To become effective learners, students need to develop good evaluative judgment skills. Unfortunately, numerous studies find that self-evaluation estimates provided by undergraduates often differ significantly from the marks awarded by the tutor. This suggests that students either have a rather poor grasp of the assessment criteria, or they find it difficult to apply standards to their own work because of their emotional investment. They may demonstrate a better understanding of standards when asked to judge the work of their peers. We use data from a cohort of 2nd year undergraduates to compare the ability of students to accurately self- and peer-evaluate an assessed essay. We find that peer evaluation is more accurate, on average, than self-evaluation but shows greater dispersion, and there is limited evidence that misconceptions about standards are consistent across self- and peer-evaluation.