This study aimed to examine (1) the differences in mathematical representation ability between students taught using the Problem Based Learning model assisted by Power Point and those in the control class, and (2) the effectiveness of the PBL model on students’ mathematical representation ability. A quasi-experimental method with a nonequivalent control group design was used, involving two fifth-grade classes in Semarang. The experimental group received PBL assisted by Power Point over four meetings, while the control group received direct instruction. Data were collected through pretests and posttests and, then analyzed using Shapiro-Wilk Test, Levene’s Test, and the Mann-Whitney U Test. The results showed that students taught with the PBL model assisted by PowerPoint achieved higher mathematical representation abilities, particularly in verbal representation aspect. However, the effectiveness of the model was not statistically significant, as it was influenced by several factors such as limited instructional time, students’ readiness for group work, the complexity of higher-order thinking questions, and the lack of interactivity in the learning media. Despite these limitations, the model shows potential for further development, especially through improving media quality, more effective time management, and designing assessments that align with students' actual abilities.