Editor’s note: The following is part of a resource collection called It’s Worth Discussing, in which we feature research articles that are especially suitable for personal reflection and group discussion with your colleagues.
Why is this article worth discussing: College doesn’t offer students much guidance or practice self-assessing. In college, teachers grade student work. Students don’t have the expertise or objectivity that accurate assessments require. But some teachers have explored approaches that develop students’ self-assessment skills and work around some of the self-grading issues. This article provides an example that honestly explores the successes and failures of one approach. It’s worth discussing because self-grading has potent benefits. It forces students to look at their work critically. It increases the chance that students will learn from their mistakes. Self-grading experiences prepare students for the self-assessment activities their futures will likely hold.
Chang, K., & de Lemos Coutinho, L. (2021). Administering homework self-grading for an engineering course. College Teaching. Advance online publication. https://doi.org/10.1080/87567555.2021.1957666
Junior students in an engineering course self-graded homework problem sets. They submitted a copy of completed problems before the teacher posted solutions. Students resubmitted their problem sets after using a rubric to grade their solutions. The instructor checked the student grading. For purposes of the study, the instructor graded the problem sets before seeing the student grades so that students could compare assessments of both.
Results showed that poorly performing students generally calculated significantly higher scores than those given by the instructor. Score differences between better-performing students and the instructor were smaller. Results also indicated no measurable correlation between the order of the problem sets and the scoring differences of the sets. In other words, students’ scores did not move closer to instructor scores across the seven problem sets.
Self-assessment, whether it involves self-grading or a more general self-reflection, has the potential to garner many benefits. Here are some the authors discuss:
“Students recognize that they can learn from their own mistakes, and self-assessment can promote self-reflection, problem-solving, and more responsibility for learning.”
“Students develop a better understanding of what they already know and what they need to learn with increased self-judgment. This awareness increases student learning by developing self-feedback habits.”
“Students may feel a greater sense of ownership with their learning process when they actively participate in a process that uniquely becomes their own.”
The problems with self-grading jeopardize the benefits. Students are strongly motivated to give themselves a grade closer to the one they want than the one their work has earned. And that’s not the only problem.
“Assessment activities require students to understand the evaluation criteria, acknowledge the level of effort and perceived task difficulty, and provide the appropriate feedback needed to improve.”
“Students benefit when the instructor identifies strengths in their solutions to problems, to confirm their understanding, and when the instructor flags errors in their solutions, to enable them to learn from their mistakes.”
“When students choose to copy a solution from either a classmate or from the solution key, or have little interest in mastering their understanding of the material, the intended purpose will not be served.”
Sadler and Good, whose work the article cites, “point out that multiple-choice and fill-in the blank problems result in more consistent outcomes.”
“The instructor and student must share assessment responsibilities and the instructor will rigorously review student activities and warn students when anomalies occur.”
“To be effective, self-assessment criteria should be understandable, measurable, realistic, and relevant to the outcomes.” (The authors cite references to support this claim.)
Students provided feedback on this self-grading experience. Using a Likert scale, with 5 being favorable and 1 unfavorable, the mean response was 3.63, SD 1.37. Students provided additional written comments, which contained both positive and negative reactions. One student wrote, “It really didn’t help me learn more because I mostly just looked if I got the answer right and if I did . . . I moved on.” The researchers write that some “some students did not embrace this opportunity.”