IELTS Writing

Learning-oriented assessment (LOA) seeks to support instruction by providing information about students ’ current learning conditions and their ultimate performance. This study scrutinizes the application of LOA in English as a Foreign Language (EFL) writing classes within Iranian language institutes. In addition, the current study examined the participants’ writing traits using the researchers-developed writing traits questionnaire during the implementation of LOA. Sixty upper-intermediate students (male & female) in two intact classes at a language institute in Tabriz, Iran, were taken as the participants. One of these classes was randomly picked as the control and the other one as the experimental group. During the treatment, learners of the experimental group carried out the relevant writing tasks based on the principles of LOA. On the other hand, the control group adopted the teacher-centered approach. 120 IELTS writing task 2 samples were collected and scored by using a scoring rubric (Wang & Liao, 2008) before and after the LOA treatment. To realize the differences in the mean scores of students in the control group and the experimental group, an independent sample t-test was used. The findings showed that the students of the experimental group benefited vastly from multiple drafting, self, and peer-evaluation, sharing learning goals with teachers and each other, and using feedforward techniques instead of feedback. The results might provide both theory and pedagogy implications for syllabus designers and language teachers with valuable guidelines on improving writing assessment in IELTS writing courses.

With the incremental integration of technology in writing assessment, technology-generated feedback has found its way to take further steps toward replacing human correcting and rating. Yet, further investigation is deemed necessary regarding its potential use either as a supplement to or replacement for human feedback. This embedded mixed-method research aims to investigate three groups of Iranian intermediate IELTS applicants who received automated, teacher, and blended (automated + teacher) feedback modes on different aspects of writing when practicing for the writing skill of IELTS. Furthermore, a structured written interview was conducted to explore learners’ perception (attitude, clarity, preference) of the mode of feedback they received. Findings revealed that students who received teacher-only and blended feedback performed better in writing. Also, the blended feedback group outperformed the others regarding task response, the teacher feedback group in cohesion and coherence, and the automated feedback group in lexical resource. The analysis of the interviews revealed that learners had high opinion regarding the clarity of all feedback modes and learners’ attitude about feedback modes were positive. However, they highly preferred the blended one. The findings suggest new ideas that can facilitate learning and assessing writing and recommend that teachers provide comprehensive, accurate, and continuous feedback as a means of formative assessment.

IELTS preparation courses have gained significant popularity in Iran in the past decade. Although teachers in such an exam-oriented context have started to use formative assessment to improve their writing instruction, their knowledge and beliefs about assessment for learning are still a myth. This mixed-methods study investigated Iranian IELTS teachers’ beliefs and knowledge about the four main aspects of formative assessment of writing in preparation courses for IELTS Writing task 2. Thirty-nine IELTS teachers provided answers to a 23-item questionnaire focusing on four areas: feedback, self-assessment, peer-assessment, and using assessment results for day-to-day classes, to illustrate how frequently they use such techniques. In the next stage, six of the teachers sat for an interview to provide their reasons for using/not using such techniques. The results showed that the teachers have good feedback literacy and make use of some self-assessment techniques such as rubric orientation while they did not value or know enough about how they can involve their students in their own learning process. The teachers seemed to overestimate their role in their students’ learning process while considering the students as somewhat incapable of monitoring their own progress and achievement, which is a crucial aspect of formative assessment. These findings have implications for teacher professional development and further formative assessment programs to be conducted in Iran.  

This study compares the effectiveness of artificial intelligence (AI)-generated feedback (specifically using ChatGPT) and human feedback in improving the writing performance of B2-level IELTS candidates. A mixed-methods approach was employed, combining quantitative and qualitative data to evaluate the impact of both types of feedback. A total of 100 B2-level English learners from various cities of Iran were selected through convenience sampling and then randomly assigned to two groups. Pre- and post-test assessments, using mock IELTS exams, provided quantitative data, while qualitative data were gathered through participant questionnaires. These findings reinforce the superiority of human feedback in enhancing higher-order writing skills; however, this study extends prior research by examining the effectiveness of AI-generated and human feedback within the specific context of Iranian IELTS candidates. Given the traditionally grammar-focused writing instruction in Iran, the results highlight the need for integrated feedback approaches that address coherence, cohesion, and argument development. The study also offers practical implications for IELTS preparation programs in non-native English-speaking countries, advocating for a structured combination of AI and human feedback to maximize writing proficiency. Learners valued the personalized guidance of human feedback, which fostered motivation and engagement. Conversely, AI-generated feedback excelled in correcting surface-level errors, such as grammar and vocabulary, offering immediacy and consistency. Nonetheless, AI's lack of interpretive depth limited its ability to address complex writing issues like argumentative development and cohesive flow. This study underscores the potential benefits of integrating AI and human feedback to enhance writing instruction for standardized assessments. Future research should investigate strategies to refine AI tools for better support of advanced writing skills.