Student variables and their influence on the reliability of peer assessment : assessing maritime engineering graduate attributes in problem based learning

The work detailed in this thesis investigates the development and assessment of graduate attributes in problem-based learning (PBL). The PBL experience provides students with realistic and challenging tasks, promoting interactive learning while developing graduate attributes. The study investigated how students perceive peer contribution and interaction, which influences the way they respond in using peer assessment to evaluate contribution, participation and the development of graduate attributes in-group work. The study involved the development of a new approach to the delivery and assessment of graduate attributes in a PBL experience. It brings into question as to whether peer assessment is a reliable method in assessing the attainment of graduate attributes in the problem-based learning environment. This study introduces a more comprehensive and theoretically informed understanding of what affects student variables (Psychological Safety, Value Diversity, Interdependence, and Trust) play in the validity and reliability of peer assessment in a team based active learning experience. The results show there is a direct link to the interpersonal influence of student variables on students performing in this environment. These student variables then have an impact on the reliability of peer assessment. A Formative PBL assessment model was developed that introduced a team contract, mentor and student reflective feedback process. The latter process enabled the student to provide instructional opportunities that related to their individual result of peer assessment. The outcome encouraged the student to develop purposeful personal and professional understanding of their attainment of graduate attributes. Furthermore, it allowed the students collectively to develop a positive interdependence approach to team development and the project, and their interaction with peer assessment. The development of positive interdependence showed a higher correlation with observed attainment of graduate attributes and a high level of psychological safety in assessing their peers. Students offered insights into the relationship between psychological safety and motivation for learning when taking on peer assessment responsibilities. The developed model is suitable for future work involving the use of peer assessment in PBL and the assessment of graduate attributes. This study provides new insight into the impact of student variables on the reliability of peer assessment. An important finding in this study is that the Formative PBL-Assessment model 2 presented in this study has clearly identified key elements that influence how students approach and participate in PBL and peer assessment of graduate attributes. The adoption of the Formative PBL-Assessment model 2 enables students to develop a positive interdependence approach when in a PBL environment. This approach has shown to allow students to develop a deeper approach to interaction in, and understanding of, their learning, attainment of graduate attributes and the outcome of peer assessment. This study contributes to the understanding of the meaning and consequences of implementing peer assessment into final year design units within Bachelor of Engineering courses and has broader implications in other academic disciplines.