<![CDATA[Junior high school students' mathematical problem-solving ability in Indonesia remains relatively low, partly because conventional, teacher-centered instruction rarely gives students the opportunity to construct their own understanding through real, contextual problems. This study aimed to examine the effect of the Problem-Based Learning (PBL) model on students' mathematical problem-solving ability compared to conventional learning. A quasi-experimental design with a nonequivalent control group was employed, involving 64 eighth-grade students from a public junior high school, divided into an experimental class that received PBL and a control class that received conventional learning. Data were collected using a validated mathematical problem-solving test administered as a pretest and posttest, then analyzed using an independent samples t-test after normality and homogeneity assumptions were satisfied. The results showed that the average posttest score of the experimental class (78.6) was significantly higher than that of the control class (61.4), with a t-value of 6.42 and a significance value of 0.000 (p < 0.05). The N-gain analysis further indicated that the increase in problem-solving ability in the experimental class fell into the moderate-to-high category, whereas the control class remained in the low-to-moderate category. These findings indicate that the implementation of Problem-Based Learning effectively improves junior high school students' mathematical problem-solving ability, particularly in the aspects of understanding the problem, devising a strategy, and evaluating the solution. Teachers are encouraged to consider PBL as an alternative learning model to develop students' higher-order thinking skills in mathematics.]]>
