The Impact of Online Dynamic Assessment on EFL Learners’ TOEFL Reading Comprehension performance

Introduction Aim of the Study: Dynamic Assessment (DA), rooted in Vygotsky’s theory of development and specifically his notion of the Zone of Proximal Development (ZPD), reconceptualizes assessment as an integrated and inseparable component of instruction. Unlike traditional forms of testing that seek merely to measure what learners already know or can do independently, DA emphasizes future development, responsiveness to mediation, and the assessment of learning potential. This approach shifts the focus from a static measurement of ability to a developmental perspective that considers learners’ potentioal to improve when provided with appropriate forms of mediation, scaffolding, or hints. Computerized Dynamic Assessment (CDA) represents a technological extension of this theoretical orientation, allowing for the integration of mediation into digital assessment environments. CDA is especially relevant in contexts such as TOEFL reading comprehension, where the ability to process text, employ strategies, and overcome comprehension obstacles can be directly supported through graduated prompts. As online learning environments expand and computerized testing becomes increasingly prevalent, CDA offers promising opportunities for delivering individualized mediation, capturing detailed learner interaction data, and creating scalable models of DA for large populations. Although DA has been investigated in various domains of language learning, its application to TOEFL reading comprehension—particularly in online and computerized formats—remains relatively underexplored. Since TOEFL reading items rely heavily on cognitive reasoning, inferencing, and text-processing skills, they provide a rich domain for examining how learners respond to mediation and how their ZPD can be engaged to promote deeper comprehension. This study, therefore, seeks to address this gap by applying CDA to TOEFL reading and examining its effects on Iranian EFL learners’ reading development as well as their learning potential. Therefore, the present study had two major aims: To investigate the effect of Online Computerized Dynamic Assessment (CDA) on the development of EFL learners’ TOEFL reading comprehension. This involved examining whether exposure to mediated, hint-driven computerized assessment would lead to statistically significant improvement in learners’ reading abilities. Then, to determine the extent to which CDA can reveal learners’ learning potential, using Kozulin and Garb’s learning potential formula. This aimed to show whether CDA could provide insights beyond static test scores by showing how much learners benefit from mediation.
Methodology: This study employed a quantitative, quasi-experimental research design. The participants consisted of 90 Iranian EFL learners selected through purposive sampling from a range of language institutes across Tehran. Learners represented intermediate to upper-intermediate proficiency levels based on institute placement tests. Only those who met predetermined proficiency criteria and demonstrated sufficient familiarity with computer-based testing environments were included.
Materials: A shortened TOEFL reading comprehension test was developed for the study. This test included five common TOEFL reading item types, such as: factual information questions, inference questions, vocabulary-in-context, reference questions, and main idea or summary questions. The test was adapted specifically for CDA by integrating graduated hints, which served as scaffolding designed to guide learners through the comprehension processes. These hints ranged from implicit to more explicit guidance, following a continuum of mediation similar to that recommended in DA literature.
Computerized Dynamic Assessment Software: A specialized CDA software package was designed for the study. The software delivered reading texts and questions, allowed learners to request or receive hints, recorded mediation use, and calculated post-mediation performance. This digital environment ensured standardization of mediation, immediate feedback, and precise tracking of learner responses.
Procedure: The study employed a pre-test and post-test design, with CDA integrated as the main intervention. Participants first completed the initial version of the test under dynamic conditions. During the test, learners could receive hints adapted to item difficulty and learner need. After the intervention period—where learners engaged with the computerized mediated feedback—they took a post-test. Both tests provided data for statistical analysis.
Data Analysis: Data were analyzed using the Paired-Samples t-test to determine whether the change in reading scores from pretest to posttest was statistically significant. Additionally, Kozulin and Garb’s (2002) Learning Potential Score (LPS) was calculated to measure learners’ achievement under mediation and to evaluate each learner’s responsiveness to DA.
Findings: The findings revealed several important outcomes such as Significant Improvement in Reading Comprehension: The Paired-Samples t-test indicated a statistically significant difference between learners’ pre-test and post-test reading comprehension scores. The increase in scores indicated that the CDA intervention had a positive effect on learners’ reading ability, supporting the pedagogical value of mediated assessment. Evidence of Learners’ Learning Potential applying Kozulin and Garb’s formula, the study found that the CDA approach was effective in revealing differences in learners’ learning potential. Learners who initially struggled demonstrated substantial gains after receiving mediation, while higher-proficiency learners showed more subtle but still measurable improvement. Mediation Patterns Revealed Strategic Weaknesses: Analysis of mediation use showed which item types triggered more hint requests, highlighting learners’ strategic gaps. Items requiring inferencing and summary comprehension drew the most mediation, suggesting areas where instruction might be strengthened. Overall, the data strongly suggested that CDA is capable not only of improving reading performance but also of identifying developmental trajectories, offering teachers deeper insights into learners’ cognitive processes.
Discussion and Conclusion: This study demonstrates that Online Computerized Dynamic Assessment can significantly enhance TOEFL reading comprehension among Iranian EFL learners. More importantly, CDA provides a richer, more multidimensional assessment of ability by integrating teaching and testing and by revealing learners’ responsiveness to mediation—something static tests cannot capture. The findings supported the incorporation of dynamic assessment into mainstream language testing practices, especially in technologically mediated contexts where individualized support is feasible. CDA offers a promising alternative to traditional assessment by focusing on future learning, not just current performance, and by making learners’ ZPD visible. Consequently, educators and test developers are encouraged to adopt CDA-based approaches to obtain a more complete and pedagogically meaningful picture of language proficiency.