<![CDATA[Mathematical Communication Skills (MCS) play a vital role in helping students structure their thinking and express mathematical ideas clearly and logically. Unfortunately, this skill remains relatively low due to limited interaction during instruction and the dominance of teacher-centered learning models. One innovative approach believed to address this issue is Project-Based Learning (PjBL), which positions students as active agents in the learning process. Numerous studies have indicated that PjBL can enhance MCS; however, it remains unclear which educational level and group size are most optimal for its implementation. This study aims to analyze the overall effect of PjBL on students’ MCS, as well as its impact across educational levels and sample sizes. A meta-analysis was conducted on eight published articles selected using PRISMA guidelines and analyzed using the Comprehensive Meta-Analysis (CMA) software. Effect sizes were calculated using Hedge’s g to measure the magnitude of the difference between the experimental and control groups. The results revealed that PjBL had a significant and moderately (0,785) effective impact on MCS, with students taught through PjBL demonstrating an average of 78.38% higher MCS compared to those in conventional learning environments. The highest effectiveness was found at the junior high school level, which aligns with Piaget’s view that students at this stage are in the concrete operational stage and are more responsive to activity-based and collaborative learning. Furthermore, PjBL was more effective when applied in smaller group sizes, as it allowed for more intensive interaction and support. However, the limited number and diversity of studies analyzed indicate a need for further research with broader coverage and more in-depth moderator analysis to obtain a more comprehensive understanding.]]>
