FOREIGN TRADE UNIVERSITY
Faculty of Economics and International Business
---------------------------

DISSERTATION
THE EFFECT OF USING SMARTPHONE ON THE RESULT OF THE
STUDENT IN FOREIGN TRADE UNIVERSITY BRANCH II (ANOVA)


List of members
No

Ho Chi Minh City – April, 2014

Full Name

Student ID

1

Phạm Thị Hạnh

1201017091

2

Nguyễn Ngọc Huy

1201017131

3

2. Theory and research methodology

2

2.1. Theoretical basis and Analysis framework

2

2.2 Methods of data collection and model estimation technique

6

3. Results and discussion

7

3.1 Descriptive data

7

3.2 One-way ANOVA model

9

3.3 HSD (honest significant difference) test

11

4. Conclusion and Policy Implication


University 2, students, one-way ANOVA, Tukey's HSD test

Smartphone Effective

Page 4


1. Introduction
Nowadays, along with the evolution of Information Technology, telephones are
not only for texting and calling purposes, but they also help us to connect with each
other via social networks, email and other online services... These smartphones are
becoming more modern and helpful day after day. However, overusing smartphone
may cause many negative effects on everyone especially college students. These
effects include decline in health, waste of time and decrease in study result... The
decrease in study result is the most serious consequence when smartphones are getting
commoner among the students.
The phenomenon that smartphones are addictive and affect many respects of
life is no new problem. It has appeared so many times on the media. This is an
unsolvable problem for the students as well as a deep concern for the parents.
Therefore we decide to carry on the topic “The effect of smartphone on the study
result of the student in Foreign Trade University branch II” by analyzing One - factor
ANOVA. We find down how the percentage, scale and usage of smartphone of the
second-year student of Foreign Trade University branch II in HCMC change their
study result as well as propose some solution to overcome this worrying problem.
2. Theory and research methodology
2.1. Theoretical basis and Analysis framework
While the analysis of variance succeeded in the 20th century, antecedents
extend centuries into the past according to Stigler. These include hypothesis testing,
the partitioning of sums of squares, experimental techniques and the additive model.
Laplace was performing hypothesis testing in the 1770s. The development of leastsquares methods by Laplace and Gauss circa 1800 provided an improved method of


x
11
x
12
…

x

…

x

…

x

x
1n1

x

21

x

22
…
2n2


Step 3: Find variances
MSW (mean square within): MSW=
MSB (mean square between): MSB=
Step 4:One-way ANOVA Table
Source of SS
Df
MS
F ratio
variation
(sum
of (degrees of (mean of
square)
freedom)
square)
Between
SSB
k-1
MSB=
F=
Samples
Within
SSW
n-k
MSW=
Samples
Total
SST
n-1

In: k number of populations

individual surveyed has their own features and cannot represent for the whole set.
We had to choose the one factor ANOVA to analyze. Compare the average of
many populations based on the average of models. Consider the effect of one factor
reason to result factor.

3. Results and discussion
3.1 Descriptive data
Below is the data collected from 238 people, questioned about how many hours
the sophomore of FTU II use smartphone and their study result / average mark? Base
on their using hour, we divided them into 4 groups (as shown in the table). The unit of
measure is hour.
Groups of factor
from 0h to 2h

>2h to 4h

>4h to 6h

>6h

8,88

7,20

7,90

7,11

Smartphone Effective



6,00

8,79

7,80

7,60

5,70

7,20

7,00

8,07

6,70

8,79

8,40

7,80

6,80

5,60

8,90


8,70

8,70

7,00

6,60

9,20

8,40

8,20

7,20

8,90

8,30

7,00

7,30

8,00

8,00

7,90


6,20

8,00

8,00

7,80

8,20

6,80

8,50

7,40

8,40

7,00

8,90

6,70

8,50

7,20

Smartphone Effective

7,90

6,60

8,00

9,00

7,00

7,70

5,80

7,80

9,03

8,80

7,00

7,60

8,00

8,60

7,12


7,12

7,90

9,00

7,40

6,19

8,00

6,90

7,40

6,95

7,80

7,70

7,90

6,95

8,80

8,00


8,60

7,32

5,00

8,00

7,40

7,32

6,45

7,80

8,30

7,67

6,45

8,30

8,29

7,55

5,54



7,00

6,95

7,80

7,60

7,46

7,85

7,00

8,23

8,20

7,67

4,00

6,50

7,60

6,00

7,90

8,89
8,90
8,45
7,40
7,60
7,90
7,30

Smartphone Effective

Page 12


9,12
8,80
9,27
8,34
8,50
6,50
8,00

3.2 One-way ANOVA model
Hypothesis:
Ho: There is no different in the average monthly food cost between 4 group. (µ1 =
µ2 = µ3 = µ4)
H1: The average monthly food cost of them are not equal.
Table 1:Anova: Single Factor by Excel
SUMMARY
Groups



57

421,66 7,39754386

0,569683145

>6h

53

356,52 6,72679245

0,961922206

Smartphone Effective

Page 13


3

ANOVA

Source
Variation

of
SS



136,1692091

234

Total

199,327758

237

7

If F > F crit, we reject the null hypothesis. As we can see in the table above:
36.178273 > 2.6432. Therefore, we reject the null hypothesis. The means of the four
populations are not all equal. At least one of the means is different. Therefore, we can
say that the hour for using smart phone does affect how much to the result / average
mark of students.
3.3 HSD (honest significant difference) test
Because the null hypothesis has been rejected, the result / average mark of four
groups are not equal. However, in order to find out how they differ from each other,
we need to do Tukey’s HSD (honest significant difference) test and compare each
couple of group.
t-Test: Two-Sample Assuming Equal Variances
Smartphone Effective

Page 14


From 0h to 2h and >2h to 4h

0,430531732

Hypothesized

Mean

Difference

0

Df

126

t Stat

1,048187211

P(T
0,485674861

5

Observations

74

57

Pooled Variance

0,522143574

Hypothesized

Mean

Difference

0

Df

129

t Stat

5,112973725


from 0h to 2h

>6h
6,72679245

Mean

8,048648649

3
0,96192220

Variance

0,485674861

6

Observations

74

53

Pooled Variance

0,683793757

Hypothesized


Smartphone Effective

Page 17


If t Stat < -t Critical two-tail or t Stat > t Critical two-tail, we reject the null
hypothesis. In this table: 8.883284671 >1.979124109. Therefore, µ1 ≠ µ4 , the result /
average mark of the two groups are different.
>2h to 4h and >4h to 6h
Hypothesis:
Ho: µ2 - µ3 = 0
Table 5: Tukey’s HSD test (>2h to 4h and >4h to 6h) by Excel

Mean

>2h to 4h

>4h to 6h

7,925555556

7,39754386
0,56968314

Variance

0,354579874

5


t Critical one-tail

1,658953458

P(T1.98196749. Therefore, µ2 ≠ µ3 , the result /
average mark of the two groups are different.
>2h to 4h and >6h
Hypothesis:
Ho: µ2 - µ4 = 0
Table 6: Tukey’s HSD test (>2h to 4h and >6h) by Excel

>2h to 4h

>6h
6,72679245



Df

105

t Stat

7,65837593

P(T
Pooled Variance

0,758538989

Hypothesized

Mean

Difference

0

Df

108

t Stat

4,03600465

P(T
It will be better for students’ health as well as study results if they do not use
smartphone after 9 pm. Smartphones only brings much benefit and convenience when
students know how to use them reasonably.

Smartphone Effective

Page 21


REFERENCE:
1.

David F.G., Patrick W.S., Phillip C.F. and Kent D.S. (2010), Business Statistics
8th edition, Pearson.

2.

Fisher .Ronald Aylmer ,sir (1890-1962) -The analysis of variance with various
binomial transformations.

3.

Fisher, Ronald Aylmer, Sir, 1890-1962-Answer to query 114 on the effect of
errors of grouping in an analysis of variance

4.

Hoang Tran Van and Van Le Hong (2013), Principles of Statistics, Vietnam
National University-Ho Chi Minh City Press, Ho Chi Minh City.