Performance assessment through problem-solving or problem-posing provides benefits in learning mathematics. This study aims to obtain empirical evidence about the effects of performance assessment and metacognition on senior high school students' mathematics achievement. A quasi-experimental method was employed to engage 163 students in four classes selected through cluster random sampling. In addition, data were collected via an achievement test and a metacognition scale and analyzed using a two-way analysis of variance. The study results indicated a statistically significant discrepancy in the mathematics scores of students who were given performance assessments utilizing problem-solving and problem posing. Depending on the students' levels of metacognition, performance assessment had varying effects on the students' mathematical achievement. Students with a high level of metacognition and performance assessments through problem-solving had more effective mathematics achievement than those with performance assessments through problem-posing. In contrast, in students with medium and low metacognition, the performance assessment through problem-solving and problem-posing did not differ significantly and were classified as having low scores. This study suggested that using performance assessment and considering the level of metacognition support further efforts to enhance students' mathematics achievement.