Background and Objectives: The changes in the new world, due to social, cultural, and the development of information and communication technologies (ICTs), have led to a modification in the objectives and educational approaches to align with the changes occurring in the learners' environment. One of the new approaches that has received attention in teaching and learning to acquire 21st-century skills is peer assessment. This study examined the impact of web-based peer assessment on the self-directed learning skills of educational sciences students at Arak University. Materials and Methods: Researchers employed a quasi-experimental research method with a pre-test-post-test design and included a control group in the study. The target population consisted of all educational sciences students in the academic year 2022-2023 at Arak University (178 students), from which 32 participants were selected through the convenience sampling method and assigned randomly into two groups: experimental and control. To gather data for the study, the researchers utilized the Self-directed Learning Readiness Questionnaire, which comprised 40 items and three subscales: self-management, desire for learning, and self-control. The questionnaire was administered before and after the course. During the 10 sessions, web-based peer assessment was integrated into the training process in the experimental group. In this way, courses, students, and the instructor were defined in the Learning Management System (LMS) of Arak University. After each teaching session, the teacher added an activity (peer assessment) in the system, in such a way that the students did homework related to the subject of the lesson. The teaching approach in both the experimental and control groups involved the teacher following a lesson plan and allowing the learners to ask questions during the teaching process. However, there was a difference in how the two groups were assessed. In the experimental group, peer assessment was utilized, where students reviewed and assessed their classmates' assignments. On the other hand, the control group followed a conventional evaluation method. In the classroom, the teacher delivered lessons and assigned homework to the students. The students needed to complete the assignments and submit them to the teacher for assessment. The teacher reviewed the student’s homework in each session and provided feedback. Statistical analyses, including descriptive statistics (i.e., mean and standard deviation), as well as inferential statistics (i.e., Analysis of Covariance), were employed to analyze the data. Findings: The findings indicated that peer assessment training had a notable impact on self-directed learning skills and their sub-scales, including self-management, desire for learning, and self-control (p<0.5). Conclusions: Based on the findings of this research, utilizing peer assessment can serve as an effective approach to enhance students' self-directed learning abilities, a crucial skill for learners in the 21st century. However, the successful implementation of peer assessment requires certain conditions that teachers should consider. These include having an appropriate e-learning platform, the instructor's formulation of assessment criteria, and providing proper guidance to learners acting as evaluators. It is important to note that the implementation of peer assessment comes with its own set of challenges, which should be thoroughly explored in studies examining its different aspects.
