THE INFLUENCE OF WATCHING YOUTUBE PRONUNCIATION VIDEOS TOWARDS STUDENTS’ PRONUNCIATION ABILITY

In language teaching, media is one of the tools that very useful to support its successful teaching. In teaching pronunciation, the students need to give additional information how to pronounce words. One of the ways is by using Youtube pronunciation videos. By using Youtube pronunciation videos, the students will get more input and knowledge how to pronounce words from the native. The objective of this research is to know whether there is a significant influence of watching Youtube pronunciation videos towards students’ pronunciation ability. There were two variables in this research, the independent variable (X) was Youtube Pronunciation videos and dependent variable (Y) was the students’ pronunciation ability. The research method used was experimental design. The researchers dealt with two classes, they were an experimental class and a control class. In the experimental class, the researchers used Youtube pronunciation videos, whereas in control class the researchers did not use any media. Each class received the same pre-test and post-test. The population of this research was the second year students of MAN 1 Model Bandar Lampung. The samples of this research were two classes consisting of 90 students. In collecting the data, the researchers used reading aloud test as the instrument. After doing the data analysis, it was found that there is a significant influence of using videos towards students’ pronunciation ability. It can be seen from the result of the T test that T-observed was 5.34615, and the result of T-critical was 1.67. It means thatT-observed was higher than T-critical (T-observed> T-critical). Therefore, the hypothesis is accepted. This result means that there is significant influence of watching Youtube pronunciation videos towards students’ pronunciation ability.


INTRODUCTION
One of the English teaching purpose in Senior High School is to develop the students' competence in communication both oral and written form to reach the informational level. According to The Rule of The

Ministry of Education no 22 year 2006 about Standard Competence and
Basic Competence (SK/KD) of Senior and Islamic High School (2006) that the target of the English teaching in Senior High School is expected that the students can reach in the informational level, because they prepared to continue their study to the university. The communicative competence taught in learning English covers four skills; listening, speaking, reading, and writing. From those skills, speaking is the most important skill that has to be master, Richards (2008) states, "The mastery of speaking skills in English is a priority for many second-language or foreign-language learners" (p.19). It means the mastery of speaking skill is important for students, because oral communication requires the ability of speaking.
In speaking class, the students should be taught how to speak. The components of English speaking skill that should be given and studied in English speaking class are pronunciation, vocabulary, grammar, fluency, and comprehension. Speaking is the most important skill, because it is one of the abilities to carry out conversation on the language. It is an interactive process of constructing meaning, receiving, and processing information. But in English class, teaching speaking often ignore by the teacher. Harmer (2007) says that almost all English language teachers get students to study grammar and vocabulary, practice functional dialogues, take part in productive skill activities and try to become competent in listening and reading. From Harmer explanation, we know that teachers make little attempt to teach speaking, thus the students are not active in speaking activities.
According to Somantri (2003), most of students in Senior High School cannot speak English well. One of the problems is because of their pronunciation ability. Zamroni in Somantri (2003) states that our education problems are tend to be social stratification and education system school just transfer the dead knowledge. From this statement, it can be concluded that speaking is requires the pronunciation ability. Furthermore, the researchers conducted a preliminary research in MAN 1 Model Bandar Lampung to know the students' pronunciation ability. Based on the preliminary research, most of the students score in pronunciation is under the standard score. In this research, one of the internet media that will be used by the researchers is Youtube videos. By using Youtube pronunciation videos the students can learn how to pronounce English sounds correctly or minimally acceptable by learned from the native.

Method
In order to know whether there is significant influence of students' pronunciation ability by watching Youtube pronunciation videos, the researchers conducted an experimental research. It is a quantitative study that has two classes; one as the experimental class and the other one is the control class. The experimental class is a class that gave the treatment and the control class is a class that did not give the treatment. For the experimental class, the researchers used Youtube pronunciation videos in teaching pronunciation.
Meanwhile, the researchers did not use any media to the control class in teaching pronunciation.
The research design is presented as follows: Where: G1= experimental class There are two variables in this research; dependent and independent variables. Based on the problem, in this research the dependent variable is students' abilityin pronouncing verbs and the independent variable is watching Youtube pronunciation videos.

Subject
Population in a research is subject, it can be people or group of people that investigated by the researchers. Arikunto (2006)   took one piece of the rolled paper. The first paper chosen opened and the class written was the experimental class. Then the researchers took the second paper and the class written was the control class.

Data Collection Technique
To collect the data, the researchers used three kinds of techniques, as follows:

Pre-test
Pretest is used to know the students' pronunciation ability before the treatments. The test is orally with the students should read a text and the researchers record it. The recorded result of each student analyzed and scored to know the students' pronunciation ability in pronouncing English vowel sounds.

Post-test
Posttest is used to know the students' pronunciation ability after they were taught by watching Youtube pronunciation videos. The system and degree of difficulty of posttest same as the pretest, because both of them used to measure the students' pronunciation ability by watching Youtube pronunciation videos the development of students' pronunciation ability after Youtube pronunciation videos applied. Arikunto (2006) states that research instrument is a device used by the writer during the data collection. The instrument in this research was test.

Instrument
The researchers used oral test form and it was read a text. The test is used for pretest and posttest. Pretest has been given to check the students' ability in pronouncing vowel sounds before treatment. Posttest is use to gain the data of the students' ability in pronouncing vowel sounds after the treatment. The raters were the English teacher and the researchers.

Pre-test Instrument
For the pre-test instrument, the students asked to read a text and the researchers recorded it. The title of the text is The Man and The Serpent.

Post-test Instrument
For the post-test instrument, the students asked to read a text and the researchers record it. The title of the text is Golden Eggs.
After giving the test above, the recorded data scored. The scoring was used to measuring the students' vowel sounds.

Validity
According to Arikunto (2011), "Validity is a measurement which shows the level of validity or the real of the instrument" (p.168).
Consequently, a valid instrument has a high validity, meanwhile the instrument which is lack of lack of goodness has a low validity. An instrument can be called valid if it can show the data of variable are researched correctly. To measure the validity of the instrument, the researchers used content validity and construct validity.

Content Validity
Content validity is a type of evidence in which the content of test is judged to be representative of a larger domain of content. To get the content validity, the researchers consulted the instrument to the English teacher grade XI of MAN 1 Model Bandar Lampung, Mrs. Heny Astuti, S.Pd to make sure that the instrument is valid. After the researchers consulted to the English teacher, Mrs. Heny Astuti, S.Pd said that the instrument was valid.

Construct Validity
Construct validity focuses on the kind of test that is used to measure the ability. According to Setiyadi (2006), construct validity is used to measure perception, language behavior, motivation, even the language ability. In this session the researchers asked the English teacher to measure the construct validity of the instrument. After the researchers asked to Mrs. Heny Astuti, she said the instrument was valid.

Reliability
Reliability shows the degree of mainstays about something. It means the data can be believed so it can be relied on. According to Arikunto (2011), reliability shows that the instrument can be believed to be used as a tool of data collecting technique when the instrument is good enough. To know the reliability of test, the researchers used the following steps: 1. Giving pre-test and post-test items to 20 students out of sample.

Collecting the result and analyzing it.
3. Analyzing the difference between the pre-test and post-test result.
To get the reliability of the test, the researchers used inters rater reliability. It has done by two raters who examined the students' speaking test with the intention of knowing the reliability of the test. The raters were the researchers and Mrs. Heny Astuti, S.Pd. The statistical formula for counting the reliability was as follows: Acording to Sudijono (2009) states that

Normality Test
To measure the normality test, the researchers used Lilliefors test as follows: a. Arrange the sample's data from the lowest until the highest b. Determine the score Z from each data by using the following formula: 1. H o is accepted if Lobserved is lower than Lcritical , means the distribution of the data is normal.
2. Ha is rejected if Lobserved is higher than Lcritical , means the distribution of the data is not normal (Sudjana, 2005).

homogenity test
This test is intended to test whether the variance of the data in the experimental class and in the control class is equal or not. The formula is:  (Sudjana, 2005 1. Ho is accepted if Fobserved is lower than or equal to Fcritical means the variance of the data is homogenous.
2. Ha is rejected if Fobserved is higher than Fcritical means the variance of the data is heterogonous.

1) Hypothetical Test
To measure the hypothesis, the researchers used formula as follows : t-test =

Reliability
Reliability shows that the instrument can be believed to be used as a tool of data collecting technique when the instrument is good enough.

Reliability of Pre-test
From the calculation above, the result of pre-test is 0.62. it means that the data is very high reliability.

Reliability of Post-test
From the calculation above, the result of pre-test is 0.66. it means that the data is very high reliability.

Normality Test
Normality test is used to know whether the population is in normal distribution or not. The hypotheses for normality test formulated as follows:

Result of Homogenity Test
Homogenity test is used to know whether the variance of the data is homogenous or not.

Post-test
In the last activity, the students did the post-test. Post-test was held to know the students score after got the treatment.
The average of the students' result = The scores of the student

Hypothetical Test
Hypothetical test is used to know whether there is significant influence of watching Youtube pronunciation videos towards students pronunciation ability towards students' pronunciation ability or not. Based on t -test result calculation (see appendix 14) obtained t -test = 5.35. t _Critical at degree of significance 5% and df = N X + N Y -2 = 45+45-2=88 gain t -test = 1.66.
Since 5.35 > 1.66, H o is refused and H a is accepted. It means that the treatment has influence of the students' score. It means that there is significant influence of watching Youtube pronunciation videos towards students' pronunciation ability of MAN 1 Model Bandar Lampung.

Discussion
According to the result of the students' post-test score, the average of students' post-test score in experimental class was 64,333. The average of students' post-test score in control class was 61,778. It shows that the students' post-test score in experimental class was higher than students' posttest score in control class.
Moreover, from the calculation of students' pre-test and post-test score using t-test, the result of t _observed was 5.35 when the result of t_ critical was 1.66. It means thatt_ observed was higher than t_ critical (Tobserved > Tcritical ). Therefore, the Ha hypothesis is accepted.
Because the Ha hypothesis was accepted, thus the researcher concludes, it could be said that there is a significant influence of watching Youtube pronunciation videos towards students' pronunciation ability at grade eleventh of MAN 1 Model Bandar Lampung at the first semester in 2017/2018 academic year.

CONCLUSION
Based on the data analysis, it can be concluded that there is significant influence of watching Youtube pronunciation videos towards students' pronunciation ability. It can be seen from the result of t test that t_ observe is 5.34615, and the t_critical is 1.66. The criteria to know whether the hypothesis is accepted or not, dealt t_ observe is higher than t_ critical (t_ observe t_ critical ). From this statement, it is clear that 5.34616 is higher than 1.66. Therefore, Ho is rejected and Ha is accepted. This result means that there is significant influence of watching Youtube pronunciation videos towards students' pronunciation ability.