ریحانه شوقی

As AI-driven tools gain prominence in informal language assessment, high-stakes proficiency exams such as the International English Language Testing System (IELTS) continue to depend on trained human raters to ensure fairness, genre sensitivity, and construct validity. This study examines how raters’ linguistic backgrounds—specifically English L1 and Persian L1—affect holistic scoring across writing genres within the IELTS framework. Some experienced EFL teachers (11 English L1, 11 Persian L1), all trained in writing assessment and familiar with IELTS rating procedures, evaluated 150 argumentative and descriptive essays produced by advanced Iranian learners under time-controlled conditions. Scoring was based on the publicly available IELTS Task 2 band descriptors, ensuring standardization and alignment with institutional criteria set by Cambridge Assessment. Findings revealed that while both groups demonstrated overall scoring consistency, English L1 raters applied stricter standards to organization in argumentative writing, whereas Persian L1 raters were more sensitive to grammatical accuracy across both genres. The results underscore the critical role of human raters in detecting discourse-level features not fully captured by AI-based scoring systems. Implications are offered for rater training, assessment fairness, and genre-specific writing instruction in standardized testing contexts.

This study investigates the effects of summative, formative, and combined assessments on the development of listening and reading comprehension skills on advanced Iranian EFL learners. Data were collected from 40 advanced EFL learners, aged 19 to 35, from intact classes at a higher education institute, divided into three experimental groups, each receiving one of the assessment types (summative, formative, or a combination of both), along with a control group. Learners’ progress in listening and reading comprehension was measured over a four-month period using pre- and post-tests administered at the end of the term. Data analysis involved ANOVA, ANCOVA, and post hoc LSD tests. The findings of this longitudinal study revealed that while the combined application of formative and summative assessments had a significant positive impact on listening comprehension, neither assessment type alone contributed to improvement in this area. In contrast, the results showed that all forms of assessment—whether summative, formative, or combined—led to notable improvements in reading comprehension. These results highlight the differential effects of assessment types on listening and reading skills, suggesting implications for assessment practices in EFL contexts.

This study investigated the effectiveness of two different scopes of corrective feedback on enhancing the complexity, accuracy, and fluency of L2 written productions among 150 intermediate EFL learners participating in intact online courses. The corrective feedback scope includes highly focused and comprehensive feedback delivered via digital platforms in synchronous and asynchronous formats. Participants were divided into five groups: synchronous highly focused feedback, synchronous comprehensive feedback, asynchronous highly focused feedback, asynchronous comprehensive feedback, and a control group receiving corrective feedback via traditional methods, characterized by the indiscriminate identification and marking of all errors. Results across the feedback groups regarding their impact on complexity, accuracy, and fluency in EFL learners' written productions indicated a significant effect on accuracy, while complexity and fluency showed no significant differences based on feedback scope and delivery modality. The results of this study have several important implications for educators, material developers, and policymakers in the field of language education. For teachers, adopting synchronous feedback strategies could significantly enhance the accuracy and complexity of students' written work.