Problem-based Learning (PBL) approach has been widely used in various disciplines since it is claimed to improve students' soft skills. However, empirical supports on the effect of PBL on problem solving skills have been lacking and anecdotal in nature. This study aimed to determine the effect of PBL approach on students' problem solving skills using a quasi-experimental non-equivalent group pretest--posttest design. Fifty management students from a premier Technical University in Malaysia were assigned to experimental and control groups. In the experimental group, students were given four problems to be solved and their solutions of the problems given were assessed in terms of their accuracy and quality. Students in the control group received conventional classroom instructional design. Results indicate that students in the experimental group have better problem solving skills (z: -4.220, p: 0.001 for accuracy and z: -2.594, p: 0.009 for quality) compared to those who were not exposed to the PBL approach. This finding substantiates the use of PBL as an effective instructional tool to improve students' problem solving abilities.