The paradigm shift from testing the outcome to assessing the learning of process shines a light on the alternative assessment approaches, among which portfolio-assessment has sparked researchers’ interest in writing instruction. This study aimed at investigating the effect of portfolio-assessment on Iranian EFL students’ writing achievement through the process-centered approach to writing. To this end, fifty-three sophomores, studying English translation at Islamic Azad University, Tehran Science and Research Branch, were chosen as the participants of this study. The researchers randomly divided them into two groups--an experimental and control. The experimental group received an instruction based on the four stages of the writing process--brainstorming, outlining, drafting, and editing--and underwent the portfolio-assessment. For ten weeks, the participants of the experimental group practiced reflecting on their writing through formative self-check, peer-review, and teacher-feedback on each of the stages written as homework assignments. The revised paragraphs were regarded as the final portfolio. The control group, however, received a product-based writing instruction, to which portfolio-assessment, individualized-feedback, and reflection did not adhere. The results of the quantitative data analysis showed that the students in the experimental group performed better than the students in the control group in their writing achievement. The qualitative findings also revealed the students’ positive perception toward the portfolio-assessment on the writing process. The results suggest some pedagogical implications for EFL writing instruction and assessment.
