UNIVERSITI TEKNOLOGI MARA

KAMPUS KOTA SAMARAHAN, SARAWAK

COURSE : STATISTICS & PROBABILITY

COURSE CODE : STA108
SEMESTER : MARCH – JULY 2020
GROUP : AS1204D

PROJECT TITLE

“A study on sleeping times and score in exams : The

descriptive measures and simple linear regression”

GROUP MEMBERS
NO NAME STUDENT ID SIGNATURE
.
1 ANGJELLY ALLYA ANAK NGAUH 2018236062
2 CASSY LUGAK 2018433928
3 THIERRY HENRY ANAK DANIEL 2018643182
4 VIVIANA ANAK VERY 2018671504
CONTENTS

TITLE PAGES
CHAPTER 1 : INTRODUCTUON
1.1 : Introduction of Study 1
1.2 : Objective(s) of Study 2
1.3 : Significance of Study 2
1.4 : Limitation of Study 2
CHAPTER 2 : METHODOLGY
2.1 : Data Description 3
2.2 : General Technique 4-5
2.3 :Numerical Technique 5
CHAPTHER 3 : RESULTS AND INTERPRETATION
3.1 : Data Presentation 6-7
3.2 : Descriptive Statistics Analysis 8-9
3.3 :Correlation and Regression Analysis 10 -11
CHAPTER 4 : CONCLUSION
4.1 :Report Summary 12
5.0 : Appendix 13-16

CHAPTER 1 : INTRODUCTION
1.1 Background of study
According to the University Health Center, college students receive 6 to 6.9
hours of sleep each night. This is often due to an intense schedule of classes, work,
extracurricular activities and social activities. Other than that, research from various
health and sleep studies have shown that college student who do not receive proper
amount of sleep per night negatively affect their mood, health and their performance
on study and exam.

Students seem to be at more risk for not receiving the right amount of sleep
compared to other individuals. Even though 8 hours is the recommended amount of
sleep for college students to get, that is bot probably going to happen. Reported by
Geneseo at The State University of New York, even not getting 10 hours of sleep at
night can make a college student feel unrest.

This study sought to investigate the relationship between the time taken for
students to sleep on Sunday and the score their get on Mathematics exam.
University students are required to perform of demands level. In addition, sleep
pattern a to university likely to change from high school due to alteration such as
school schedule and lifestyle preference.

The time taken for sleeping on the day before examination defined the test
scores by students for upcoming test on Monday and was measured from 12 am to
11.59 pm on Sunday. The test scores by the student on Mathematic are in range
50% to 100%. This data will show whether the number of minutes the students to
sleep spend affected their test result in examination.

From the study, the time taken for students to sleep (minutes) is an
explanatory variable and the test scores by the students is response variable. This is
because test scores are depending how long the students spent their students spent
their Sunday to sleep.

1
1.2 Objective of study
The objective of this study are as follows:
I. To investigate the relationship between the time taken for students to
sleep on Sunday and the score their get on Mathematics exam
II. To describe the numerical measure for time taken for students to
 sleep.

1.3 Significance of study

The findings of this study will redound to the benefit of society
considering that the amount of sleep per night plays an important role in our
life. An inconsistent sleeping pattern is a bad lifestyle that can affect health,
mood and also daily performances especially students and those who work.
This study can help student or other people to reconsider back their sleeping
habit and change for a better lifestyle. An enough amount of sleep per night
surely can balance their health and education. For the researcher, this study
can help them to uncover critical areas in the processing the data using
statistic that many researchers were not able to explore. Thus, a new theory
on how to read a data deeply may be arrived in future.

1.4 Limitation
The limitation of this particular study is that the we only measured one
activities that affect by the amount of sleeping per night.. Although we know that
studying can come at the expense of sleeping, we do not know how other
activities interact with studying and sleeping, nor how this activities might be
associated with academic functioning. Further research could use similar daily
and yearly methodology to examine wide array of activities that can be affect by
the amount of sleeping.

Similarly, our study is limited in that we only assessed the measure of

sleeping amount and the effect on exam score. We did not, for example, assess
the quality of student studying and their sleeping amount. Without knowing
exactly how students were spending their study time before the test, its is unclear
how study strategy might interact with study time, sleeping amount, and the
academic problems. It is possible that studying more than usual is associated
with poorer quality of sleep which reduce the amount of sleep per night.

2
CHAPTER 2 : METHODOLOGY
2.1 DATA DESCRIPTION
The data that we use for this case study STA108 is form secondary data group,
it is because we collect it for the internet. The secondary data is the data that
collected for other parties. Such as journals, newspapers, internet etc. The data that
we collect have two (2) variable which is the time taken to sleep and the score that
the student get from the Mathematics test. The data is from the quantitative
continuous variable because it can be measure by numerically such as the score that
the student get from their test and it can assume any numerical value over a certain
interval or intervals.

The independent variable (X) from the data that we use is sleeping times. It is
because sleeping times is the variable does not depend to other variable and it set by
the experimenter. On the other hand, the dependent variable (Y) in this case study is
score in exam. This variable is depending on the independent variable, so score in
exam is depending on the sleeping times before student taking the examination.

The sample size of this data is 14 students that take the test. The link that we
used to collect the data is ( https://www.scribd.com/document/463060120/STA-104 ).

3
2.2 GRAPHICAL TECHNIQUE
In the case study we use the histogram to label our data, there are 2 histogram
the we get. The first histogram is related to the independent variable which is the
time taken to sleep. For the histogram the graph that displays that data by using
vertical bars of data to represent the frequency of the class. For histogram it was
used class boundaries for (x-axis) and the frequency of the classes at (y-axis).

Figure 2.2a : Independent variable

The second histogram is the histogram for the dependent variable which is the
score that the students get in examination. Moreover, it still the same that For the
histogram the graph that displays that data by using vertical bars of data to represent
the frequency of the class. For histogram it was used class boundaries for (x-axis)
and the frequency of the classes at (y-axis).

4
 Figure 2.2b : Dependent variable

2.3 NUMERICAL TECHNIQUE

The numerical technique that we use to measure the independent variable is
time taken that the students use to sleep before their examination in minutes. The
scale that we use is in the ratio scale. It is because, the time taken that we use in this
case study have the true zero, so, it make the variable is in the ratio scale.

Secondly, the numerical technique that we use to measure the dependent

variable is by the score percentage that the students get from their examination mark
in percent (%). The scale the we use is the same as the independent variable
because the score percentages have the inherent point which meant by having the
true zero in the data.

5
CHAPTER 3 : RESULTS AND INTERPRETATION
3.1 Data Presentation

Fig 1.1

Fig 1.2

6
3.1.1 Sleeping Times (x-variable)

From the graph shows in figure 1.1 it is a symmetric data set,it represents the
amount of sleeping time among the students.Based on the histogram,the most
average sleeping time by the students to make revision for Mathematics subject is
170 minutes.Other than that,the mean and the standard deviation for this graph are
281.79 minutes and 83.404 respectively.

3.1.2 Score in exams ( y-variable)

The graph from figure 1.2 represent a y-variable which is score in exams and the
graph are symmetric data set.Based on the histogram,the most highest test score of
the students is 90%.The mean and the standard deviation for this graph is 72.93 and
13.691 respectively.

7
3.2 Descriptive Statistic Analysis
This section explain the descriptive analysis for quantitative variables which was
obtained from this study which is sleeping time and score exam of the students.

3.2.1 Sleeping time of the students

Statistics on Sleeping Time

Valid : 14
Missing :0
Mean 281.7857
Median 285.0000
Mode 300.00
Std. Deviation 83.0372
Variance 6956.181
Range 245.00

BOXPLOT OF SLEEPING TIME

0 50 100 150 200 250 300 350 400 450

minutes of sleeps

The median number of minutes of sleeping is 285.00.The range is about 245.00

minutes of sleep which indicates about 30.53% of the students between 180 to 240
minutes. The first quartile is 202.5 and the third quartile is 345.0 The distribution is
skewed to the left.

8
3.2.2. Exam Score of the students
 Statistics on Exam Scores
Valid : 14
Missing :0
Mean 72.9286
Median 70.0000
Mode 55.00
Std. Deviation 13.69146
Variance 187.456
Range 40.00

EXAM SCORES

0 10 20 30 40 50 60 70 80 90 100

The median of the exam scores by the students is 70. The range of exam score
by the students are 40 marks. The first quartile is 61.25 while the third quartile is
84.75. The distribution is skewed to the right.

9
3.3 Correlation and Regression analysis
From this part,there will be a scatter plot that show the relationship between
independent variable and dependent variable.
3.3.1 : Scatter plot
Scatter plot for sleeping times and exam scores for the Mathematics
test.

Figure 1.3

For the figure 1.3 the scatter diagram is showing the positive correlation.So,for
positive correlation it was shown the existence of positive relationship between two
variable which is sleeping time (mins) and exam scores in their examination.The
direction of change for both x and y axis,if x-axis is increases,the y-axis is increases
too.

10
3.3.2 Correlation
Next,from this part the Pearson correlation,r actually to measured of linear
relationship between two quantitative variables which is sleeping times and
 exam scores in Mathematics exam.The value of pearson correlation,r is
0.957.Hence,there is very strong positive linear relationship between the
sleeping times and exam scores.

Other than that,the coefficient ,r2 which is to determine the ratio of the
explained variation to the total variation.If the correlation coefficient r = 0.957,the
the coefficient ,r2 = 95.7%.Therefore,this implies that 95.7% of the total variation
in exam scores is due to the sleeping times and 4.3% is due to other factors.

3.3.3 Simple linear regression

The regression equation for scatter diagram in figure 1.3 is y= 28.659 +
0.157x. The value of a=28.659 and for the value of b = 0.157 .b= 0.157 indicates that
when sleeping time increases by 1 minutes , the exam score in Mathematic will
increases by 0.157.

11
CHAPTER 4: CONCLUSION
4.1 Report Summary
In conclusion, exam score with sleeping time (mins) will affect the marks of
student gain in exam. In this data, we can see that the highest percentage 10.77% of
students sleeping for 425 minutes in order to get the highest mark in exam which is
95 marks. Thus, 4% of the students that sleeping less ( 180 to 190 minutes) score
averagely 55 to 60 marks.

Thus, we can conclude that in order to have a highest mark in exam, students
should sleeps more than 425 minutes before exam in order to obtain a good score.
Hence, this data shown that a good sleeping time will affect the students score in
exam.

Moreover, from the correlation and regression analysis, we find out that there is
a very strong positive linear relationship between the exam scores and the minutes of
sleeping time for exam in Mathematic. Furthermore, as there is an increase in
minutes of sleeping time the exam score of the students will increase as well.

12
5.0: APPENDIX
SPSS OUTPUT

CHAPTER 2
 Descriptive Statistics
Maximu Std.
N Range Minimum m Sum Mean Deviation Variance Kurtosis
Std
Statistic Statistic Statistic Statistic Statistic Statistic Statistic Statistic Statistic Err
Sleeping Time 14 245.00 180.00 425.00 3945.00 281.785 83.40372 6956.18 -1.309 1
(mins) 7 1
Exam Scores (%) 14 40.00 55.00 95.00 1021.00 72.9286 13.69146 187.456 -1.404 1
Valid N (listwise) 14

Statistics
Sleeping Time
(mins) Exam Scores (%)
N Valid 14 14
Missing 0 0
Mean 281.7857 72.9286
Median 285.0000 70.0000
Mode 300.00 55.00a
Std. Deviation 83.40372 13.69146
Variance 6956.181 187.456
Skewness .268 .165
Std. Error of Skewness .597 .597
Range 245.00 40.00
Minimum 180.00 55.00
Maximum 425.00 95.00
Sum 3945.00 1021.00
Percentiles 25 197.5000 59.5000
50 285.0000 70.0000
75 356.2500 85.7500
a. Multiple modes exist. The smallest value is shown

13
14
CHAPTER 3

Correlations
Sleeping Time
(mins) Exam Scores (%)
Sleeping Time (mins) Pearson Correlation 1 .957**
Sig. (2-tailed) .000
N 14 14
Exam Scores (%) Pearson Correlation .957** 1
Sig. (2-tailed) .000
N 14 14
**. Correlation is significant at the 0.01 level (2-tailed).

15
 Model Summary
Adjusted R Std. Error of the
Model R R Square Square Estimate
1 .957a .916 .909 4.13335
a. Predictors: (Constant), Sleeping Time (mins)

Coefficientsa
Standardized
Unstandardized Coefficients Coefficients
Model B Std. Error Beta t Sig.
1 (Constant) 28.659 4.028 7.116 .000
Sleeping Time (mins) .157 .014 .957 11.430 .000
a. Dependent Variable: Exam Scores (%)

16