A bias-neutral approach to major project assessment in mechanical engineering

All engineering students completing the final year of manufacturing and mechanical engineering at Swinburne University of Technology (SUT) must undertake a comprehensive final year project. The project may encompass many areas of design, analysis, research, development or management. Often the projects integrate a number of these aspects. It is often the assessment of the project, not the pursuit of it by the student that presents difficulties for the academic faculty. In the past, techniques of assessing the project have proved difficult, particularly where the faculty supervisor has control over the execution of the project and has some positive or negative bias towards either the student or their project. This bias is often introduced in the final assessment, making it difficult to compare and quantify the final results of a cohort of projects resulting from many different supervisors. In an attempt to make the assessment process bias-neutral a comprehensive set of criteria have been introduced; in all seven steps are involved in the assessment procedure, only one of which is dependent on the supervisors, thus reducing their influence on the final grade. The assessment is classified according to the quality of the work (two assessments), a formal verbal presentation (two additional assessments), a large poster display (two further independent assessments) and a project performance assessment by the students’ supervisor (the only direct influence by the supervisor). The results of this assessment procedure have produced a “bell curve” series of grades or results, whereas in previous years a skew towards one end of a bell curve was usually obtained. This has been shown to be bias-free with regards to high or low grades and was a reflection of students; grades in other subjects. Overall, there seems to be satisfaction by both faculty members and the student cohort in the final grades achieved.