DISERTASI dan TESIS Program Pascasarjana UM, 2016

Ukuran Huruf:  Kecil  Sedang  Besar

The Effect Of Planning on EFL Students’ Writing Performance Across Different Levels of Self-Efficacy

Lestari - Setyowati




Setyowati, Lestari. 2015. The Effect Of Planning on  EFL Students’ Writing 


Performance Across Different Levels of Self-Efficacy. Dissertation, English Language Education, Graduate Program, State University of Malang. Advisors: (1) Prof.  H. Moh. Adnan Latief, M.A., Ph.D, (II) Prof.  Dra. Hj. Utami Widiati, M.A., Ph.D., (III) Prof. A. Effendi Kadarisman, M.A, Ph.D.


Key words:  Planning, Writing Performance, Self-Efficacy


Aiming at investigating the effect of planning on EFL students’ writing performance across different levels of self-efficacy beliefs, this research was directed to answer three research questions. Firstly, do students who use pre-writing strategy  achieve better writing performance than those  who use rough drafting? Secondly, do high self-efficacy students who use  pre-writing strategy  have better writing performance than  high self-efficacy students who use rough drafting? And lastly, do low self-efficacy students who use  pre-writing activities have better writing performance than low self-efficacy students who use rough drafting? To answer the research questions, a quasi experimental, posttest only nonequivalent group-design was used.

This study investigated the writing performance of fifty sophomores majoring in the Teaching English as a Foreign Language at STKIP PGRI Pasuruan, Indonesia in the academic year 2014-2015. From the accessible population  of all students taking essay writing class in this institution, two intact classes were chosen, in which each consisted of 25 students. These two intact classes were taught by the same instructor for 8 weeks. Thus, the treatment took 10 weeks, excluding the try-outs, conducted from April 2015 to June 2015.

The data of this research consisted of scores obtained from administering a writing test to the participants in the posttest and general writing test prior the treatment. The writing tests given to the subjects of the study were the researcher’s made test which had been tried out and validated. There were two sets of scores gained from two different  rubrics used, the first one is Jacobs ESL Composition Profile to score the students’ writing, and Primary Test Scoring Rubric for argumentative essay which was used to score the students writing posttest. All students’ compositions were rated by two raters. The stastitical procedure used was   t-test for independent sample analysis by using IBM SPSS 20.

The result of the experiment shows that there is no difference between the students who were taught by prewriting strategies and those who were taught by rough drafting strategies at .05 significance level,  (U = 210, p = .144 > α =  .05). Although there is a  mean rank  difference between the pre-writing group (26.85) and the rough drafting group (21.27) the difference is so small that it is not significant at .05 significance level. The effect size of planning is small (.013) which informs that  only 13 percent of the population will be affected by planning.  There are  possibly five factors that  influence the results of the study, namely  1) the participants’ characteristics, 2) the time length for planning, 3) the size of the sample, 4) the time length of the treatment, and 5) the sensitivity of the  rubric used to score the posttest.

In relation to the hypothesis testing number 2 and 3, the researcher concludes that  there is no difference in the students’ writing performance between the high self-efficacy ones who were taught to plan by using rough drafting or prewriting, as well as the low self-efficacy ones who were taught by using these strategies. The answers to these questions are inferred from the main research question which shows no differential effect between the prewriting strategies and rough drafting in terms of the learners’ writing performance.

Therefore, the researcher suggests future researchers who are interested to replicate this study to improve the sample size so that the power and the effect size can be increased and the null hypothesis can be rejected at the predetermined significance level. The researcher also suggests that future researchers use random assignment and devote the whole semester time for data collection in order to find out the real effect of planning, as well as use a more sensitive scoring rubric to score the posttest.

Although the result of this research shows no statistical significance difference between pre-writing strategies and rough drafting strategies, some recomendations still can be made. Firstly, the writing teacher should introduce different types of planning activity for writing, model them and help the students to practice. When they are already familiar, the teacher can suggest that the students  select one of the strategies to be used for their future prewriting task. And secondly, for the writing test developers, there should be a specific instruction in the writing test that encourages the learners to plan their writing with specific time demand. There is a great chance that the provision of planning in the writing test will enable the students to produce a better quality of composition.