Self-grading in a project-based software engineering course

Abstract

The Web Application Design and Development course offered at Rensselaer at Hartford uses a project-based approach where students construct a web application of their own choosing. For the past two years, the instructor of this course has used a student self-grading approach where, as part of defining the requirements for the project, students also define grade specifications for project grades of 'A', 'B', and 'C'. This paper discusses the motivation and approach to self grading and reports on a survey-based study used to determine student attitude toward the self-defined project and the self-grading approaches. Results of the survey indicate that students are moderately satisfied with the self-grading approach when used in conjunction with the student-defined project.

Self-grading

“…The self-grading approach has been employed in eight offerings of a graduate web application design and development course [1,2], including two distance offerings. Results indicate that the approach definitely and positively supports self-directed learning, contributing to a foundation of life-long learning.…”

Spanning the gap between software engineering instructor and student

“…A number of these courses are project-based where teams of students complete a project under the guidance of the instructor and possibly one or more teaching assistants (TAs) [1][2][3][4]. Typical student projects range from applications with real clients [5] to modifying open source software [6] to service-learning experiences [7].…”

Self-Corrected Homework for Incentivizing Metacognition

“…This is a question that has been addressed previously in the literature in other disciplines 4,5,6,7,8,9,10,11 where it has been demonstrated repeatedly that most -but not all -students tend to assign themselves slightly lower grades than instructors, and we seek to compare our results with others. To do so, we graded a photocopied version of each student's initial homework submission without the students' knowledge; meanwhile, the students used an answer key and grading rubric to grade their own initial homework submission, which they subsequently resubmitted, and the results were compared.…”

“…While previous studies have investigated the use of self-grading 4,5,6,7,8,9,10,11 and self-correcting 12,13 , most of them have left traditional incentive mechanisms in place. This work considers an alternate incentive mechanism with the aim of encouraging students to attempt problems on their own while removing the incentive to plagiarize.…”