for the number of explanatory terms in a model
relative to the number of data points. Its value is always less than or equal to R 2.It is used to
check the correctness of the linear regression model (it used to check the goodness of fit of the
model).Adjusted R2=1-(1-R2) (n-1) /n-k Where,

R2=coefficient of multiple determination n= number of observation, k= number of paramet

CHAPTER FOUR

4. RESULT AND DESCUSSION

4.1 Descriptive statistics

Table 4.1

Variable Frequen perc Valid Percent Cumulativ

cy ent e percent
Library service 3 53.7 53.7 53.7
Yes 6 46.3 46.3 100
No 3
1
Parental Involvement 3 52.2 52.2 52.2
poor 5 34.3 34.3 86.6
good 2 13.4 13.4 100
very 3
good 9
School Administration 3 47.8 47.8 47.8
Yes 2 52.2 52.2 100
3
No 5
Rule and regulation Yes 3 53.7 53.7 53.7
No 6 46.3 46.3 100
3
1
Text book 1:3 2 32.8 32.8 32.8
1:2 2 32.8 32.8 65.7
1:1 2 34.3 34.3 100
2

18
 2
3
Relation b/n teacher poor 6 9 9 9
And student good 3 52.2 52.2 61.2
Very 5 38.8 38.8 100
good 2
6
Number of student 31-40 3 46.3 46.3 46.3
In a class 41-50 1 29.9 29.9 76.1
Above 2 23.9 23.9 100
50 0
1
6
Advantage of time No 3 49.3 49.3 49.3
Spent on study Yes 3 50.7 50.7 100
3
4
Distance of home 0.5km 1 28.4 28.4 28.4
1km 9 25.4 25.4 53.7
2km 1 23.9 23.9 77.6
Above 7 22.4 22.4 100
2km 1
6
1
5
Study hour 1- 1 25.4 25.4 25.4
2hour 7 26.9 26.9 25.2
2- 3hour 1 23.9 23.9 76.1
3- 4hour 8 23.9 100
Above 1
4hour 6
1
6
Relation b/n parent Yes 3 47.8 47.8 47.8
And school administration No 2 52.2 52.2 100
3
5
Name of school model school 2 32.8 32.8 32.8
Bate school 2 41.8 41.8 74.6
Gada school 2 25.4 25.4 100
8
1
7

19
The result in table 4.1 shows about 36( 53.7%) of the respondents (students) have say there is
enough library service and about 31(36.3%) of the students say there is no enough library
service. Generally the majority of the students say there is no library service.
Also from the above table shows about 35(52.2%) of the students are says parental involvement
on economically are poor, about 23(34.3%) of the students are says good and about 9(13.4%) are
says very good. Generally more than half of the students are somewhat good parental
involvement on economically.
Similarly, about 32(47.8%) of the students are says the school administration are affects yours
academic performance and about 35(52.2%) says the school administration are not affects yours
academic performance. Generally more than half of the students are says there is no problem
your academic performance behind your school administration.
Similarly, shows about 326(53.7%) of the students are says the rule and regulation of school
affect your academic performance and about 31(46.3%) says the rule and regulation of school
doesn’t affects your academic performance. Generally more than half the student says the rule
and regulation school doesn’t affect your achievement.
Again shows about 22(32.8%) of the students says one text book for three students or 1:3, about
22(32.8%) of the students says one text book for two students and about 23(34.3%) of the
students are says one text books are for one students. Generally more than half of student is take
one textbook for three.
Also shows about 6(9%) of the students says the relationship between teacher and students are
poor, about 35(52.2%) of the students are says good and about 26(38.8%) of the students are
says very well.
Similarly, shows about 31(46.3%) of the students are says the number of student in a class
from31-40, about 20(29.9) of the students are says from 41-50 and about 16(23.9%) of the
students are says above 50.
On the same table shows 33(49.3%) of the students are says the advantage of time spent on study
is important and 34(50.7%) of the students are says important.
Similarly, shows 19(28.4%) students are 0.5km far from school, 17(25.4%) students are 1km far
from school, 16(23.9%) students are 2km far from school and 15(22.4%) students are above
2km’s far from school. This implies that the majority of students are 0.5km far from the school.

20
Also, shows about 17(25.4%) of the students are uses one hour for the study, about 18(26.9%) of
the students are uses two hours for study, about nine students are uses three hour for study and
about thirteen students are uses above three hour for study.
And, shows about 16(23.9%) of the students are there is no relationship between yours parents
and school administration and about 16(23.9%) of the students are there is a relationship yours
parents and school administration.
4.2 Analysis of Variance (ANOVA)
One-way ANOVA: performance of student versus school name

ANOVA
The averarge of students

Sum of Squares df Mean Square F Sig.

Between Groups 919.250 2 459.625 2.207 .118

Within Groups 13330.335 64 208.286
Total 14249.586 66

Source DF SS MS F P
School 2 919.250 459.625 2.207 .118
Error 64 13330.335 208.286
Total 66 14249.586

Since p-value is greater than alpha level of significance so fail to reject null hypothesis. Thus, all
mean are same i.e. there is no mean difference.

4.3 Results from Inferential Statistics

4.3.1 Multiple Linear Regression Model

ANOVAa
Model Sum of df Mean F Sig.
Squares Square
Regressio
6844.823 12 570.402 4.160 .000b
n
1
Residual 7404.763 54 137.125
Total 14249.586 66

21
a. Dependent Variable: The averarge of students
b. Predictors: (Constant), The name of school, Relationship betwen
students and teachers, Parental involvement, Rule and regulation of
school, Time of study, Availability of text books, Advantage of more time
spent on study, Library service, Number of students in the class, The
relationship betwen administration and parents, School location, School
administration

Table 4.2
The ANOVA table is used to test the hypothesis of the overall significance of the model. It tests
The overall significance test
Since P-value less than alpha level of significance (i.e. 0.00<0.05) so we can reject the null
hypothesis. I conclude that, over all F-test are significance differences. That means the
explanatory variables (library service, parent’s involvement, and delivery ability of teacher and
distance of home from school, school administration, rule and regulation of school, number of
student in class, name of school) are significant effect on response variable (academic
performance of student).

Table 4.3

Coefficientsa

Model Unstandardized Standardized t Sig. Collinearity

Coefficients Coefficients Statistics

B Std. Error Beta Toler VIF

ance

1 10.73
(Constant) 74.385 6.931 .000
2

Library service -8.986 3.049 -.307 - .005 .885 1.130

2.947

22
 Parental involvement 7.913 2.169 .386 3.648 .001 .860 1.163

School administration 10.130 3.425 .347 2.958 .005 .699 1.430

Rule and regulation of -

-7.092 3.086 -.242 .025 .865 1.157
school 2.298

Availability of text books -.449 1.877 -.025 -.239 .812 .865 1.156

Relationship betwen
-.723 2.555 -.031 -.283 .778 .807 1.239
students and teachers

Number of students in the -

-4.078 1.891 -.226 .036 .878 1.139
class 2.156

School location 3.653 1.456 .281 2.509 .015 .769 1.301

Time of study -.459 1.345 -.035 -.341 .734 .917 1.091

Advantage of more time

2.248 3.018 .077 .745 .460 .899 1.112
spent on study

The relationship betwen

administration and 1.876 3.129 .064 .599 .551 .838 1.193
parents

-
The name of school -5.889 2.162 -.307 .009 .760 1.316
2.724

a. Dependent Variable: The averarge of students

From the above table 4.3, p-value for the coefficient of library service is less than alpha level of
significance (0.005<0.050), indicating rejection null hypothesis. Thus library service has
significant effect on academic performance of students and for unit increment of library service
the academic performance of student decreased by -8.986.

Similarly, p-value for the coefficient of parent’s involvement is less than alpha level of
significance (0.001<0.050), indicating rejection the null hypothesis. Thus, the parent’s
involvements on economically have significant effect on academic performance of students and
for unit increment of parent’s involvement economically the academic performance of student
increase by 7.913.

Again, p-value for the coefficient of school administration is less than alpha level of significance
(0.005<0.050), indicating rejection the null hypothesis. Thus, the availability of school
administration has statistically significant effect on academic performance of students and for

23
unit increment of availability school administration the academic performance of student
increase by 10.130

Also, p-value for the coefficient of rule and regulation of school is less than alpha level of
significance (0.025<0.050), indicating rejection the null hypothesis. Thus, rule and regulation of
school has statistically significant effect on academic performance of students and for unit
increment of rule and regulation of school the academic performance of student decrease by -
7.092.

And, p-value for the coefficient of distance of home from

school is less than alpha level of significance (0.00<0.015),
indicating rejection the null hypothesis. Thus, the distance
of home from school has statistically significant effect on
academic performance of students and for unit increment
of distance of home from school the academic performance
of student increase by 3.653.

Model Summary

Adjusted R Std. Error of

Model R R Square Square the Estimate

1 .480
.693a .365. 11.71005

Table 4.4
According to the given fitted model, about 79.7% of the variation in the performance of students
is explained by the explanatory variables and this indicates the regression model well fit model.
Checking assumption of multiple regression model
This residual plot is necessary to see weather error term are normally distributed or not as stated
by in the assumption of regression equation and to see linearity between the response
variable(academic performance of student) and explanatory variable some of the diagnostics

24
tools checking regression assumption include:

From above histogram indicates the error term is assumed to be normally distributed with mean
zero and constant variance. Therefore normality assumption is satisfied.

25
The above p.p plot of regression standardized residual indicates that the point in the plot form
straight line, is normally distributed this show approximately linear pattern that is consistent with
normal distribution. Therefore the linearity assumption is satisfied.

26
The above scatters plot the residual shows that there is no pattern; it indicates that there is no
hetroscadisticity problem in the model. This implies that whatever the value yields, the error
term has constant variance indicating homoscedasticity.

27
 CHAPTER FIVE

5. CONCLUSION AND RECOMMENDATION

5.1 Conclusion
Based on the above result and discussion the following points were conclude on school related
factors and school variation on academic performance of students. From the above SPSS and
Minitab output result by using multiple regression analysis and analysis of-variance ANOVA
output library service, parents involvement, availability of textbook, school location and number
of student in the class and rule and regulation of school are the main school related factors on
academic performance of students. That is the academic performance of student is affected by
those explanatory variables.

5.2 Recommendation
Based on the final findings or results the following recommendations were forwarded. Since
library service, parental involvement, availability of textbook and distance of school from home
are main school related factors on academic performance of students. To increase the academic
performance of student’s governments, school administration and any concerned body give more
attention to solve this problem by fulfilling the solution for problem. That is producing enough
library service for students in the school to increase the academic performance of students.
Parents fulfill their child’s facilities to make the academic performance of students also as
availability of text book for students, build school in society or in each Keble and also
government employed the performed and delivered teachers for the students.

28
 Reference

 Amitava. R (2010). Factors affecting student’s academic performance.

 Christenson and Sheridan (2001) - parental involvement on academic performance
of students.
 Cochran William (1997, 3rd edition sampling technique)
 Globa (1994). Absence of reference material may affect student’s achievement.
 Hansen .J (2000). Academic achievement of student depends on socio-economics,
physiological and environmental factors.
 Harb. N (2006). Factors affecting student’s performance united Arab Emirate
University, access on http:/mpra. Ub. Uni
 Kebede Likebo (2001). Major factors affecting student’s Academic performance in
secondary school of Wolayita zone.
 Olaniyi Bojuwaye (1993) A case study of school related factors affecting Nigerian
secondary school pupil’s academic performance, Journal of education.
 Sanders (1999) .About teacher Quality.
 Syed .T and Raza .N (2006). Factor affecting student’s performance.
 Taddesse Lema (1999) - The student’s low level of English proficiency in some
selected senior high school.
 UNESCO (1999)- school administration
 Usdan and Clark (1990) - about children’s school performance.

29
 APPENDIX A
HARAMAYA UNIVERSITY

Department of statistics

Questionnaire

First of all, would like to thank you voluntarily and willingness to fill this questionnaire. The
main objective of the research to investigate the school related factors and school variation on
academic performance of the student.

Instruction: - For the following question fill the appropriate information in the space provided
or tick the right mark () for question in the form of choice.

1. Write the name of your school? - - - - - - - - - - - - - - - - - - - - - -

2. What is your average? - - - - - - - - - - - - - - - - - - - - - -
3. Do you have enough library service? Yes  No 
4. What is your parent’s involvement on economically to support you?
Poor  Good  Very good 
5. Is there a problem on your academic performance behind your school
administration in your school? Yes  No
6. Do you think that the rule and regulation of your school affect your achievement?
Yes  No
7. If your answer is “yes” for question 5, how? - - - - - - - - -- - - - - - -- - - - - - -
8. Do you have enough text books? 1 by 3  1 by 2  1 by 1 
9. What is the relationship between you and your teacher?
Poor  Good  Very good 
10. Number of student in a class? 31-40  41-50  above 50
11. Is there advantage more time spent on study?
Yes  No 
12. Distance of your home from school in kilometers?
0.5km  1km  2km  above 2km
13. How much time spent on study?

30
 1hour  2hours  3hours  above3 hours

14. Is there a close relationship between school administration and the parents?

Yes No

15. Write the general learning or teaching style of the school?

31
 APPENDIX B
Variable coding

{
library service = 0=Yes
1=No

{
1=Poor
Parent’s involvement = 2=Good
3=Very good

School administration= 0=Yes

1=No {
Rule∧regulatio= 0=Yes
1=No {

{
0=1 :3
textbook = 1=1 :2
2=1 :1

{
0=Poor
relation shipbetween teacher∧student = 1=Good
2=Very good

{
0=31−40
number of student ∈ a class= 1=41−50
2=above 50

Advantage of spent more time on study=¿

{
0=0.5 km
distance home ¿ school= 1=1 km
2=2 km
3=above 3 km

32
 {
0=1 hour
study hour= 1=2 hour
2=3 hour
3=above 3 hour

{
relation shipbetween parent ∧schol administration= 0=Yes
1=No

33