CORRELATION INQUIRY BETWEEN STRUCTURE KNOWLEDGE AND PARAGRAPHS WRITING PERFORMANCE

This paper aims at investigating the correlative aspect between two variables; structure knowledge and writing performance. It comes up in analyzing the theories and results of some studies that structure knowledge would or would not emphasize in writing performance. It indicates that this phenomenon is being still debated now on. On the sake of this purpose, the research methodology used is correlation design, where it took all the second semester students English Study Program of FKIPIslamic University of Indragiri as sample. In collecting data, there are two kinds of instrumentation tests; structure knowledge and writing performance tests. While analyzing the data, the researcher used Pearson Product Moment correlation statistical analysis. The finding of the study reveals that there is a positive correlation between the students’ structure knowledge and their writing performance. It is obviously proven that the coefficient correlation is 0.84, which is categorized as high correlation. After testing the hypothesis, the researcher could determine that there is a significant correlation between two variables, since the critical value of rxy is greater than the critical value of rtable in significant level 5%; (rxy = 0.84 > rtable = 0.227). It emphasizes that null hypothesis is rejected and alternative hypothesis is accepted. It infers that who want to have better writing performance should learn structure.


INTRODUCTION
The goal of English learning is to be able to master the four language skills.Those skills support each other.Language learners who intent to explore productive language skills (speaking and writing) should enrich receptive language skills (listening and reading), because of having better the receptive skills knowledge assists them to gain better productive ones.Certainly, language components cannot be separated from them, since it plays essential role to acquire the language skills well.Dawson (1984:4) views that learning a language purposes learning pronunciation, vocabulary and grammar or structure by practicing the four skills listening, speaking, reading and writing.In other words, the language skills work integratively.
Grammar, as one of language components, displays substantial point in language skills.For instance, someone's grammar competence effects to language skills competence.Harmer (2001:12) defines that the term of grammar as the description of the ways in which words can change their forms and can be combined into sentences in that language in order to gain meaningful sense.In addition, Thornbury (2006:13) states that grammar adds meanings that are not easily inferable from the immediate context.In the same point of view, Chin (2000) says that grammar is the sound, structure, and meaning system of language.Each language possesses of its own grammar.People who interact the same language enable communicate as they naturally detect the grammar rules of that language which is the rules of making meaning spoken and written forms as well.That means grammar is the basic component to create good English, particularly on writing as an important part in a process of learning a language itself.As one of the major skills in language learning, writing is a tool of transferring one's notion to the audiences.It is also stated that writing is a vitally complex demanding process (Saadian & Bagheri: 2012:1).Moreover, it is a complex performance because there are some components should be focused on writing such as the purpose of writing and writer's knowledge of writing (Reid, 1988).
In learning writing there are several aspects should be considered, namely fluency (how easy a writing to understand), organization (form), vocabulary (wording), language use (grammar), and mechanics.
Theoretically, both (structure knowledge and writing performance) have closely relationship.This notion is supported by Hughes (2005); Heaton (2005) that grammar is the component of writing performance.It means that the students' grammatical knowledge and their writing performance influence each other.To strengthen this assumption, Doff (2000) states that in the case of writing, grammar permits writers to generate their opinions into well-organized sentences so that they enable successfully communicate in a written form.In other words, by learning grammar, writers enable to transfer meaning in the forms of phrases, clauses, and sentences.Then, Frodesen and Eying (2000:23) sharpen that a concern on grammatical rules in writing can assist writers explore more linguistics resources needed to express idas effectively.
To support those notions, some related studies had been carried out Istiqomah, et.al (2012) investigated the relationship between grammar mastery and descriptive writing ability at SMAN 1 Terusan Nunyai.As a result, they determined that the students' grammar mastery had high correlation to their writing skill.In other words, the students who want to improve their writing ability should learn grammar.Other inquiry was also conducted by Saadian & Bagheri (2013).It purposed to find out if grammar and vocabulary knowledge could contribute significantly to Iranian EFL learner's writing competence.The result revealed that there is a high and significant correlation between grammar and vocabulary knowledge and the learners' writing performance.In other words, grammar and vocabulary knowledge can assess writing scores.
On the other hand, other study which was conducted by Septiani (2012), she investigated the relationship between grammar mastery and writing performance at the 6 th students of English Department of UIN Syarif Hidayatullah Jakarta.Her finding shows that grammar mastery and writing performance had very low correlation.In short, there is no correlation between both of them.It inferred that students who have minimum level of grammar, it does not mean that the students also get poor achievement in writing.Moreover, Huang (2011) has conducted a study for discovering the answer whether EFL students' grammatical ability account for writing ability.The results indicated that the most students' grammar subtests outperformed their writing subtests.The finding implied that there was no strong relationship between the knowledge of grammar and usage among lower-intermediate learners.
Those previous studies findings set over that the role of grammar in writing has been extensively argued and yet not reached a consensus conclusion.It works out the researcher to drive other findings if or not grammar knowledge hand over positive and significant relationship toward writing performance.
Based on the curriculum of English Study Program FKIP-Islamic University of Indragiri, the second semester students learn Writing I. Its main aim is how students are able to write a good paragraph.Moreover, they also learn Structure II, which focuses on English grammatical systems that are very essential to enhance their English proficiency, particularly on written forms.
Relating to the content of their curriculum, this inquiry was limited on variables; students' structure knowledge, which was limited on six kinds of tenses; Simple Tense, Future Tenses, Past Tense, Present Perfect, Present Progressive, Past Progressive, and Past Perfect.Then, the students' writing performance was focused on writing paragraphs.
Dealing with the phenomena and the theories above, it could be inferred that one of factors influencing the students' writing performance is grammar (structure knowledge).Therefore, in order to master writing, the students have to possess a large amount of structure knowledge.It is expected by having adequate structure; the students would be able to write well.In other words, can the students write English well if they posses more structure knowledge?
This study was done to investigate: (1) how the students' structure knowledge is, (2) how the students' paragraphs writing performance is, (3) how the correlation coefficient between the students' structure knowledge and their paragraphs writing performance is, and (4) whether there is a significant correlation between the students' structure knowledge and their paragraph writing performance.

METHOD
The participants were 77 the second semester students of English Study Program FKIP-Islamic University of Indragiri in 2013/2014 academic year.They were selected from 3 parallel intact classes by using total sampling technique, which all the population desired are selected as sample (Riduwan: 2005:64).
The correlation research design is used to determine the research purpose.Correlational research involves collecting data in order to determine whether, and to what degree a relationship exists between Vol 3, No 2, August 2016 Correlation Inquiry between Structure Knowledge and Paragraphs Writing Performance two or more quantifiable variable.Then, the degree of relationship is expressed as correlation coefficient.If a relationship exists between two variables, it means that scores within a certain range on one variable are associated on the other variable (Gay and Airasian;2000:321).

Table 1. Blue Print of Structure Test
The instrumentation devices of the current study were as follows: Structure Test: It purposed to determine how well the students' structure knowledge.It consisted of several kinds of tenses; Simple Tense, Past Tense, Future Tense, Present Perfect Tense, Past Perfect Tense, Present Progressive, and Past Progressive Tense.The instrument was adapted from textbooks for preparation and grammar books on structure.In this case, the students were asked to complete the sentences in a paragraph by using the appropriate tenses for the verbs in parentheses.They must comprehend how to utilize correct verbs forms based on the time signal handed over.There are 35 questions or verbs that must be answered by the students.The duration time for doing the test was 50 minutes.
Writing Test: It aimed at measuring the students' performance in writing paragraphs.They were asked to select two of five topics given: (1) The Seasons in My Country, (2) English for Future, (3) Unforgettable Experience, (4) An Ideal Campus, and ( 5) An Ideal Student.The length of paragraphs was not be decided, that means it depended on the students' capability in developing their ideas in writing.The time provided for doing the test was 50 minutes.Furthermore, in analyzing the structure knowledge test, the researcher utilized descriptive statistics analysis.Meanwhile, the writing performance test was analyzed by using writing scoring rubric which was proposed by Hughes (2005: 91-93).
It was comprised five writing indicators; grammar, vocabulary, mechanics, fluency, and organization.Each component was scored within scale 1-6.The students' level ability could be classified into five interval classes: Good to Excellent (80-100), Average to Good (60-79), Poor to Average (50-59), and Poor (00-49) (Haris, 1974:134)  Once all the tests were administered and corresponding data were gathered, data analysis phase began using the SPSS package.To evaluate the test takers' performance in writing paragraphs with their structure knowledge as variable, bivariate correlation analysis was run to measure the degree of relationship between variables.The results would us understand how structure knowledge could improve the candidates' writing competence.Moreover, a correlation coefficient between the obtained scores was calculated which would assist us make the results of the inquiry more reliable.

FINDINGS AND DISCUSSION a. The Description of the Research Variables
This research was conducted in order to find out how close the relationship between two variables; the students' knowledge of structure as the independent variable (X) and their ability in writing a paragraph as dependent variable (Y).
The scores of the students' structure knowledge were obtained through a structure test.The test consisted of 35 items.And the scores of the students' writing ability were obtained by using written test.In this case, the students were asked to write a paragraph based on the topics given.The research data of structure knowledge were gained as the test to the desired sample.The students' scores of structure knowledge can be seen as follows: In this case, it could be inferred that the students still got confuse in using some kinds of tenses to complete sentences in paragraphs.Supportive point was shown that more than 50% of the students were classified in poor to average level.In other words, using appropriate verbs forms based on the situation given was not easy.The students needed to focus more on analyzing the time signal and then deciding the correct tenses should use.

c. The Students' Ability in Writing a Paragraph
To avoid the subjective scoring in analyzing the data of writing test, the researcher was assisted by the three raters.Then, he combined the scores of three raters to gain the average score of students' paragraphs writing performance, which can be seen as follows:

e. Linearity Testing
One of requirements as a ideal data in correlation research designs is the data must be in linear.The linearity testing, in this research, was analyzed by using SPSS program.The data is categorized as linear if the F-significant value is less than 0.05 (sig.value < 0.05) and vice versa.The results could be seen in table 6.
It shows that the F-significant value was 0.000.It could be found that the F-significant value was less than 0.05 (0.000 < 0.05). in other words, the data was linear.

f. Correlation Coefficient
Two variables investigated (structure knowledge and paragraphs writing performance) were analyzed by using statistical analysis.To determine how close the relationship between them, the researcher used Person Product Moment Correlation analysis.Then, after computing the data it could be found that the correlation coefficient between variables observed was 0.842.It was classified as high and positive correlation.This infers that the students' structure knowledge has strong relationship toward their paragraphs writing performance.Moreover, R square score (the contributions of variable X toward Y) was 0.709; it means that variable X (structure knowledge) contributed 70.9% toward variable Y (paragraphs writing performance), and 29.1% is effected by other variables.In a nutshell, the language learners who intent to be better in writing skill could learn much in structure knowledge.

g. Hypothesis Testing
There are two hypotheses need to be answered of this research, the hypotheses are there is a significant correlation between students' ability in mastering structure and their ability to write a paragraph (H 1 ) and there is no significant correlation between students' ability in mastering structure and their ability to write a paragraph (H 0 ).
To test the hypotheses, the researcher utilized two systems; using Person Product Moment statistical analysis and SPSS program.These following criteria are as consideration: If the "r" observed is greater than "r" table the null hypothesis is rejected and alternative hypothesis is accepted and vice versa.Or if the sig.(2-tailed) value is greater than 0.05 the null hypothesis is accepted and alternative hypothesis is rejected and vice versa.
After calculating the data, it was found that the value of r xy is 0.842, and the value of the degree of freedom (df) is 75 (df = N-nr ) (df = 77 -2 = 75).Then,  (084 > 0.227).Next, the hypothesis could be also tested from the value of Sig.(2 tailed) was 0.000.This score was less than 0.05 (0.000<0.05).It can be concluded that the Null hypothesis is rejected, and Alternative hypothesis is accepted.In a nutshell, there is a significant and positive correlation between students' structure knowledge and their ability in writing paragraphs.

DISCUSSION
Dealing with the data analysis description results, it was gathered some findings that need to be discussed.
The students' structure knowledge results revealed that most of the students were classified into poor level.42.9% of the students were in that interval range.Then, the mean score supported that the students were placed in poor to average level.It seemed that the students still got troubles in using some kinds of usage components, particularly on tenses.Margaret (2005) clarifies this finding that using a appropriate kinds of tense is not easy due to time signals.It means that the students should comprehend when the sentences structure will be used.To support the results of the study, Rahman and Ali (2015:131) on their study which investigated about problems in mastering English tenses and aspect and the role of practitioners found that a lot of EFL leaners fail to master tses and aspect even though they spend substatial amount of time in order to master them.The use of verbs forms must take main priority.Cowan (2008:350) emphasizes that use of verb forms is one of the most difficult areas for English language leaners to master.It can be deduced that mastering English tenses is not easy, due to tense is related to time, and time refers to 'when' an action takes place.
The writing paragraphs performance results showed that the students' skill was still low.It can be proven dealing with 31 (40.2%) of samples were ranged in to average level, and not more than 10% of them who could achieve good to excellent interval.Moreover, after summing up all the score, it revealed that the students' average score placed on poor to average level.On other words, they still got some problems in writing paragraphs.Jimenez, el.al (2013:6) state that writig skill problems are one of the major substantial diffculties that impact both native speakers and hundreds of langauge learners around the world.No having interest in witing field leads them to be poor writers, have low scores in their courses, increase errors in their assignments, write run-on sentences, and create incoherent paragraphs.Furthermore, Jimezed, et.al (2013:11) also emphasize that one of the major difficulties among language leaners in fact that a lot of them cannot develop their writing profeciency.Possibility, those problems which lead the sample of this current inquiry had low level in writing performance.
Investigated variables, structure knowledge and writing paragraphs performace, had positive relathionship.This finding was analyzed statistically that correlation coeffiient between observed variables was 0.842, which placed as high correlation.Moreover, the hypotesis testing also showed up that the null hyphotesis was rejected, meanwhile work hyphothesis was accepted.Thus, there was a positive and significant relationship between the structure knowledge and writing paragraphs profiency.According to Best and Khan (2006:379) that a perfect positive correlation is +1.00, a perfect negative correlation is -1.00, and a complete lack of relationship is zero (0).In line this finding, Fatemi (2008:24) supports that the role of structute Additionally, several studies results have shown that one's structure knowledge has positive contribution toward his/her writing performance.Suthiwartnarueput and Wasanasomsithi (2012) have done an inquiry which investigated the effect of using facebook as a medium for discussion of English grammar and writing of lowintermedaite EFL students.One of their research findings revealed that the students' grammar competence contributed to their writing performance significanly.Then, Dewi (2013) also executed a research which proposed to gain empirical data of determining the correlation among mastery of sentence structure and dictation, either individually and simultaneously toward writing skill at IX grade of SMP Al-Syuro Tangerang Selatan.The results revealed that the contribution of mastery of sentence structure and diction toward writing skill was 22.75%.
Though, they could contribute only less than 50%, among those variables had significant and positive relationship.In a short, the researcher summed up that language writers need to pay attention to form (grammatical system) in developing writing performance.Moreover, Suryanto (2007) also has executed an inquiry which examined the correlative aspect between language learners mastery in understanding past tense and their ability in expressing past activities in writing.The results pointed out that there was a substantial correlation (r xy =0.81) between the mastery of past tense and the competence in showing up past activities in the written forms.Other research finding which was done by Adhiyatma, et, al (2015) revealed that t-observed (6.748) was higher than t-table (2.712).The result of this study was classified essential.In short, students' mastery of grammar positively correlated to their writing ability.
There is a significant correlation means the scores within a certain range on particular variable are associated with scores within a particular range on the other variable (Gay and Airasian: 2000:321).It indicates that the students who have high score in structure knowledge tend to have high score in composition profiency, menwhile those who have low ability in structure knowledge tend to have low profiency in composition.Chin (2000) states that a lot of studies wonderfully recommend that the most useful method of backing up students to grade up their understanding of grammar in writing is to use students' writing as the basis for discussing grammatical concepts.Researchers believe that it is more effective to teach punctuation, sentence variety, and usage in the context of writing than to approach the topic by teaching disintegrated skills.

CONCLUSIONS
This current study is a nonexperimental research which applies in correctional design.The primary its objective is to retest the theories and determine if the language learners' knowledge of structure has relation to their writing proficiency.The findings of this study have been discussed in the previous part which reveals that the language learners still gained difficulties in using appropriate tenses.It was proven from the ability level classification which was in the interval poor to average ability.Furthermore, their writing paragraphs performance also indicated in the same ability level.performance] has positive and substantial relationship.It can be implied that the language learners who have adequate knowledge of structure language rules tend to have adequate composition proficiency.
To conclude, in the current study, the structure knowledge hands over positive and essential contribution on the writing performance.
In teaching writing performance cannot be isolated from structure knowledge.Due to one factor that effects on the writing performance is the structure knowledge.
It recommends that integrating grammar instruction into the revising and editing process helps students make immediate applications, thus allowing them to see the relevance of grammar to their own writing.Moreover, to gain better writing proficiency, writers should not only learn much structure systems knowledge but also learn other components of writing as how to use appropriate mechanics, to make composition easily to comprehend, to organize writing well based on the kinds of composition, and to choose suitable diction and vocabulary. .

Table 3 . The Percentage of Students' Knowledge of Structure No.
The rest of them, 33 students (42.9%) were considered Poor.The mean score of students in mastering structure was 56.38, which was categorized in the interval Poor to average level.
J-SHMIC,Vol 3, No 2, August 2016Correlation Inquiry between Structure Knowledge and Paragraphs Writing PerformancePoor to average.

Table 7 . Model Summary
a. Predictors: (Constant), Writing Performance J-SHMIC, Vol 3, No 2, August 2016 Correlation Inquiry between Structure Knowledge and Paragraphs Writing Performance

Table 8 . Correlations Between Structure Knowledge and Paragraph Writing Performance
table of Product Moment Correlation it can be seen that the value of r table of degrees of freedom with the level of significant 5% (0.05) is 0.277.That means the value of r xy is bigger than the value of r table of Product Moment J-SHMIC,Vol 3, No 2, August 2016Correlation Inquiry between Structure Knowledge and Paragraphs Writing Performance from the r