Abstract:

A gradeless form of assessment is still a relatively rare assessment method across higher education, even with evidence suggesting that graded courses exacerbate student stress and degrade students’ intrinsic motivation to learn course material. To foster student success, in Fall 2022 the Software Engineering Technology (SFWRTECH) program at McMaster University implemented a Residency Model of education. Similar to a medical residency, instead of students receiving a numeric grade at the end of each course, students are assessed on a variety of skills and pass when they satisfactorily demonstrate these skills. Within each course, students undergo iterative and continuous learning through assignments, short-challenge projects, and a midterm and final portfolio. Students are also allowed and encouraged to reappear for assessments that did not meet the required benchmarks, based on the qualitative commentary from the teaching staff. Previous studies on the SFWRTECH Residency Model have found that students’ responses to gradeless learning was mixed with some elevation of stress due the stiffer benchmarks needed to attain a satisfactory rating in demonstrating a skill, but at the same time appreciating the opportunity to reappear for assessments and recognizing the value of demonstrating skills through portfolios. Teaching staff similarly recognized the importance of the portfolios and the resubmission/reappearance opportunities on assessments to foster student intrinsic learning. Here, we add to these studies by examining how learning outcomes are affected by the Residency model and how it compares with the outcomes from the normal numeric assessment method that was previously used. We demonstrate the superiority of the Residency Model via a comparison of the performance of the students in various cohorts in a fourth year course (ENGTECH 4MA3 – Advanced Mathematics) prior to and after implementing the Residency Model. We find that the students educated through the Residency model have significantly higher competencies and skills compared to their peers who took the same course in a graded format. Specifically, assuming the equivalence of meeting competencies for the course at a numeric grade of 75% in both graded and Residency assessments, a greater proportion of students satisfactorily met competencies in the Residency Model compared to the normal graded style. Taken together, we show, using numeric grades as a proxy, the effectiveness of the Residency Model to promote student learning.

Keywords:

Engineering education, graceless learning, residency model, cohort comparison.