CAPTER IV

RESULT AND DISCUSSION

A. Data descriptions
In this chapter, the writer explains the result of the research. The writer

will attempt to submit the data as outcomes of research has hold in Five grade at

SDN IV Cilegon. The writer takes 76 students as a subject this research. It is

divided into two classes. They are 38 students from V A as the Experimental class

and 38 students from V B as the control class. The data of this research were the

score of the students’ pre-test and post-test both experimental class and control

class. The score of pre-tests was taken before the treatment, while the score of

post-tests was taken after the treatment. The result of pre-test is to know students’

pronunciation before receiving the treatment, meanwhile the result of post-test is

to give the information whether there is any improvement on students’

pronunciation achievement after receiving the treatments.

In this research, the writer gave treatments to experimental class and

control class in teaching pronunciation. In the experimental class, the writer

applied Duolingo Application to teach pronunciation, while in control class the

writer applied conventional technique. The writer measured students’

pronunciation achievement by using a test essay forms. Below are the data of pre-

test and post-test in experimental and control class.

1. The Data of the Experiment Class

Table 4.1
The Result of Pre and Post Test in Experiment Class

Experiment Class (X)

No Students Pre-Test Post-Test Difference
1 ADR 70 75 5
2 AIA 40 60 20
3 AFF 50 70 20
4 AFR 30 60 30
5 AK 50 70 20
6 AFS 30 50 10
7 AN 50 60 10
8 ACD 40 55 15
9 ARM 60 75 15
10 AR 50 60 10
11 AF 55 75 20
12 BRA 60 70 10
13 CP 80 85 5
14 FA 45 60 15
15 FPC 40 65 20
16 HNW 30 50 20
17 HYS 55 75 20
18 JRA 50 60 10
19 KA 30 40 10
20 KN 35 50 15
21 KAI 40 60 20
22 MRA 60 70 20
23 MBLM 50 55 5
24 MIA 80 90 10
25 MAI 50 60 10
26 MD 80 95 15
27 MRN 30 90 60
28 MN 40 65 15
29 NNY 45 60 15
30 N 55 75 20
31 NSN 70 75 25
 32 NNA 45 55 5
33 RRSA 50 60 10
34 RZ 55 70 15
35 RRS 80 90 10
36 RAP 70 75 5
37 VES 60 70 10
38 ZAP 55 65 10
TOTAL 1965 2545 580
MEAN 51,7 66,9 15,2

According to the results of the experiment class's pre-test and post-test, the

lowest pre-test score was 30 and the best score was 80. In the meantime, the

lowest post-test score was 40, and the highest score was 90. The pre-test

mean was 51,7 and the post-test mean was 66,9. As a result, the difference in

mean from pre-test to post-test for the writer was 15,2. Based on the

discrepancies between the students' pre-test and post-test scores, it can be

inferred that employing the Duolingo application had a good effect on

improving students' English pronunciation.

2. The Data of the Control Class

Table 4.2
The Result of Pre and Post Test in Control Class

Experiment Class (X)

No Students Pre-Test Post-Test Difference
1 AFD 50 60 10
2 AG 40 50 10
3 ACS 45 60 15
4 AJS 30 35 15
5 CAN 50 55 5
6 CML 60 70 20
7 DRI 40 50 10
8 FF 40 55 15
9 FA 80 85 5
 10 GS 30 35 5
11 HS 40 50 10
12 HAA 50 60 20
13 HPN 60 65 5
14 JC 50 55 5
15 MPC 55 65 10
16 MHR 30 35 5
17 MKA 30 40 10
18 MMM 20 25 5
19 MNR 40 45 5
20 MNA 30 40 10
21 MRAM 50 65 15
22 MRP 60 65 5
23 MSP 55 60 5
24 NAM 70 80 10
25 NAR 30 40 10
26 NRH 35 40 5
27 NP 40 50 10
28 PN 35 40 5
29 RBS 40 50 10
30 RFC 30 40 10
31 RPP 55 60 5
32 RPS 50 55 5
33 RAT 60 65 5
34 SSG 70 75 5
35 SFN 80 85 5
36 SR 60 65 5
37 TA 45 55 10
38 WP 50 60 10
TOTAL 1785 2085 330
MEAN 46,9 54,8 8,6

According to table 4.2, the lowest pre-test score was 20 and the highest

pre-test score was 80, with an average pre-test score of 46,9. Aside from that,

the average post-test score increased to 54,8. The highest post-test score was
 85, while the lowest was 25. As a result, the conclusion is that the Duolingo

application as teaching learning media can improve students' English

pronunciation, as seen by the table, which shows that the experiment class

had higher differences scores than the control class.

6% 4%
21%
Pre-Test Experimental
22% Post-Test Experimental
pre-Test Control
Post-Test Control
Difference Experimental
difference Control
27%
19%

This paragraph describes the results of a study comparing the effectiveness of

an experimental intervention in a treatment group compared to a control group.

The pre-test results showed that the experimental group had a score of 21% and

the control group had a score of 19%. After the intervention, the experimental

group had a post-test score of 27%, while the control group had a score of 22%.

The difference in means between the pre-test and post-test scores in the

experimental group was 6%, indicating an improvement in performance, while the

difference in the control group was 4%. These results suggest that the Duolingo
Application had a greater impact on improving students pronunciations in the

experimental group compared to the control group.

Table 4.3
The Comparison between the Experiment Class and Control Class

NO Experiment Control X Y X2 Y2
Class (X) Class
(X- (Y-
(Y)
MX) MY)

1 5 10 -10,2 1,4 104,04 1,96

2 20 10 4,8 1,4 23,04 1,96

3 20 15 4,8 6,4 23,04 40,96

4 30 15 14,8 6,4 219,04 40,96

5 20 5 4,8 -3,6 23,04 12,96

6 10 20 -5,2 11,4 27,04 129,96

7 10 10 -5,2 1,4 27,04 1,96

8 15 15 -0,2 6,4 0,04 40,96

9 15 5 -0,2 -3,6 0,04 12,96

10 10 5 -5,2 -3,6 27,04 12,96

11 20 10 4,8 1,4 23,04 1,96

12 10 20 -5,2 11,4 27,04 129,96

13 5 5 -10,2 -3,6 104,04 12,96

14 15 5 -0,2 -3,6 0,04 12,96

15 20 10 4,8 1,4 23,04 1,96

16 20 5 4,8 -3,6 23,04 12,96

17 20 10 4,8 1,4 23,04 1,96

18 10 5 -5,2 -3,6 27,04 12,96

19 10 5 -5,2 -3,6 27,04 12,96

20 15 10 -0,2 1,4 0,04 1,96

21 20 15 4,8 6,4 23,04 40,96

22 20 5 4,8 -3,6 23,04 12,96

23 5 5 -10,2 -3,6 104,04 12,96

24 10 10 -5,2 1,4 27,04 1,96

25 10 10 -5,2 1,4 27,04 1,96

26 15 5 -0,2 -3,6 0,04 12,96

27 60 10 44,8 1,4 2007,04 1,96

28 15 5 -0,2 -3,6 0,04 12,96

29 15 10 -0,2 1,4 0,04 1,96

30 20 10 4,8 1,4 23,04 1,96

31 25 5 9,8 -3,6 96,04 12,96

32 5 5 -10,2 -3,6 104,04 12,96

33 10 5 -5,2 -3,6 27,04 12,96

34 15 5 -0,2 -3,6 0,04 12,96

35 10 5 -5,2 -3,6 27,04 12,96

36 5 5 -10,2 -3,6 104,04 12,96

37 10 10 -5,2 1,4 27,04 1,96

 38 10 10 -5,2 1,4 27,04 1,96

Tota 3397,52 684,48

580 330
l

Explanation:

X = Difference score of the experiment class

Y = Difference score of the control class

x = The different score of X after compared with the mean of difference

score of the experiment class

y = The different score of X after compared with the mean of difference

score of the control class

x2 = The square score compared to the mean of difference score of

experiment class

y2 = The square score compared to the mean of difference score of control

class

MX= The mean of difference score of the experiment class

YX = The mean of difference score of the control class

ƩX = Total of difference score of the experiment class

ƩY = Total of difference score of the control class

Ʃx2 = Total of The square score compared to the mean of difference score

of experiment class

Ʃy2 = The square score compared to the mean of difference score of

control class
 According to the data gathered from the results, the square score of the

experiment class compared to the mean of difference score was 580 and the

square score of the control class compared to the mean of difference score

was 330. As a result, the writer concluded that using the Duolingo application

is more effective to improve pupils' English pronunciation.

B. Data Analysis

The formula of t-test as follow:

∑ M 1−M ²
TO¿ 1
SE M −M ²

The calculation can be seen as follow:

1. Determining mean of variable X

X 580
M1 = ∑ = ∑ = 15,26
N1 38

2. Determining mean of variable Y

Y 330
M2 = ∑ = ∑ = 8,68
N2 38

3. Determining standard of deviation score of variable X

SD1 =
√ ∑X²
N1
=
√
3397,52
38
= √ 89,40 = 9,4

4. Determining standard of deviation score of variable Y:

 SD2 =
√ ∑y²
N1
=
38 √
684,48
= √ 18,01 = 4,2

5. Determining Standard Error Mean Variable X:

SD ¹ 9,4 9,4
= = = = 1,5
SEm1
√N 1
−1 √38−1 √37

6. Determining Standard Error Mean Variable Y:

SD ² 94,2 9,4
= = = = 0,6
SEm2
√ N −1
²
√38−1 √37

7. Determining Standard Error of Different Mean Variable X And

Variable Y, with formula:

SE m1-m2 = √ SEm 1−SEm 2 = √ 1,5−0,6 = √ 0,7 = 0,8

8. Determining t0 with formula:

1
M −M ² 15,26−8,68
To = = = 7,4
SE m 1−m2 0,8

9. Determining Degrees of Freedom, with formula:

𝐷f = (𝑁𝑁1+𝑁𝑁2) − 2

𝐷f = (38+38) – 2

𝐷f = 74

10. Determining t-table in significance level 5% and 1% with df.

 Df = 74

Significance level 5% of df 74 is 1,66

Significance level 1% of df 74 is 2,37

11. Comparing the t-score with t-table:

The researcher compares tO (tobservation) with tt (t-table) that

if tO > tt it means that H0 (Null hypothesis) is rejected and Ha

(alternative hypothesis) is accepted, but when t 0 < tt it means that H0

is accepted, and Ha is rejected.

t0: tt = 7,4 > 1,66

t0: tt = 7,4 > 2,3

C. Discussion and Interpretation of the Data

The mean of the pre-test 51,7 before employing Duolingo Application is

described in the data obtained from 38 students in the experimental class. The

mean of the post-test was 66,9 after the writer used the to improve pupils' English

pronunciation four times for the experimental class. As a result, the writer

received a gain score of 15,2 on average. In the pre-test, the lowest score was 30

and the highest score was 80. The statistics revealed that the lowest score was 40

and the best score was 95 in the post-test. In summary, the lowest and highest

scores in the post-test were higher than in the pre-test.

 The writer then calculated the mean of the pre-test 46,9 from the description

of score in the controlled class. The writer did not teach the students Duolingo

application in this lesson, but rather a theory of English pronunciation. The writer

received a mean of 54,1 after giving four sessions without employing Duolingo

application. The average gain score for the writer was 8,6. In the pre-test, the

lowest score was 20 and the highest score was 80. The statistics revealed that the

lowest score was 25 and the best score was 85 in the post-test. In summary, the

lowest and highest scores in the post-test were both higher than in the pre-test.

The final calculation was testing the hypothesis. This was the main calculation

to answer the problem formulation of this research that whether there is

significant different between students’ English pronunciation at controlled class

without using Duolingo application and students’ English pronunciation at

experiment class which using Duolingo application. The writer used t-test formula

in the significance degree. The value of the to was 7,4 with the degrees of freedom

(df) was 74. In this research, the writer used the degree of significance in the level

of 5% and 1% which in the table showed that the level 5% of df 74 was 1,68 and

1% of df 74 was 2,3. So, the result was 1,68 < 2,4 < 7,4. It means that the

alternative hypothesis (Ha) is accepted, and the Null Hypothesis (Ho) is rejected.

In addition, there is a significance difference between students’ English

pronunciation by using Duolingo application and without Duolingo application.

From the data above the students in experiment class and control class. The

experiment class had higher score in the aspect of pronunciation, but this was

happened after the writer gave them the treatment. Before it, they very lack about
that. Because of control class was not given treatment, almost of pronunciation

aspects they lacked. The writer awarded also that pronunciation is important in

English skills. there were the advantages in using Duolingo application, it

increased the interaction between the students and teacher