THE USE OF MODIFIED THINK-PAIR-SHARE IN TEACHING WRITING

Penelitian ini bertujuan (i) untuk mengetahui perbedaan prestasi menulis antara siswa yang diajar menggunakan strategi Modifikasi Think-Pair-Share (TPS) dan strategi Think-Pair-Share (TPS), (ii) untuk mengetahui korelasi antara sikap menulis pada siswa dan hasil kemampuan menulis siswa. Subjek penelitian ini adalah siswa kelas dua SMA IT Ar-Raihan Bandar Lampung. Hasil penelitian menunjukkan bahwa (i) ada perbedaan signifikan secara statistik dalam kemampuan menulis antara siswa yang diajar dengan strategi Modifikasi TPS dan strategi TPS, (ii) tidak ada korelasi antara sikap menulis pada siswa dan hasil kemampuan menulis siswa. Ini menunjukkan bahwa strategi Modifikasi TPS lebih efektif untuk meningkatkan prestasi menulis siswa. This study was aimed (i) to find out whether there was a significant difference in writing achievements between the students taught through the Modified Think-Pair-Share (TPS) and those taught through Original think-pair-share, and (ii) to find out whether there was a correlation between the students’ writing attitudes and their writing achievements. The subjects of this research were the second grade students of SMA IT Ar-Raihan Bandar Lampung. The result showed that (i) there was a statistically significant difference in the students’ ability between the students taught through modified TPS and those taught through TPS, (ii) there was no correlation between the students’ writing attitudes and their writing achievements. It can be said that modified TPS is more effective to increase the students’ ability in writing achievements.


INTRODUCTION
Writing is one of the skills in English that is perceived as difficult to teach as well to study. A lot of people consider teaching writing is not a simple task because of a variety of writing competences, arranging grammatical structures, organization, vocabulary, ability to give and appropriate assessment of students' written works. Richards & Renandya (2002) state that the skills involved in writing are highly complex. L2 writers have to pay attention to higher level skills of planning and organizing as well as lower level skills of spelling, punctuation, word choice, and so on. It becomes more difficult when their language proficiency is weak.
Learning to write a good writing is also not a simple task. Fareed & Bilal (2007) also states that most ESL learners find it hard to make a fine and coherent writing since they might feel there are some barriers. In relation to this latest problem faced by EFL learners, writing skill is considered challenging to EFL learners. Other than that, Nik et al. (2010) states that there are double problems for the EFL learners since the learners have to struggle on the acquisition of grammar, syntactic structure, rhetorical structure and the idioms of a new language. Regarding the statements previously, teachers have to find a way to support the teaching-learning process in order to be succesful. A suitable technique is needed to keep the students' willingness to write the whole paragraph with less difficulties. Besides, the technique should improve students writing achievement and encourage them to enjoy the teachinglearning process.
Think-Pair-Share (TPS) might be the suitable technique to improve students' writing skill. TPS requires the students to think on their own, then they should discuss their thinking with the partner near them, and the last is, they should share before the whole class what they have got. This technique has been proved to improve writing skill. The accountability of this technique is constructed due to the answer the students give to their partner and then partner must report to the class. According to Johnson & Johnson (1999) it could provide an opportunity for students to work in groups towards a common goal, increasing their own and others' understanding in a safe environment to make mistakes. According to Kagan & Kagan (2009) TPS technique consists of some steps, such as organizing students into pairs, posing the topic or a question, giving time to students to think, and asking students to discuss with their partner and share their thinking.
There are so many studies related to Think-Pair-Share. Siahaan (2014) conducted TPS technique to improve writing procedure text in the first grade of high school. The study showed that Think-Pair-Share (TPS) technique gives contribution to improve students' achievement in writing procedure text. Rosadi (2016) also tried to use TPS in improving writing skill of junior high school. It resulted the technique was able to improve writing skill. Another study conducted by Oktaviani (2017) also found that TPS is an effective technique to improve writing recount text for the first grade of high school. Yaqin (2018) in her study which used TPS in classroom action research also stated that the students' writing skill on descriptive text improved in each cycle after they were taught by using Think-Pair-Share technique However, according to Yulanda (2018) during her research in improving writing by using Think-Pair-Share, she realized that students still have weakness in writing skill such as difficult in gaining idea and less of vocabulary. Hence, the students must improve their ability in English writing by doing a lot of practice. In addition, Sugiarto & Sumarsono (2014) also state that students constraint in TPS was they had difficulties in delivering their idea and opinion during the pairing and sharing steps. In the typical application of TPS, the students are only expected to write and discuss without any guidelines before they construct the whole paragraph.
In accordance to help students to write, the teacher could also use outline as a technique in teaching writing. It also helped us to write the paragraph quickly because the ideas had arranged. It means that by learning an outline, the students improved their writing and made their paragraph easy to be understood, comprehended and organized. According to Muirhead (2006), outlining could promote a deeper understanding of the material because it requires the students to take time to create and to arrange information into sentences which reflect distinct aspects of critical thinking. It can stimulate them to have a logical thinking and organizational structure of ideas.
In addition, Reinking & von der Osten (2016) state that an outline can show you how to organize and develop your paragraphs. This also means that by using outline, the writing will focus on the topic. An outline is the blue print that shows the division and subdivision of your paper, order your ideas, and relationship between the idea and supporting details. It means that, the listed idea which is collected to arrange the paragraph consist of main idea and supporting details. Outline will lead to writing a better composition, as the subject of the composition and its purpose are clearly stated. It will help the writer stick to the subject and force the writer to list the ideas that will be presented. Thus, the outline assures that the composition will be completed in logical order. It means, outline is very useful to make a good writing, especially in describing the details of the subject.
Based on the previous studies stated above, Think-Pair-Share successfully proves to improve students' writing skill. It can activate their 'critical thinking' as well as their social intelligence. However, the studies are not yet further explored in organizing their ideas and opinion. It would be beneficial if the students are given the chance to write down an outline before they begin to make paragraphs.
Writing attitude, however, is also believed to have influence in students' writing achievements. Generally, Graham et al. (2007) define writing attitudes as an aff ective disposition involving how the act of writing makes the author feel, ranging from happy to unhappy. The complexity of writing as a task tends to add anxiety levels in students who are taking writing courses. This anxiety can often demotivate the student or lead to discouragement, and thus may result in negative attitudes towards writing. In addition, Dhadhodara (2017) believes that writing attitudes plays an important role in forming our view towards writing skill. The attitudes is highly effective on improving or hindering writing achievement.
In accordance to that explanation, the researcher was interested to investigate whether there is a significant difference between the students' achievements taught through modified TPS and those taught through original TPS. The researcher also wanted to find out whether there is any correlation between the students' writing attitudes and their writing achievements.

METHOD
This research used quantitative approach. The design for the first research question was control group pre test-post test design, where there were experimental and control groups. As for the second research question, ex post facto design was used. The independent variables were the strategies i.e. modified TPS and original TPS. The population of this research was the second grade students of SMA IT Ar raihan Bandar Lampung. The research used two classes at this school as the samples. They were XI A and XI B. For the data collection instruments, writing test and writing attitudes questionnaire were administered.
Independent sample t-test was applied to answer the first research question regarding the significant difference of students' writing achievements between experimental and control groups. Meanwhile, to answer the second research question, the researcher used pearson product moment correlation to investigate whether there was any correlation between the students' writing attitudes and their writing achievements.

RESULTS AND DISCUSSION
Independent group t-test on SPSS version 16 was used to analyze the difference in students' writing achievements between the students taught through modified TPS and those taught through original TPS. Meanwhile, repeated measure t-test was used to analyze the significant improvement of writing aspects in both classes. The table above indicates that the students' writing achievement has significant difference between those taught through modified Think-Pair-Share and those taught through the original one. It is revealed from sig 2tailed is lower than 0.05 (0.017<0.05) and also t-value is higher than t-table with the significance level of less than 0.05 (2.469 >2.011). As a result, the hypothesis is accepted.   Table 1 presents that there is a statistically significant difference between the modified TPS and the original TPS. It can be used to improve students writing achivements' gain better than the original TPS because in the modified TPS, the students used outlining first where they could generate ideas and organize it before they construct complete paragraphs. The outlining process could make them learn how to write the main ideas and supporting details that they want to write later. Besides that, the experimental class seemed to show better understanding and critical thinking skills on receiving the instruction and was able to pour their thoughts into composition better.
This study is supported by Muirhead (2006) who claims that creating outlines encourages students to devote more attention to organizing their ideas and make thoughtful decisions about their choices of words to communicate their ideas. Outlines can play a vital role as part of a comprehensive set of instructional strategies that can improve the quality of student writing and foster critical thinking skills. Another research from Salija (2017) also proved that writing strategy with outlining was an efficient and a successful strategy for essay writing. It was efficient since it could make the writing run smoothly as the result of already having the outline of ideas at hand, and was successful since it could help a writer successfully organize his ideas systematic and give focus on relevant materials as well as organize logical supporting details. Therefore, there is a significant different between those taught through the Modified TPS and those taught through the original TPS.
Both table 2 and table 3 show that all aspects in control class improved signficantly, while mechanics was the only aspect that did not improve significantly in experimental class. Furthermore, organization gained the highest in experimental class and language use got the highest gain in control class.
It is believed that outlining could contribute in organization aspect of writing. This research fully supported Kellogg (1988) who found that planning, in this case also outlining, improved the quality of the students' writing in terms of idea development and effectiveness of communication when it was compared to no planning. In the second experiment of Kellogg's (1990) study, he investigated whether planning, no planning, and mental outlining had an impact on the persuasive writing of 20 university students. This time, Kellogg (1990) found that planning improved the language use, organizational coherence, idea development, and effectiveness of communication of the text, but not fluency or mechanics.
On the other hand, this study did not fully supported Sahardin et al. (2017) who claim that TPS do not actually improve mechanics and grammar. This is probably because it is generally much easier to make improvements in organization, vocabulary and content but it requires a much longer effort with much more practice to significantly improve mechanics and grammar. In this present research, the language use aspect gained the highest in control class, and the second highest in experimental class. This could happen since Think-Pair-Share technique were implemented in both classes. The pair and share steps allowed the students to communicate and share a lot to their friends.
The students might get the highest gain on organization in experimental class because by using modified TPS the students could arrange their ideas on the outline form given. It required detail information that was needed to construct paragraph. So, the students could develop and organize their own ideas.
The last aspect to improve was mechanics. Both classes had the lowest gain of mechanics. This study supported Astika (1993) who states that among the five components, mechanics contributed the least. This may indicate that the subjects, when they wrote, they were more concerned with organization and development of ideas and meanings with less attention given to mechanics. Moreover, Kellogg (1990) found that planning improved the language use, organizational coherence, idea development, and effectiveness of communication of the text, but not fluency or mechanics.
Furthermore Sahardin et al. (2017) state that this is probably because it is generally much easier to make improvements in organization, vocabulary and content but it requires a much longer effort with much more practice to significantly improve mechanics and grammar. This condition might be caused by the students who only focused on the content of the text they write, and they were less aware of the little things of sentence construction and the conventions of grammar. In addition, Bacha (2002) explained that in her research that in writing process first drafts were read by the teacher mainly for content and organization, leaving language and mechanical aspects for a further draft.
The second research question was regarding the correlation between the students' writing attititudes and their writing achievements. The researcher gave questionnaire consisted of 19 statements to the students. The students' writing attitudes were categorized as positive attitudes (almost always and often), neutral (sometimes), and negative attitudes (seldom and almost never). The questionnaire of writing attitudes was 5 Likert scaled and was adopted from Knudson (1993).

Table 4. The Correlation between the Students' Writing Attitudes and Writing Achievements in Experimental Class
The table above reveals that the Pearson Correlation was 0.052, and the significant 2-tail was 0.806. Based the table above, the significant value was 0.806 which is higher than 0.05. Therefore, there is no correlation between the students' writing attitudes and writing achievement in control class.

Table 5. The Correlation between the Students' Writing Attitudes and Writing Achievements in Control Class
The table above reveals that the Pearson Correlation was 0.024, and the significant 2-tail was 0.910. These two score described the correlation between learners' writing attitudes and writimg skill. Since the significant value on the table above was 0.910 which is higher than 0.05, hence, there is no correlation between the students' writing attitudes and writing achievement in control class.
Based on the results, the researcher found out that most students in both experimental and control class have positive attitudes towards writing. Certain studies showed that negative attitudes towards writing resulted in no correlation with their writing achievements. A research done by Hanane (2015) showed that the results and findings of the research revealed that students have negative attitudes towards writing which affect their level of educational achievement. Previous research also showed that writing attitudes would go more negative as children get older. This is supported by Knudson (1991Knudson ( , 1992Knudson ( , 1993) who confirmed the attitudes decline with surveys in grades 1-3, 4-8, and 9-12. However, the statistics showed that there is no correlation between the students' writing attitudes and their writing achievements. The result did not fully support several studies where the relationship between the students' writing attitudes and their writing achievement has widely proven to have positive correlation.
Hashemian & Heidari (2013) stated that those participants who had positive attitude showed better operation in L2 writing than their peers with negative attitude. In addition, Bulut (2017) in his research, found out that writing attitudes and writing self-efficacy beliefs are highly correlated to each other and writing attitudes and writing self-efficacy beliefs directly and significantly affect summary writing. Al-khayyat (2018) in his research of University EFL students' attitudes towards writing claimed that there is a significant positive correlation between the students' attitudes and achievement. Another research from Al-Sobhi et al.
(2018) revealed that the Arab ESL secondary school students have a high positive attitude toward English spelling and writing. In addition, the study revealed that there is a significant positive relationship between the students' attitude toward spelling and writing.
On the other hand, this present result is supported by Kotula et al. (2014) who state that although students' attitudes about the Value of Writing component appeared to show reasonable levels of internal consistency (Cronbach's alpha coefficients of 0.80), that attitudes variable demonstrated the lowest correlation with students' writing. It appears that although students may value writing, and they recognize that it is an important skill that is valued in school, their performance did not correlate highly with it.
Furthermore, Olinghouse & Graham (2009) in their study showed that writing attitude was not correlated with a measure of writing achievement or quality. They reported negative correlations between writing attitude and story quality and between writing attitude and story length. The studies reviewed by Ekholm et al. (2018) suggest that writing attitude is a malleable belief. The reason is the students' attitudes toward writing seem to be sensitive to various types of writing interventions, including technology interventions, strategy interventions, and changes in the classroom's general approach to writing. Other than that, the review also indicated that students' writing attitudes decline as they age. Thus, it might be concluded that writing attitudes could not directly affect writing achievements, yet the technique or strategy used could influence the connection.
From the discussion above, it can be concluded that there is no correlation between the students' writing attitudes and their writing achievements. As children are getting older they might have attitudes' decline towards writing. In addition, most of the students, although they have positive attitudes to writing, they see writing as difficult and stressful. This might also be the reason their result of attitudes and their achievement were not correlated significantly.
Furthermore, attitude is something that someone's certain feeling to refuse or accept something based on his opinion, view and also experience towards something. So, writing attitudes might influence writing achievements. However, it could not affect the achievements directly. The asssociation between writing attitudes and writing achievements might be influenced by the activities applied in the classroom.

CONCLUSIONS
Referring to the discussion of the research findings on the previous chapter, the researcher comes to these following conclusion. Based on the research, it can be concluded that there is a statistically significant difference in writing achievements between the students taught through modified Think-Pair-Share (TPS) and those taught through the original Think-Pair-Share. In addition, the students who were taught through modified TPS have the higher gain than those taught through the original TPS.
Yet, both classes had differences in aspects' significant value. The original TPS increased all the aspects significantly. Meanwhile, the modified TPS got all aspects significantly increased, except for mechanics. Thus, the original TPS is better in increasing all aspects significantly, while the modified TPS could significantly increase the gain in the students' writing achievement.
Furthermore, modified TPS could enhance the students' writing achievements in organization aspect. The result shows that organization has the highest increase in modified TPS could be caused by the outline applied on the second step of modified TPS. In contrast, language use reached the highest gain in original TPS. The 'pair' and 'share' steps might be the reason why their language use improved.
However, both classes get mechanics aspect at the very least to improve. In the control class, all writing aspects improved significantly, yet only mechanics which did not increase significantly in experimental class. The students might be more focus on the organization and development of ideas when they wrote. Furthermore, the time limitation could also affect the result of mechanics, since they had to do the writing twice; making the outline and making the whole composition.
Lastly, the result shows that most of the students of both classes have positive attitudes towards writing. However, there is no correlation between the students' writing attitudes and their writing achievements. It is believed that writing attitudes is a malleable belief since it seems to be sensitive to various types of writing interventions. Furthermore, as the children get older, they might have attitudes' decline towards writing. In addition, it is also believed that writing attitudes could not directly affect the achievement, yet activities or language learning strategies applied might play a role in determining the result.
beliefs on their summary writing achievement.