The study sought to establish the level of students' self-assessment skill--particularly "inexperienced" students--and to examine the relationship between self-assessment skill and learning style, student perceptions of academic locus of control and academic self-efficacy. Students were asked to evaluate and provide estimated marks for their own work, were which compared with tutors' actual marks. Students also completed measures of learning style, academic locus control and academic self-efficacy. Comparisons of student estimated and tutor marks indicated a good level of self-assessment skill in the majority of students. A significant minority of students did however fail to exhibit such skills. There was also some evidence of a tendency for students to underestimate their performance. While both strategic and deep approaches to learning were shown to be positively correlated with tutor mark, only surface approach was negatively correlated with students' estimated mark, suggesting that surface learners are inclined to provide lower evaluations of their own performance. Deep approach was also correlated with accuracy of student self-assessment skill, suggesting that deep learning is associated with self-assessment competency. No clear or convincing associations between self-assessment skill and perceptions of academic locus of control or academic self-efficacy were identified. Findings suggest that while self-assessment skill undoubtedly develops, becoming more effective during students' academic career, inexperienced students do have the capacity for self-evaluation and should therefore be included in self-assessment activities. In the light of findings related to learning style and the heterogeneous nature of student groups, student monitoring and skill development are proposed in order to allow the integration of self-assessment into the learning and assessment process. (Contains 1 table and 2 figures.)