THE EFFECTIVENESS OF JOINT CONSTRUCTION AND INDIVIDUAL CONSTRUCTION ON THE STUDENT WRITING ABILITY AT ENGLISH DEPARTMENT OF UNIVERSITY OF ISLAM MALANG

THE EFFECTIVENESS OF JOINT CONSTRUCTION AND INDIVIDUAL CONSTRUCTION ON THE STUDENT WRITING ABILITY AT ENGLISH DEPARTMENT OF UNIVERSITY OF ISLAM MALANG

Danafi
English Education, Faculty Teacher Training and Education,
University Islam of Malang
Jl. MT. Haryono 193 Malang, 65144

Abstract: This article aims to know the effectiveness of join construction and individual construction on student writing ability. Implementation of this technique was to know the performance of students writing before and after giving teaching technique of join and individual construction. They include mechanical, vocabulary, organization, content, and language use. The design of the study was quasi-experimental, non-randomized, pretest-posttest control group design. The subject of the research was student semester 3 of UNISMA. It consists of 37 students. There were two class in this research, control class and experimental class. The control class was taught by individual construction while the experimental class was taught by join construction. The instrument of the research study was test. The stages in the research covered to test before treatment (Pre-Test), treatment, and test after test (Post-Test). To analyze the score of student used statistics. The computation score of the student used ANCOVA computation with the significance level of 0.013. The research found that the mean of experimental class was 78.015 while the control class was 75.445. It means that the students taught by using join construction in writing are better than students taught by using individual construction.
Key Words: join construction, individual construction, and writing ability
INTRODUCTION
Learning English need to pay attention in some skills which build English ability of the students well. One of them is writing ability. This skill is categorized as a productive skill (Harmer, 2004 p. 270-275). Writing is categorized as productive skill because a writer generates an idea and translates it into a readable text (Richards and Renandya, 2002, p. 303).
To master in the writing language, a writer is not only thinking about the grammatical rules which is discussed, but also how the writer can communicate their thoughts in the writing forms (Laksmi, 2006). Laksmi’s statement means that the understanding in the written language is important owned by a writer because writing skill also includes in the communicating skill.
According to Lopes (2009), the writing the students could not be afraid to make or avoid the mistake. Concentrating in the contents of writing is more important in order to the students write the text well and clearly. By guiding teacher in the teaching process is also needed by students. Lopes also said that teacher should help the students develop a different attitude toward writing by encouraging them to concentrate on thinking of contents and on expressing their thought clearly, rather than concentrating on avoiding mistakes.
In the teaching writing, in order to several of Indonesian students write the written language well, the teacher should give the approaches that are needed for students because without using the appropriate approaches the students often get problem in the getting ideas, choosing the right words, and organizing the details in their written (Laksmi, 2006). Therefore, the teacher should have approaches which have several stages. As Flower and Hayes (1980 in Beard, 1984) suggest that the stages include prewriting, drafting, revising, editing, and publishing. Also in Britton (1994) in Tompkins (1970) suggests that the stages of writing cover conception, incubation, and production.
Although the teaching writing aims to make students can communicate via written language, however, in the reality of teaching writing is more dominant in the grammatical rules and vocabulary. It can be seen in the survey of Alwasilah (2000). Alwasilah (2000) reports that in Indonesian teaching writing only dominant in the grammatical and vocabulary issue and less in the content of the text. Thus, the students problematize to construct the text clearly.
To make the students understand in the constructing writing, the researcher conducts the study aims to know the effectiveness of constructing the writing ability which uses two technique, those are join construction and individual construction in English department of University of Islam Malang. Join construction includes in the collaborative writing because the students have a small group or in pair then they have to discuss the topic and construct the topic into good text (Dobao, 2012).
In the join construction sure that in the process of constructing text will be there any discussing about the language and correcting themselves or others about text (Swain  & Lapkin, 1998, p. 326). The students can elaborate correct the grammatical into the right written. Meanwhile the individual construction is letting the students construct the text their own self.
According to Department of Education and Early Childhood Development in State of Victoria (2008), “join construction supports the students to understand the purposes and audience impact on language choices.” It is clear that the join construction help the students to know their goal in their writing task and make the reader understand about their language terms.
As in Storch (2005), “the use of small group/pair works in writing classes seems quite limited. I tends to be limited to the beginning stages (brainstorming), or more commonly, to the last stages of writing—the peer review stage.” It indicates that the activity in the small group/pairs has limited time in the beginning, brainstorming or final section or in the review step. It is caused the activity only focus on the process of the writing task. In the Dobao (2008) also said that, “both studies pairs were assigned more time to complete the task than individual learners.” it can be drawn the conclusion that the limitation of join construction spent more time to construct the writing tasks. So, the teacher less involve the activity of writing process.
Dobao (2012) found that two studies both of pairs and individual written task are same in terms of fluency and complexity. However, in pairs or join construction have more accurate in the written ability than individually. Eisenhower and Ettinger (2009) have also found there are many advantages in the collaborative learning for the student and also the teacher. Working groups emulate workplace models in which participants solve problems by committee. Automatically, the opinions of two experts have been clear to explain that both join construction and individual construction is suggested to implement in the university students.
 Based on the researcher’s experience in University Islam of Malang found that there were some instructors differences in the constructing written, those are organizing the written, grammatical and structural the written, and improvement the content of the written. There was an instructor always focus on the organizing the written and for the grammatical and structural of the written that instructor wasn’t too discussed. 
Therefore, to know the effect of the join construction and individual construction the researcher conducts the study that entitled ‘The Effectiveness of Join Construction and Individual Construction on Writing Ability of The Students First Semester at The English Department of University of Islam Malang.

METHODE
This study is experimental research. Experimental research is a scientific investigation in which an investigator manipulates and controls one or more independent variables and observes the dependent variable or variables for variation concomitant to the manipulation of the independent variables (Ary, Donald, et al, 1979). This research is quasi-experimental, non-randomized, pretest-posttest control group design because condition of the classes cannot be reorganized or disrupted to conduct research study. Therefore, researcher uses the classes that have already been organized into classes (Ary, Jacobs, & Razavieh, 2002).
Based on Ary, Jacobs, & Razaviech (1979), stated that all members of any well-defined class of people are called population. The populations that will be observed in this research are UNISMA’s students in academic year 2017-2018, exactly in English Department. There are 5 classes of writing class of 3rd semester with the population about 125 students and each of class consists of 20 students. The researcher chooses two classes randomly with lottery and the researcher get class C and D. In the class 3C consists of about 15 students and in the class 3D there are 22 students.
The instrument which is used in this study research is a test. Based on the Ary, Jacobs, & Razaviech (1979) note that a test is a set of stimuli presented to an individual in order to elicit responses on the basis of which a numerical score can be assigned. The students are tested in a prompts composition test that is given before treatment (pre-test). In the pre-test, the students in the two classes do same tasks about procedure text but difference in time. After conducting pre-test in two classes, researcher does the treatments of join and individual construction but treatment is conducted only in the class D which as experimental class. Meanwhile in the class C which as control class, there is no treatment from researcher. After the treatment have finished, researcher conducts test after treatment (post test). All of students in two classes are tested. The allotted time in this test is 90 minutes for the join construction group and also 90 minutes for the individuals. The students are asked to construct the procedure text.
To take the data of the research, the researcher conducts pre-testing to ensure the test both of groups is same. As Rahayu L. N. (2014) cited in the Lodio et al. (2010) notes that “a pre-test measures whether the experimental and control groups are starting out equal”. Pre-testing is checking the ability of the students both of two classes of join or individual construction. In this research, both of class C as individual construction and D as join construction are given 60 minutes to construct the written. The students of class C and D construct the written individually. The type of text in class join and individual constructions are same, that is making procedure text. In the pre-test both of class C and D, the researcher only once in conduct pre-test.
In this step is very important. The researcher gives two treatments which are different. First, in the class D as join construction is given to the experimental class and the second, the individual construction is given to class C as control group. The meeting in this research is done in once meeting, only in the class join construction.
The final step to take the data collection is Post-testing. In the post-test process, both of classes join and individual constructions are taken the last scores. The text that will be made by students is procedure text. The time allotment which is provided to the students is 60 minutes each of class.
FINDINGS
In the conducting ANCOVA, the researcher has to know the assumption of ANCOVA first. There are three assumptions before conducting ANCOVA those are linearity, homogeneity of variances, and homogeneity of regression slopes.
In that figure is distribution of the score of the students both of joins construction class and individual construction class. Based on that figure looks a linear (straight line) relationship or not curvilinear because many of “dots” is near the straight line. In order word linearity is met or not violated.
Table 1 Levene's Test of Equality of Error Variancesa
Dependent Variable:   Posttest 
F
df1
df2
Sig.
2.581
1
35
.117
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Design: Intercept + Pretest + Class

Table 1 shows the Levene's Test of Equality of Error. The levene’s test is the method to check whether equality of variance is violated or not. The assumption of quality of variance is if significance value more than 0.05 the variance is homogenous, and if smaller than 0.05 is not homogenous or violated, (Pallant, 2000). Based on table 4.1 the researcher assumes that the value of significance is Sig. 0.117, it indicates that the assumption of quality of variance is met or not violated.

Table 2 Tests of Between-Subjects Effects
Dependent Variable:   Posttest 

Source
Type III Sum of Squares
Df
Mean Square
F
Sig.

Corrected Model
364.392a
3
121.464
14.545
.000

Intercept
57.880
1
57.880
6.931
.013

Class
.724
1
.724
.087
.770

Pretest
219.545
1
219.545
26.290
.000

Class * Pretest
.226
1
.226
.027
.870

Error
275.581
33
8.351



Total
219859.000
37




Corrected Total
639.973
36




a. R Squared = .569 (Adjusted R Squared = .530)


Table 2 shows that the interaction term (Class*pretest) is more than 0.05. The significance value in that table show 0.870, it is able to be concluded that interaction between covariate variable and dependents variable or treatment is statically significant. After knowing the assumption of significance value, the ANCOVA computation will explore the differences between Independent variables.
Homogeneity of regression slopes is testing the interaction between covariate variable and independent variable to predict the dependent variable or treatment. In the interaction between covariate variable and independent variable to predict treatment, there is assumption that if significance level is smaller than 0.05 or equal to 0.05, it indicates that the significant or violated. Thus, ANCOVA computation cannot be proceeded. Accordance to the Table 4.2 the significance level of Class*Pretest is 0.870 or in other word is more than 0.05. It indicates that the significance level is not violated or computation of ANCOVA is able to be proceeded.
In the table below provides the data analysis of significance value. The significance value in the data analysis is 0.13. It means that the significance level is less than 0.05. Thus, the experimental class and control class is significantly different and the null hypothesis saying that the teaching writing by using technique join construction and individual construction are equal should not be accepted.
Table 3 One way Analysis of Covariance
Dependent Variable:   Posttest 
Source
Type III Sum of Squares
Df
Mean Square
F
Sig.
Partial Eta Squared
Corrected Model
364.166a
2
182.083
22.446
.000
.569
Intercept
65.790
1
65.790
8.110
.007
.193
Pretest
237.626
1
237.626
29.293
.000
.463
Class
55.929
1
55.929
6.895
.013
.169
Error
275.807
34
8.112



Total
219859.000
37




Corrected Total
639.973
36




a. R Squared = .569 (Adjusted R Squared = .544)

Table 4 Estimated Marginal Means

Dependent Variable:   Posttest 
Class
Mean
Std. Error
95% Confidence Interval
Lower Bound
Upper Bound
join construction
78.015a
.614
76.768
79.262
individual construction
75.445a
.747
73.926
76.963
a. Covariates appearing in the model are evaluated at the following values: Pretest = 74.22.

In the table 4 reported differences adjusted means for two classes. For the Estimated Marginal Means of join construction or experimental group is 78.015 and individual construction or control group is 75.445. According to the Estimated Marginal Means above, students who get join construction have better than individual construction in writing ability is accepted. As conclusion, technique of join construction can enhance writing ability of students.
Discussion
In this past, the researcher discussed the information about research findings. In the finding of research, the research divided into two points of view; in practical and statistical view. In the statistical view, researcher found the significant effect of students who are taught by using join construction than individual construction. It was proven by scores of mean join construction as experimental class was 78.015 and individual construction as control group was 75.445. it means that the students who taught by join construction have better in writing ability than individual construction.
In the join construction, the researcher believes that the technique join construction is appropriate for constructing writing, in this case constructing procedure text. By using that approach, the students is more clear in the arranging the component of the written, such as mechanical, language use, vocabulary, organization, and content because the student who do not understand about material will be more understand when they are taught by join construction approaches. According to Lara (2013) wrote that writing is not an isolated activity, but collaborative one that also promotes self-reflection about one’s own errors through listening to different opinions.
Many of join construction begin to decide the material or genre to be taught. The researcher gave procedure text that become material to be taught. Asking the question about definition is the first activities to recall or build knowledge of the student about the procedure text. They could mention the definition of the text by differences explanation. Not just the definition about procedure text, general structure of the text, and language use was also discussed in the first activities of the class.
In the join construction there are two ways to be applied; those are the teacher can write together with the student, while the teacher acts as writer in front of classroom and the student may write in a group (Emilia, 2010).  In this research the researcher applies to first ways; the researcher acts as writer. While the treatment use join construction or collaborative learning, the researcher found interaction among the students such as sharing opinions about definition, language use, and general structure, etc. that interaction can impact the learners’ cognitive processes (such as sharing the opinion about structure text with other students etc). Thus, the finding of the research showed that join construction have impact to the cognitive processes of the student.
In the last activities the researcher asks the student to summarize about discussion about procedure text enthusiastically. When the student could not mention the structure of the text, researcher gives the clues to ease them mention it.
This strategy will be more effective if the participant or the student can participate in the teaching learning process. As the facilitator, the teacher should be able to create the attracting atmosphere so that the student comfort with the learning writing.
REFERENCE

Alwasilah, A. Chaedar. 2006. From Local to Global: Reinventing Local Literature Through English Writing Classes. Universitas Pendidikan Indonesia. Bandung

Beard, R. 1984. Children s Writing in the Primary School. London: Hodder & Stoughton.

Department of Education and Early Childhood   Development. 2008.  State of  Victoria

Dobao, Ana Fernandez. 2012. Collaborative  Writing  Tasks  In  The  L2 Classroom: Comparing  Group, Pair, And Individual Work. University of Washington. United States

Emi, Emilia. 2010. Teaching Writing. Rizqi Press. Bandung

Harmer, Jeremy. 2004. How to Teach Writing. Pearson Education ESL. England

Laksmi, E. Dewanti. 2006. Scaffolding Students Writing in Efl Class: Implementing Process Approach. Universitas Negeri Malang. Malang

Rahayu, N. R. Lutin, 2014. The Effect of Pair and Individual Writing Tasks on the Quality of Writing by Efl Learners of STKIP PGRI Blitar. Unpublished. Islamic University of Malang.

Richard, J. C. and Renandya, W. A. (2002. Methodology in Language Teaching Cambridge: Cambridge University Press

Storch,   N.  (2005). Collaborative writing: Product, process and students’ reflections. Journal of Second Language Writing, 14, 153–173.

Swain,  M., & Lapkin, S. (1998).  Interaction   and   second   language   learning: Two  adolescent French  immersion  students  working  together.  The Modern Language  Journal, 82, 320–337.

Tompkins, G. E. 1994. Teaching Writing: Balancing Process and Product. New York: Macmillan Publishing Company. 


                                                                                           Malang, 10 February 2018
                                                                                                             Advisor 1

                                                                                                 M. Yunus, S.Pd, M.Pd

                                                                                                 NPP: 209.02.000.02

Comments