Putri Anggraeni, et al / Journal of English Language Teaching 6 (1) (2017) 63 ELT FORUM 7 (2) (2018) Journal of English Language Teaching http://journal.unnes.ac.id/sju/index.php/elt THE INFLUENCES OF TONGUE TWISTER IN TEACHING PRONUNCIATION OF ASPIRATED SOUND [ph] Anik Kholisoh, Alief Noor Farida  English Department, Faculty of Languages and Arts, Universitas Negeri Semarang, Indonesia Article Info ________________ Article History: Received in October 2018 Approved in November 2018 Published in January 2019 ________________ Keywords: aspirated sound; pronunciation; tongue twister. ____________________ Abstract ___________________________________________________________________ Speaking English for Indonesian students needs more than an effort to master as they cannot produce intelligibly pronunciation. Besides, their mother tongue interferes them when there is different sound production between Indonesian and English; hence, they have difficulty in pronouncing aspirated sound that does not exist in Indonesian. This study aims to find out how Indonesian students pronounce aspirated sound [ph] and the influences of tongue twister to teach pronunciation of aspirated sound [ph]. This research uses pre-test post-test quasi-experimental design. The data were collected from the result of pronunciation tests of seventh graders of SMP Negeri 12 Magelang in academic year 2016/2017. The findings showed that students mispronounced [ph] into /p/ as what they had in Indonesian. After being taught with tongue twister, the mean score of experiment group increased from 13.28 to 50.16. There was significant difference of the ability between students who were taught pronunciation using tongue twister and those who were not. The t-test result was 0.000 sig 5%; it means there was different ability between control and experiment group after receiving the treatment. Based on the result, the researcher concluded that tongue twister helps Indonesian students to improve their pronunciation of aspirated sound [ph]. © 2018 Universitas Negeri Semarang  Correspondent Address: ISSN 2252-6706 B3 Building FBS Unnes Sekaran, Gunungpati, Semarang, 50229 E-mail: anikkholisoh.10595@gmail.com A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 64 INTRODUCTION The fact that English is considered as inconsistence language is very crucial since its spelling is different from its pronunciation. For English native speaker, it would not be a big problem, but for English as the Second Language (ESL) and English as the Foreign Language (EFL) learners, it may take a long time and need more than an effort to master the language. Indonesian students are one of the ones who find it difficult to learn English as Indonesia is included in the expanding circle of Braj Kachru’s circle together with China, Egypt, Israel, Nepal, Saudi Arabia, Zimbabwe and some other countries (Yoneoka, 2002). Thus, the problem arises in speaking field as the learner cannot produce the intelligibly pronunciation or pronunciation that is good enough to be always understood (Harmer, 2001). Sometimes ESL and EFL learners who are good at vocabulary and grammar have some difficulties in pronunciation because they do not learn pronunciation from the early time as they start learning English. In addition, Brown (2001) stated that for general speaking, children have an excellent opportunity to sound like native speaker if they keep being in authentic context. Consequently, learners in the beginning level need to have more practice if they want to have good ability in pronunciation. Unfortunately, Indonesian learners do not have sufficient training in this language aspect when they study English at school. As the compulsory subject, each English aspects does not have balance portion in teaching learning process. Harmer (2001) revealed that “almost all English teachers get students to study grammar and vocabulary, practice functional dialogues, take part in productive skill activities, and become competent in listening and reading. Yet some these same teachers make little attempt to teach pronunciation in any overt way and only give attention to it in passing.” In line with this statement that, in fact in Indonesia, pronunciation practice do not have enough portion to be taught in the class because teachers tend to focus on the main lesson material such as reading text. Students has not been introduced to any segmental feature such as English sounds nor supra-segmental features such as stress and intonation. Another thing to be concerned about learning new language for ESL and EFL learners is the influence of the native language or first language (L1) to the target language or second language (L2). If some sounds are very similar in both native and target language, the influence is called facilitating. For example, both Indonesian and English have sound /n/; for Indonesian students, the existence of sound /n/ helps them in acquiring sound /n/ in English because they are somewhat similar. However, the interfering influence comes when Indonesian students face English sound /ʒ/ which does not exist in their native language, so does happen for sound [ph] From all those discussions, the main problem for Indonesian students is in learning pronunciation practice. It is in line with Wulandari (2011) who stated that students meet difficulties in learning English pronunciation due to the different elements between their mother language and the target language. They do not have enough basic knowledge in pronunciation to go one step ahead to speaking so that many Indonesian people consider speaking English is very difficult to do and to master. Besides, mother tongue interfering problem comes to learners who have learned English but finding hard to be fluent in speaking and hard to catch native speaker spoken due to the different sound production between Indonesia and English. Thus, learning English pronunciation in the very early level can help to overcome this problem because learners stand a good chance to sound like native speaker. Moreover, the way to teach pronunciation needs to be effective so that the learning process gets maximum result. One of the effective method to teach pronunciation is using tongue twister that is fun and challenging as being suggested by Prošic-Santovac (2009) that “tongue twisters were A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 65 generally perceived as a useful exercise for improving pronunciation, and their potential for providing context for the target phonemes was valued the most.” This study aims to find out how Indonesian students pronounce aspirated sound [ph] before this research was conducted. Further, researcher also wants to know whether tongue twister gives influences for Indonesian students when teacher uses it to teach pronunciation of aspirated sound [p h ]. METHODS This research used quantitative research in which the data analysis could be done by statistical calculation. There are some types of quantitative research, but in this study quasi-experimental design was chosen. For the reason of not making a mess on the class which had been set by the school, the pre-test post-test quasi-experimental design was quite suitable for this study. The population of this study was the seventh grade students of SMP Negeri 12 Magelang in the academic year of 2016/2017. The seventh graders were chosen because the study about pronunciation practice was very effective given to English learners in beginner level to help them overcame the fossilized accent. Furthermore, cluster sampling technique was selected because the design of this research was quasi- experimental which inquired nonrandom control and experiment group. It was simpler if the sample was taken as group or cluster than as individual subject. In this study, VII A consisted of 31 students was chosen as the control group, and VII D consisted of 32 students was chosen as the experiment group. Paper [ˈpeɪ.pə r ] Pencil [ˈpen t  .s ə l] Parking [ˈpɑː.kɪŋ] Pepper [ˈpep.ə r ] Picture [ˈpɪk.tʃə r ] Party [ˈpɑː.ti] Partner [ˈpɑːt.nə r ] Parent [ˈpeə.rənt] Pretty [ˈprɪt.i] Peter [ˈpiː.tə] Paris [ ˈpær.ɪs] Proper [ˈprɒp.ə r ] Apology [əˈpɒl.ə.dʒi] Repeat [rɪˈpiːt] Peanut [ˈpiː.nʌt] People [ˈpiː.pl  ] Post office [ˈpəʊst ɒf.ɪs] Personal [ˈpɜː.s ə n. ə l] Practice [ˈpræk.tɪs] Power [paʊə r ] Table 2.1. List of Words Consisting of Aspirated Sound [ph] Using test as the instrument of collecting data, the researcher conducted two kinds of tests, they were pre-test and post-test. However, before the instrument was tested, researcher tried it out to know whether the test was eligible and suitable to be used using validity and reliability test. The try A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 66 out test was conducted once in VII C consisting of 31 students. The researcher handed in student worksheet in order to give chance for students to study at a glance first before the test was delivered. There were twenty words consisting of aspirated sound [ph] in the worksheet that is presented in Table 1. After giving short explanation about the objective of the test and the material of the test, students came forward one by one to read twenty words consisting aspirated sound [ph]. The result of the study was score which represented students’ ability in pronouncing aspirated sound [ph]. After the data was collected, the researcher analyzed them using statistical analysis. The statistical analysis in this research was the analysis of pre-test and post-test result which included descriptive statistic, normality, homogeneity, and t-test calculation using SPSS 20.0 to gain the hypothesis. The hypotheses of this research were: H0 : there is no influence of tongue twister in students’ ability in pronouncing aspirated sound [ph] Ha : there is influence of tongue twister in students’ ability in pronouncing aspirated sound [ph] The researcher used the probability level to determine the result of hypothesis. If the probability > 0,05 it means H0 is accepted and Ha is refused. If the probability < 0,05 it means H0 is refused and Ha is accepted. RESULTS AND DISCUSSIONS Based on the researcher’s observation and experience during teaching practice for three months in SMP Negeri 12 Magelang, students’ ability in pronunciation was in general low especially in pronunciation of aspirated sound [ph]. Their sound production on [ph] sound was influenced by their mother language. They mispronounced [ph] into /p/ as what they had in Indonesian. When they started learning English, they were difficult to acquire aspirated sound. It was proved by the low score of pre-test that was delivered before students got any treatment. 3.1 Pre-test Pre-test was test given before conducting the treatment. Both control and experiment group received similar test. This test was used to measure the initial ability in both groups. It was also used as a guide for the researcher to know that students’ ability in pronunciation in control and experiment group was in the same level. Using the reliable and valid test, researcher tested students one by one pronouncing words consisting aspirated sound [ph] and gave score. Score 1 was given for the correct sound production and score 0 was for the incorrect one. The final score was drawn from the following formula. 𝑆𝑐𝑜𝑟𝑒 × 5 = 𝐹𝑖𝑛𝑎𝑙 𝑆𝑐𝑜𝑟𝑒 The result of the test was analyzed in statistical analysis using SPSS 20.0. The mean of pre-test in control group was 13.39 with the highest score was 30 and the standard deviation was 8.697. Whereas the mean of the experiment group was 13.28 with the highest score was 30 and standard deviation was 8.855. Kolmogorov-Smirnov df Sig. Pre-test control group 31 0.186 Pre-test experiment group 32 0.116 Table 3.1. Pre-Test Normality Test A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 67 The result of pre-test statistical analysis was then analyzed in normality and homogeneity test. The normality test in this study was conducted by Kolmogorov-Smirnov in SPSS 20.0. The distribution is normal if the significance value is > 0.05. Table 2 showed that the sig value of pre-test control group (0.186) was higher than 0.05 and the sig value of pre-test experiment group (0.116) was also higher than 0.05. In short, the distribution of the test score in pre-test both in control and experiment group was normal. Figure 1 and 2 present the normal distribution of pre-test of control and experiment group. Figure 3.1. Score Distribution of Pre-test of Control Group Figure 3.2. Score Distribution of Pre-test of Experiment Group Levene Statistic df1 df2 Sig. 0.000 1 61 0.996 Table 3.2. Pre-Test Homogeneity of Variances Test The homogeneity test was conducted using the formula of One Way ANOVA. From the calculation, the probability/sig was 0.996 which meant that it was higher than 0.05. Therefore, this test was homogeneous or the pre-test of control and experiment group came from the same variance. A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 68 Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2- tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper nilai Equal variances assumed .000 .996 .048 61 .962 .106 2.212 -4.317 4.529 Equal variances not assumed .048 60.987 .962 .106 2.211 -4.316 4.528 Table 3.3. Pre-Test Independent Samples Test In this study, the researcher used Independent Sample T Test. The result showed that the sig (2-tailed) which was higher than 0.05. It meant that there was no difference initial ability between students in the control and experiment group. From all pre-test calculations, it can be concluded that in the initial condition, both control and experiment group were in the same level. The mean scores of the two were merely similar. If we look at the histograms in Figure 1 and 2, they performed less skewness or the shape was almost close to bell shape which showed the normal distribution of mean scores (Tuckman in Saleh, n.d.). Furthermore, the spread or variance of the two groups was equal or homogeneous proved by the value of homogeneity of variance was 0.996 > 0.05. At last, the t-test score that was 0.962 > 0.05 means H0 is accepted and Ha is refused or there is no different ability between control and experiment group before the treatment was given. 3.2 Treatment The experiment group was taught pronunciation of aspirated sound [ph] using tongue twister. The researcher provided videos as the media in delivering tongue twister, modeled how to pronounce and guided students to practice. On the other side, control group was taught using conventional method (unstructured discussion). However, this group were still given brief explanation about the material of pronunciation of aspirated sound [ph]. Below is the list of tongue twister used in the treatment. Pirates private property Pirates private property Pirates private property Peter Piper picked a peck of pickled peppers. A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 69 A peck of pickled peppers Peter Piper picked. If Peter Piper picked a peck of pickled peppers, Where's the peck of pickled peppers Peter Piper picked? Based on researcher observation during the treatment, students quite enjoyed practicing tongue twister since it is fun and challenging. Automatically they trained their tongue to produce aspirated sound while pronouncing tongue twister. 3.1 Post-test The post-test for experiment group was done after this group was given special treatment which was practicing tongue twister. Students did the same test as they did in pre-test; they pronounce twenty words one by one. The researcher’s observation focused only on students’ production of the aspirated sound [ph]. Score 1 was given for the correct sound production and score 0 was for the incorrect one. Control group had exactly the same post-test, technique and scoring system as experiment group. The mean of post-test in control group was 29.03 with the highest score was 85 and the standard deviation was 17.341. Whereas the mean of post-test in experiment group was 50.16 with the highest score was 100 and standard deviation was 20.260. Kolmogorov-Smirnov df Sig. Post-test control group 31 0.129 Post-test experiment group 32 0.200 Table 3.4. Post-Test Normality Test The result of pre-test statistical analysis was later analyzed in normality and homogeneity test. The distribution is normal if the significance is > 0.05. Table 5 showed that the sig value in post-test control group (0.129) was higher than 0.05 and the sig value in post-test experiment group (0.200) was also higher than 0.05. Consequently, the distribution on the test score in pre-test both in control and experiment group was normal. Figure 3 and 4 present the normal distribution of post-test of control and experiment group. Figure 3.3 Score Distribution of Post-test of Control Group A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 70 Figure 3.4 Score Distribution of Post-test of Experiment Group Levene Statistic df1 df2 Sig. 0.956 1 61 0.332 Table 3.5. Post-Test Homogeneity of Variance Test Table 5 presents the result of post-test homogeneity of variance test. The result of homogeneity test showed that the probability/sig was 0.332 which meant it was higher than 0.05. Therefore, this test was homogeneous or the post-test of control and experiment group came from the same variance. Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2- tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper nilai Equal variances assumed .956 .332 -4.440 61 .000 -21.124 4.758 -30.638 -11.610 Equal variances not assumed -4.451 60.099 .000 -21.124 4.746 -30.618 -11.630 Table 3.6. Post-Test Independent Samples Test Table 6 presented that sig (2-tailed) or probability was 0,000 while the significance level was 0,05 (5%). It showed that the probability was less than the significance level (0,000 < 0,05). From all post-test calculations, it can be concluded that after receiving treatment, the mean score increased quite significant for both control group (13.39 to 29.03) and experiment group (13.28 to 50.16). Moreover, the post-test score was also in normal distribution. The histogram of post-test control on Figure 3 performed a little bit skewness, but the normality value was still in normal category (0.129 > 0.05) while the distribution of post-test experiment score on Figure 4 was almost symmetrical which meant normal distribution. Both groups score spread or variance was also homogenous or equal proved by the value of post-test homogeneity of variance was 0.332 > 0.05. Finally, the final condition of the two groups was shown by the value of t-test 0.000 < 0.05. The probability is less than 0.05 then H0 is refused and Ha is accepted or in other word there is A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 71 different ability between control and experiment group after receiving the treatment. From the result, the researcher concludes that teaching pronunciation of aspirated sound using tongue twister influences students’ ability shown by the score after the treatment was higher than the score before the treatment and the significance difference in t-test of post-test. From the result of the research, it can be drawn a theory that English pronunciation of Indonesian students is influenced by their mother language especially in aspirated sound [ph]. They are hard to pronounce [ph] because they do not have aspirated sound in Indonesian. When they learn [ph] sound, it is interfered by /p/ sound of Indonesian or /p/ without aspiration. This problem makes the researcher to find out the solution that is applying tongue twister to teach pronunciation of aspirated sound [ph]. The result of the research proves that students’ performance or ability in pronouncing aspirated sound increases. The theory is strengthen by a study on the influence of mother language to the target language conducted by Andi-Pallawa (2013). In the comparative study of phonological system between Indonesian in English, he found that both languages supply several similarities and differences. The similarities did not give significant impact but the difference put learners into trouble. Researcher found out that phonemic features: / p, t, k / were never aspirated in Indonesian words wherever they occurred. That is why Indonesian students had difficulties to pronounce [ph]. To find the solution of this problem, a previous researcher also used tongue twister to teach pronunciation. Iriyani (2015) tried to improve fifth graders' pronunciation of “th” by using tongue twister. The result of the study showed that there was improvement after the treatment was given. By pronouncing words over and over, students were aware of their mistakes and memorized vocabulary at the same time. Tongue twister also gave different atmosphere in classroom learning as it could break the boredom of monotonous learning process. The present study proved that fun media once again was well accepted by the student especially young learners and tongue twister as one of them gives significant influence in pronunciation practice of /th/ sound. In addition, tongue twister also works on other English sound such as [ph]. CONCLUSIONS According to the data analysis in the previous chapter, the conclusions could be drawn as follow 1. Students’ ability in pronunciation was in general low. Students’ sound production on aspirated sound [ph] was influenced by their mother language that does not have aspirated sound. Students mispronounced [ph] into /p/ as what they had in Indonesian. In addition, students did not get enough pronunciation practice in the classroom as teacher focused on the material such as grammar and reading. 2. There was difference in students’ ability in pronouncing aspirated sound [ph] between those who had been taught tongue twister and those who had not or who had been taught using conventional method in SMP Negeri 12 Magelang. Besides, based on the researcher’ observation, students enjoyed practicing tongue twister since it is fun and challenging. Students kept repeating unique series of words after a model on videos and they automatically trained their tongue to produce aspirated sound while pronouncing tongue twister. In short, students’ ability in pronouncing aspirated sound [ph] increased by using tongue twister and it was effective to use. A. Kholisoh, A. N. Farida / Journal of English Language Teaching 7 (2) (2018) 72 REFERENCES Andi-Pallawa, B. (2013). A Comparative Analysis between English and Indonesian Phonological Systems.International Journal of English Language Education, 1(3), 103-129. Retrieved February 9, 2017, from http://www.macrothink.org/journal/index.php/ijele/article/viewFile/3892/3180 Brown, H. D. (2001). Teaching by Principles: An Interactive Approach to Language Pedagogy (2nd ed.). White Plains, NY: Pearson Education. Harmer, J. (2001). The Practice of English Language Teaching (3rd ed.). Essex: Longman. Iryani, T. (2015). Improving Fifth Graders' Pronunciation of “TH” by Using Tongue Twisters (A Classroom Action Research at Fifth Graders of SDN Mojoagung 01, Trangkil-Pati in the Academic Year of 2014/2015). Final Project. Semarang State University. Prošic-Santovac, D. (2009). The Use of Tongue Twister in EFL Teaching. Annual Review of the Faculty of Philosophy. Novi Sad, 34, 159-169. Retrieved July 18, 2017, from https://www.researchgate.net/publication/277173037_The_Use_of_Tongue_Twisters_in_EFL_Teachi ng Saleh, M. (n.d.). Statistics in Educational Research: Handouts and Assignments. Semarang: UNNES Press. Wulandari, C. (2011). The Description of Pronunciation Practice in the Classroom (A Case Study of the Seventh Grade Students of RSBI Classes of SMP N 2 Wonosobo in the Academic Year of 2009/2010). Final Project. Semarang State University. Yoneoka, J. (2002). Englishes in Asia. Retrieved February 6, 2017, from www2.kumagaku.ac.jp/teacher/~judy/gyoseki/30asianenglish.doc