This study aims to test the effectiveness of the Teams Games Tournament (TGT) learning model on the mathematical communication skills of grade II elementary school students. The research method used is a true experimental design with a posttest-only control group, involving two classes at SDN 3 Podomoro (n = 51). The instrument consists of six descriptive questions, which were tested for validity and reliability (α = 0.708). The results of the analysis using the Mann–Whitney U test showed a significant difference between the experimental and control classes (p < 0.001), with average values of 67.6 and 44.0, respectively. The magnitude of the effect was calculated using Cliff's Delta (δ = 0.808), which indicated a significant impact. However, the test instrument was dominated by easy items, causing a ceiling effect in the experimental class (36% of students achieved the maximum score). It is recommended to revise the instrument with "moderate and difficult" questions and integrate interactive technology. This study is limited to a single school and level, so the generalizability of the results needs to be done carefully.
Copyrights © 2025