<p>This study examined the task-specific self-efficacy and self-evaluation of students related to mathematical problem posing as compared to problem solving. Students were randomly assigned to either a Self-efficacy Group or a Self-evaluation Group in which they responded to problem-posing tasks and nonroutine problem-solving tasks as well as a self-efficacy or self-evaluation measure. There were significant differences between reported self-efficacy and self-evaluation. For problem solving, the Self-efficacy Group reported higher confidence about their ability to complete the task than the Self-evaluation Group reported retrospectively about their performance. In contrast, for problem posing, the Self-efficacy Group reported lower confidence than the Self-evaluation Group reported. In addition, the relationships between problem-solving performance and self-efficacy and self-evaluation were found to be linear, but this was not true for the relationships between problem-posing performance and self-efficacy and self-evaluation.</p>