The Effect of Using Portfolio-based Writing Assessment on Language Learning : The Case of Young Iranian EFL Learners

This study investigated the effectiveness of portfolio-based writing assessment in EFL situations. Participants were 40 pre-intermediate young Iranian English learners. They were randomly divided into experimental and control groups of 20 each. The experimental group wrote on five pre-established topics from their coursebook. Their writings were checked in terms of ideas, organization, voice, word choice, sentence fluency, and conventions of writing by two raters. They were given another opportunity to revise their writings to be corrected again. In contrast, the control group wrote only once and their writings were corrected only by their own teacher. The participants were also required to complete a questionnaire to assess their reflection and self-assessment. Results of the study indicate that portfolio-based writing assessment has a positive effect on language learning and writing ability. It also shows that it helps students’ self-assessment and almost all students are satisfied with this method of assessment.


Introduction
The use of portfolio-based assessment is now well established as a valuable assessment tool (Barrett, 2000;Biggs & Tang, 1997;Cooper, 1997;Education Department of Western Australia, 2000a, 2000b).Portfolio-based assessment is beneficial pedagogically because the format can encompass evidence from a wide variety of sources; it can help educators overcome many assessment difficulties, especially in relation to equity and moderation (Cooper, 1999;Cooper & Love, 2000); it provides a 'richer picture' of the student (Barrett, 2000b); and portfolio-building actively involves students in the learning process (Bowie, Taylor, Zimitat, & Young, 2000).Hedge (2000) maintains that portfolio assessment is seen as a more comprehensive portrait of students' writing ability than one essay composed under restricted circumstances.Thus, evaluating portfolios instead of only one impromptu timed writing sample of students will put teachers in a better position to make informed judgments about students' writing ability.In addition to this summative function of portfolio assessment, it also enables teachers to provide ongoing feedback that informs both teaching and learning (Dysthe, 2008).This formative function of portfolio assessment is under-explored especially in the EFL context (Lam & Lee, 2010).The purpose of the current research is, therefore, to investigate whether language learning and writing ability of the students are influenced by the portfolio-based writing assessment.Additionally, students' perceptions about portfolio use are to be investigated in this study.

Background
The notion of portfolio-based assessment began to attract attention around the mid-1980s as a reaction against the psychometric climate prevailing at the time.In the United States in the 1980s, there was growing concern about declining educational standards.This atmosphere led to intense pressure to place more emphasis on testing as a means of raising standards, in accordance with the belief that the more students are tested, the more they will be motivated to improve efforts and performance.Elbow and Belandoff (1997), in looking back on that period, observed that "in retrospect, what was striking was urgent and growing pressure for assessment, assessment, assessment: test everything and everyone for a score; don't trust teachers" (pp. 22-23).A tendency in portfolio-based writing assessment emerges from the long history of writing assessment: from indirect multiple choice tests to direct timed impromptu essay tests to portfolio-based writing assessment (Hamp-Lyons, 2001/ 2002).
Since the 1980s, portfolio-based writing assessment has gained increasing popularity mainly for two reasons.One driving force is growing dissatisfaction with timed impromptu essay tests.The other is the development of writing instruction.Instructional approaches in ESL/EFL writing have in some respects parallel developments in English L1 composition.The traditional product-oriented model is out of date while process-oriented writing pedagogies are increasingly pervasive at least in American educational milieu (Hedgcock, 2005).Process approaches to ESL/EFL writing suggest that "it is unnatural for a learner to write a draft of composition and submit for a grade" (Cohen, 2001, p. 534).Therefore, a change of assessment paradigm is called for (Hamp-Lyons & Condon, 2000).The portfolio-based writing assessment is perhaps the standard-bearing alternative instrument in ESL/EFL composition assessment (Hedgcock, 2005;Weigle, 2002).
These days, portfolio-based writing assessment is considered to be a fulfilling substitute for the traditional essay test in English L 1 contexts.Even for ESL/EFL contexts the use of portfolios is claimed to be beneficial (Delett, Barnhardt & Kevorkian, 2001;Hamp-Lyons, 1995;Hamp-Lyons & Condon, 2000;Song & August, 2002).However, most research findings indicating the usefulness of portfolio-based writing assessment come from English L 1 contexts and little attention has been given to its application to ESL/EFL contexts (Hamp-Lyons, 2001;Hirvela & Pierson, 2000;Hirvela & Sweetland, 2005;Liu, 2003;Weigle, 2002).Further, research indicates that the Western society students are encouraged to decide on their own goals and take the responsibility of learning themselves (Liu, 2003) while there has been little research done on the use and value of portfolios in L 2 contexts, in general, and writing instruction, in particular.Thus, finding an effective way of writing assessment and its influence on language learning can be of great importance in language learning and testing.Since portfolio-based writing assessment focuses on the process of learning and evaluates students' progress overtime, it can be a more useful way of assessment.Thus, the present study set itself the goal of investigating the effect of portfolio-based writing assessment on students' language learning.Specifically, it aimed at investigating the following research questions: 1) Does portfolio-based writing assessment improve young Iranian EFL learners' language learning?2) Does portfolio-based writing assessment improve young Iranian EFL learners' writing ability?
3) What are students' perceptions about portfolio use?

Participants
The participants included 40 pre-intermediate male and female students selected from a larger group of 68 students who took the OPT, so that we could include in the study those who were homogeneous.They aged between 17and 25 studying English in a language learning center in Iran.Further, they were randomly divided into two groups of control and experimental of 20 students each (10 male and 10 female students).All the students in the experimental and control groups attended an English class twice a week.

Materials
The instruments used in the study were the OPT to guarantee the participants' homogeneity in terms of their English proficiency level, Top Notch series by Saslow and Ascher ( 2008), which is a new conversation series, writing portfolios (files and folders) produced by the participants in the experimental group during the term, and a questionnaire used for students self-assessment.It consisted of some questions about areas of writing to improve and areas of strength of students while writing.Students' overall impression about this kind of assessment and the effect of portfolio-based writing assessment on language skills and sub-skills were supposed to be investigated by this questionnaire.

Procedure
First, the OPT was administered to select 40 pre-intermediate students out of a larger group.They were randomly divided into experimental and control groups of 20 students each.All the students in the experimental and control groups attended an English class twice a week, that is, about 3:30 contact hours per week during a semester of study (22 sessions) with the same female teacher.They studied 5 units of Top Notch 2.
In order to find about whether the writing ability of the students would improve the students were asked to write on two specific topics selected from Oxford Placement Book once at the beginning of term and then at the end of the semester.To assess their writing ability Pearson Education (2007) rubric was used by two rates (See Appendix C).
During the term the students in the experimental group wrote on 5 pre-established topics corresponding to the topics of each unit of their coursebook.Their writings were checked based on ideas, organization, voice, word choice, sentence fluency, and conventions by the two raters who were qualified and quite familiar with the scoring rubrics.The scoring rubric used at this stage to correct their writings was the one adapted from Rog (2001, See appendix B).The students were then given another opportunity to revise and resubmit their assignments after self-assessment or peer assessment in accordance with the feedback they received.These assignments reached the raters after three careful drafts.They were classified into groups of strong, competent, developing and emerging.In contrast, the control group wrote only once on each topic and their writings were corrected by their own teacher.The students in the experimental group were also asked to complete the questionnaire (See Appendix A) used for students self-assessment at the end of the semester to find about their perceptions about the use of portfolio.

Results
To compare progress in language learning between the control and experimental groups at the end of the semester, the OPT was administered again.The t-test was used to compare the results.As can be seen in Table 1, the observed t (obs.t = 2.13) is high enough to reject the hypothesis ( p<0.05) that portfolio-based writing assessment does not lead to language improvement (crit.t = 0.684, df = 38, and p ≤ 0.05).In other words, there is a statistically significant difference between the experimental group and the control group, which means portfolio-based writing assessment has been effective.Thus, portfolio-based writing assessment has a positive effect on young Iranian EFL learners' language learning.Moreover, to compare the participants' growth in writing, as outlined above, they wrote on two specific topics selected from Oxford Placement Book at the beginning and the end of the term.The t-test was used to compare progress of both groups in writing.Table 2 illustrates a summary of the t-test for the experimental and control groups in the writing posttest.It indicates that the observed t (obs.t = 8.20) is high enough to conclude that the difference between the control and experimental groups is statistically significant (crit.t = 1.684, df = 38, and p ≤ 0.05).Thus, portfolio-based writing assessment has a positive effect on young Iranian EFL learners' writing ability.
In order to check the students' reflection and self-assessment, a questionnaire (see Appendix A) was prepared.The students in the experimental group were supposed to answer all the questions at the end of the semester.According to the results (See Table 3), voice (42.9%) and ideas (35.7% ) were rated as the two most important areas of writing needing improvement.The other areas of writing were judged to need less improvement.Also, as the results in Table 4 show, the participants agreed that organization (31.6%) and word choice (23.8%) were the two most important areas of strength in writing.
The students were also asked to rank the skills and sub-skills that they could improve better than the other ones by means of portfolio-based writing assessment.According to Table 5, 50% of the students believed that their grammar improved more than the other skills and sub-skills while reading and speaking improved less.
Last but not least, almost all the students were strongly satisfied with this method of writing assessment; however, %5 of the total was only satisfied to some extent and no one was dissatisfied by this method (see Table 6).

Discussion and conclusions
The results of quantitative data analysis showed that portfolio-based writing assessment had a positive effect on language learning.This study was conducted over a short period of time so it is not surprising that over a longer period of time more significant results may show up.The results of this study are consistent with those of Bowie, Taylor, Zimitat, and Young (2000) according to which portfolio-building actively involves students in the learning process.
It was also found that the students' abilities in writing were not significantly different in the pretest but after implementing portfolio-based writing assessment it was observed that the scores of the students in the experimental group were significantly higher than those of the students in the control group.In contrast to Herman, Gearhart, and Aschbacher (1996), who voiced their concern over the lack of research demonstrating conclusively that portfolios are more effective than other forms of writing assessment, this study could show that portfolios were more effective than the traditional method of writing assessment.
According to findings of this study, portfolio-based writing assessment is a kind of balanced assessment that focuses on all aspects of a student's writing such as fluency, content, conventions, syntax, and vocabulary in contrast to the other kinds of writing assessment through which teachers focus their attention primarily on surface features of a student's composition related to the mechanical aspects of writing or conventions.In this line, Moran (1982) observes that teachers traditionally have been more strongly influenced by length of paper, spelling, word usage, and appearance than by appropriateness of content or organization, which is in contrast with portfolio-based writing assessment that focuses on all aspects of writing.

Punctuation and Mechanics-5 points
There is a period after every sentence.Capital letters are used correctly.
The spelling is correct.
Commas are used correctly.Total -based writing assessment has a positive effect on which of the following skills and sub-skills?Rank the skills from the highest to the lowest.
The paragraph begins with a topic sentence that has both a topic and a controlling idea.The paragraph contains several specific and factual supporting sentences that explain or prove the topic sentence, including at least one example.The paragraph ends with an appropriate concluding sentence.Education, Inc Duplication for classroom use is permitted.

Table 1 .
Summary of the t-test for the experimental and control groups in posttest

Table 2 .
Summary of the t-test for the experimental and control groups in the posttest

Table 5 .
The effect of portfolio-based writing assessment on language skills and sub-skills