Improving Students’ Speaking Confidence And Fluency Using Guided Speaking Technique With Heading And Sub-Heading

The English Proficiency Index of the English learners in Indonesia is considered low. It showed an average score of 52,15. The research objective is to examine the difference in speaking fluency and confidence of students taught using guided speaking technique and conventional technique, and whether there is any correlation between their fluency and speaking confidence. This research applied quasi experiment method. The respondents were the 11th graders of SMA Smatag Surabaya. The fluency t-test result Sig (2-tailed) was 0,000 < 0,05 which showed that there was significant difference in fluency, where treatment group achieved significantly higher. It confirmed that the method proved effective to improve speaking fluency. With the speaking confidence variable, the t-test result Sig (2-tailed) was at 0,25 > 0,05 showed that there were not any significant differences in the confidence between both groups confirming the method of guided speaking was not able to significantly raise the confidence level. Furthermore, overall Pearson Correlation value between speaking fluency and the confidence were > 0,05 which confirmed that there was not any significant correlation between the two variables. Based on the finding it can be concluded that guided speaking technique is effective in improving the speaking fluency and is recommended to be applied in school.


INTRODUCTION
The English Proficiency Index of average Indonesian falls on the low band category at 52,15 making Indonesia ranks 39 out 180 countries worldwide (Renandya, Hamied, & Nurkamto, 2018).The rate has been consistent for the past 5 years.English is learned for knowledge and skill.It is an indication that average knowledge and skill of Indonesian has to be up graded.Of all the four skills learned in school, speaking skill is the most challenging for the following reasons, big classes in school, short learning hour, lack of practice outside class, lack of confidence, it is carried out spontaneously and real time, and that

Journal of Education Research
English is taught with focus more on grammar accuracy and less on the communicative aspect (Andriani Putri, Amri, & Ahmad, 2020).
Speaking is a communicative activity.Students learn English by using it to communicate.In the process the students need competences which include linguistic competence, socio cultural competence, interactional competence, strategic competence, and discourse competence to be able to carry out effective communication.The central competence is discourse competence which include knowledge and discourse.Discourse competence the production of intelligible, organized speech and the ability to connect stretches of utterance to produce socially and pragmatically appropriate speech (Burns, 2019) This ability contributes substantially to the improvement of fluency.
Fluency consists of two aspects, affective and competence.The affective aspect is reflected in the confidence and competence in the fluency.Fluency is strongly correlated with confidence.Confidence ensures higher fluency and vice versa fluency ensure higher confidence.(Tridinanti, 2018).Therefore, the low competence as shown by the English Proficiency Index is partly due to low confidence.Low confidence is due to several reasons, fear to make mistake, self-consciousness or lack of knowledge.
Speaking activity in classes is not about learning speaking but it is about doing speaking.Doing speaking constitutes the active use the language in the class as much as possible and give the student all the opportunity to spend the best of their time in an authentic, meaningful, and contextual communication (Burns, 2019).
Speaking activity comes in two broad category, free speaking and guided speaking (Farabi, Hassanvand, & Gorjian, 2017).In the free speaking activity learners are not given enough guidance before they do the speaking task.this may work well with student with high level of confidence.However average students do not possess such ability due to self-consciousness, worry to make mistake and shyness (Anggreni, 2019) And the activities do not proceed as expected.Whereas in guided speaking students can better proceed since before doing the task students are given guidance in form of work sheet, mind map or question.Students are provided with track to follow to ensure a more fluent speech.
To solve the problem a method should be applied that can provide guidance, hence guided speaking.The type of guidance focused in this research is the use of heading with sub heading with which student can proceed in the speaking task more comfortably and be less worried with mistake, shyness et cetera as they are provided with track to follow.
Research on the English proficiency of Indonesian English speakers showed that their average proficiency score was 52,15 which fell on low category and put Indonesian 20 th in rank in Asia.The low proficiency is attributed to the on-going current method of English teaching which should be improved especially in the area of speaking skill teaching.A new approach needs to be introduced.
The research objective is to examine the difference in the speaking fluency and confidence between the students taught using guided speaking technique with heading and sub-heading and taught using conventional technique.Furthermore, another aspect to be examined is whether there is any correlation between their speaking fluency and speaking confidence.
Previous researches on the guided technique have given proof that the technique was effective and worked well.Each of the study applied a variety of method in applying the guided speaking technique among others, worksheets (Chandra & Hayati, 2017), mind-maps (Mustajib, 2017), and pair-taping (Herlina & Holanddyah), 2016).In this research, a novelty is presented which is the use of heading and sub heading to provide students with track to follow to produce more fluent, more accurate and better organized speech.

RESEARCH METHOD
This research was conducted using quasi-experimental and pre and post-test designs involving two groups, a treatment group, and a control group, to get information on the causal effect of a particular treatment.Moreover, the assignment is non-random.This method does not result in the same way as a pure experiment, due to the impossibility of strictly controlling or manipulating the relevant variables such as age, intelligence, interest, et cetera (Wibawa, 2019) This study focuses on using the guided technique to improve students' speaking fluency and confidence.The sample of the research is divided into control and treatment groups.The control group Journal of Education Research was treated using the conventional method, while the treatment was treated using guided speaking method.The research was done in 3 stages: pre-treatment, treatment, and post-treatment.In the pretreatment stage, the students of the control and experiment groups as the sample underwent a pre-test speaking (Appendix 1) and pre-test confidence (Appendix 2).Speaking test students are required to speak on one topic with the heading and subheading provided (Appendix 3) and complete confidence questioner.During the treatment stage the control group was treated using the conventional method, and the experiment group was treated with a guided technique of which the lesson plans are provided (Appendix 4).During the post-treatment stage, both groups did a post-test speaking and post-test confidence.The speaking test required students to speak on one topic and complete confidence questioner.
The research of this study was conducted in SMAN 17 Agustus 1945 Surabaya, SMATAG Surabaya.There are 167 students in grade 11 divided into six classes and by drawing 2 classes were designated as the sample which were class IPS 2 as treatment group and IPS 1 as control group The research instruments were pre and posttest for speaking fluency and pre and post-test confidence questionnaire.The purpose of the speaking test was to measure the fluency and the reference of the test was the English syllabus for class eleven based on Curriculum 2013 stipulated by the Ministry of Education (Permendikbud, 2018).Specifically refer to Standard competence 4 and Basic Competence 4.2.The speaking test consist of one test reflecting current and trending topic titled The Media Social Today.The students were required to do oral presentation in 3 minutes on the above-mentioned topic.The scoring used speaking test rubric consisting three elements namely fluency, accuracy, task completion.Before being administered the test went through validation and reliability test and considered valid and reliable.
The second test was confidence questionnaire.The purpose was to measure the speaking confidence.The questioner was based on the definition of Confidence from a psychological journal 'Improving Self-Confidence through counselling Directive' ( Indrawati, 2018).It consists five dimensions namely believe or trust in one's ability, goal, opinion, strength and believe in the guided speaking method.Before being administered the test went through validation and reliability test and considered valid and reliable.Below is the figure of the research procedure.

FINDING AND DISCUSSION
The findings are classified into three parts as specified by the objective of the research.First on determining the difference on level of confidence between students taught using guided speaking technique and using conventional method, second on the level of speaking fluency and third on the correlation between their speaking confidence and speaking fluency.
First, the score of post confidence test of treatment group and control group are as follow, the mean score of post-confidence in Ability dimension by the experiment group was 72,00 whereas by the control group was 68,83.Treatment group showed a higher score.On Goal Dimension The result showed that the experiment group obtained the mean score 75,60 while the control group gained 79,87.The treatment group showed slightly lower score.On Opinion Dimension The result showed that the experiment group obtained the mean score 74,03 while the control group gained 69,50.The treatment group showed higher score.On Strength Dimension The result showed that the experiment group obtained the mean score 75,33 while the control group gained 74,00.The treatment group showed slightly higher score.On Guided Speaking Dimension The result showed that the experiment group obtained the mean score 76,20 while the control group gained 71,20.The treatment group showed higher score.In summary the Treatment Group scored higher in 4 dimensions.
Furthermore, the t-value shown in the table for Ability Dimension was 1,160 and sig (2-tailed) was 0,251 which was higher than 0,05.On Goal Dimension the t-value was 1,302 and the Sig (2-tailed) was 0,198.On Opinion Dimension the t-value was 2,017 and the Sig (2-tailed) was 0,48.On Strength Dimension the t-value was 0,471 and the Sig (2-tailed) was 0,639.On Guided Speaking Dimension the tvalue was 1,788 and the Sig (2-tailed) was 0,79.In all of the dimensions the Sig (2-tailed) are higher than 0,05.If the sig (2-tailed) is higher than 0,05, there is not any significant difference between score achieved by experiment and control group.Consequently, null hypothesis is accepted and alternative hypothesis is rejected.There is not any significant improvement of the students' speaking confidence taught by using guided speaking technique with heading and sub-heading rather than taught by using conventional technique It indicated that the confidence level of the experiment group was slightly improved.
The following is the graphic illustration of the post confidence test between the Treatment Group and Control Group.

Confidence Level Treatment Group & Control Group
Second, the mean score of the post-test speaking fluency by treatment group is the experiment group was 78,87 whereas by the control group was 54,33 The result showed that the experiment group achieved substantially higher Furthermore, the t-value shown in the table was 5,352 and sig (2-tailed) was 0,000 which was lower than 0,05.If the sig (2-tailed) is lower than 0,05, there is a significant difference between score achieved by experiment and control group where the experiment gained considerably higher.Consequently, null hypothesis is rejected and alternative hypothesis is accepted.Third, the Pearson correlation between the post-speaking fluency score of the experimental group and the post-confidence score on ability dimension was 0,320 and Sig (2 tailed) was 0,084 and The Pearson correlation between the post-speaking fluency score of the experimental group and the post-confidence score on goal was 0,382 and sig (2-tailed) was 0,37.
The Pearson correlation between the post-speaking fluency score of the experimental group and the post-confidence score on opinion was 0,210 and Sig (2-tailed) was 0,264.The Pearson correlation between the post-speaking fluency score of the experimental group and the post-confidence score on strength was 0,252 and Sig (2-tailed) was 0,179.The Pearson correlation between the post-speaking fluency score of the experimental group and the post-confidence score on guided speaking was 0,234 and sig (2-tailed) was 0,214.
The Pearson Correlation value and Sig (2-tailed) of the correlation analysis between the postspeaking score and the post-confidence score obtained by the treatment group on all dimensions were more than 0,05.If the Pearson Correlation and Sig (2 tailed) are more than 0,05, there was not significant correlation.It is therefore confirmed that there was not any significant correlation between score of postspeaking test and score of post-confidence on all five dimensions of confidence questionnaire in this study

T H E RE S ULT O F PRE AND PO S T S PE AKING T E S T O F T RE AT ME NT GRO UP
Pre Test Post test by the treatment group taught using guided speaking technique.Therefore, hypothesis null is accepted and hypothesis alternative is rejected.
The following is the graphic illustration of the correlation between the post speaking test and post confidence test of the treatment group.

Table 4. Pearson Correlation of Post Speaking & Post Confidence of the Experiment Group
In the following section the finding above are further discussed including the three topics, speaking confidence, speaking fluency and correlation between confidence and fluency 1. Confidence During pre-treatment stage, both groups in the confidence test showed high score whereas in the speaking test both groups showed low score.These figures seem unusual since the score of speaking is considered low but the scores for confidence are relatively high.It indicated that the students as research object show unusually high confidence in English although their level of English in speaking skill in particular is low.
Furthermore, Indrawati (2018) states that the aspects of self-confidence includes, the belief in one's ability, goal, opinion, and strength.The high score in confidence test reflects that they have strong believe in their ability, their strength, their opinion and in their English fluency.But again, their level of English in speaking skill does not equal to their level of confidence.
Tridinanti, (2018) stated that speaking fluency and confidence are correlated.Therefore, the increase in confidence will increase the level on speaking fluency and vice versa.However, the result of this study showed the increase in the speaking confidence of the treatment group after treated with guided speaking did not increase the level of confidence.It is an indication that the majority of students are in possession of high level of confidence which however, unmatched to their low speaking ability as shown in the pre-test result in speaking fluency test Several dispositions are required to master English.One of them is confidence.High number of students fail to be able to speak English fluently due to the lack of confidence However, they should remain realistic about their ability ("Confidence | Psychology Today," n.d.) It is a strange phenomenon where some group of people show unusually high confidence which is termed as overconfidence.The overconfidence phenomenon or overconfident effect occurs when people's confidence in their capability is higher than their accuracy.They are more confident in their judgement than they are correct.It is a kind of cognitive bias.The overconfidence or overconfidence effect is observed when people's subjective confidence in their own ability is higher than their objective or actual performance (Parker & Stone, 2014) It can be confirmed that there was a factor of overconfidence as shown by the high average confidence score at 74,00 for experiment group and 71,2 for control group Journal of Education Research although the average speaking fluency was 57,77 for experiment and 53,73 for control group in the pretreatment stage.Therefore, due to the phenomena of overconfidence, the effectiveness of the guided speaking was not proven 2. Fluency During pre-treatment stage, either experiment group or control group showed the level of fluency and accuracy at the rate of 57,77 for experiment group and at 53,73 for control group.These figures confirm that they have low fluency.They showed hesitation and pauses in delivering the oral presentation that indicated low fluency and affected the low figure above.Language proficiency depend on fluency and accuracy.Fluency is the capability to express idea or intention with minimum hesitation and pauses that can impede or hamper communication.There are observable signs that can be used to measure changes in fluency these include speech rate (as measured in words or syllables per minute), number of filled or unfilled pauses.(Newton & Nation, 2009) .The most important indicator for language proficiency are fluency, accuracy, followed by complexity and appropriacy (Renandya et al., 2018).The speaking fluency is an important component of the communication competence.Apart from fluency, other aspect of language skill is accuracy.The figure above also indicate that they have at the same time low level of accuracy.Accuracy refers to the use of correct and proper form of language without error that can affect the linguistic and discourse feature of language However, after the treatment, the experiment group showed significant increase in the level of fluency with the mean score at 78,87 using guided technique whereas the control group was stagnant with the score at 53, 33.It is solid indication that the guided speaking method proved to work well and are able to give opportunity to English learners alike to be able to communicate in English more effectively since learner learn a language is to be able to speak it (Richards, and Rodgers, 2014) The increase in fluency is in agreement with the previous study.In this study, speaking fluency rubric is used to determine the speaking proficiency.With proven increase in speaking fluency score, the guided speaking in general and guided speaking with heading and subheading has made contribution to the improvement of English teaching method.3. Correlation between speaking fluency and speaking confidence.
The following previous studies presented contradicting results . Tridinanti, (2018) states that there is a strong correlation between confidence and speaking fluency.It implies that the more fluent a speaker is, the higher the confidence is.However, in this study especially for the experiment group, the level of speaking fluency increases significantly from 57,77 to 78,87 while the confidence level remains low.The reason was due to the phenomena of overconfidence.
There is significant correlation between confidence and speaking fluency.Speaking serve as facilitator to start any language interaction, without confidence effective interaction cannot take place.Gürler (2015), This theory implies that low confidence makes it impossible to increase speaking ability.This significant increase proved that there is not any correlation between confidence and fluency, which contrary to the theory above.the factor that interferes in this study is the existence of the phenomena of overconfidence or Miscalibration of confidence which made the level of confidence higher than the actual one Miscalibration of confidence occurs when confidence is not a signal of knowing or being fluent.How students' experiences and beliefs about fluency can create miscalibration of confidence (Finn & Tauber, 2016) Students rate their confidence higher than the actual one.

CONCLUSION
There was significant difference in the speaking fluency between the treatment group taught using guided speaking technique and control group taught using conventional method where the first group achieved higher.Whereas in the level of confidence there was not any significant difference in the level of confidence between both groups.Furthermore, there was not any correlation between speaking fluency and speaking confidence.
The application of guided technique showed positive result in improving the speaking fluency.Therefore, to the teacher or educator alike the technique is highly recommended to be implemented in the classroom by providing them with accurate, simple heading and subheading in accordant to their level of

Figure1.
Figure1.Procedure of Research The data was analyzed using statistical tests.To answer the first problem question finding the difference of speaking fluency and confidence between treatment group and control group, group statistic test was used which included Levine Test and t-test for equality of means to obtain the F and Sig (2-tailed) value.Furthermore, to answer the third problem question, Pearson Product Moment was used to obtain the correlation value and Sig (2-tailed) value.
A N D P O S T S P E A K I N G T E S T improvement of the students' speaking fluency taught by using guided speaking technique with heading and sub-heading guides rather than taught by using conventional technique It indicated that the speaking fluency of the experiment group was more improved.That meant that teaching using guided speaking proved effective in improving fluency.The following is the graphic illustration of the post speaking test between the Treatment Group and Control Group, pre and post-test of treatment group, pre and post-test of control group Table 2. Result of Post Speaking Test of Treatment & Control Group Tabel 3. Comparison of Pre and Post Test Results of Treatment Group