Love ’em or hate ’em, student evaluations of teaching (SETs) are here to stay. Parts <a href="https://www.teachingprofessor.com/free-article/its-time-to-discuss-student-evaluations-bias-with-our-students-seriously/" target="_blank"...
Not if grades are involved, would be the likely answer of most faculty. The need for good grades does cloud student objectivity. But what that doesn't change is the fact that the ability to accurately assess your work contributes much to learning experiences in college and it's a virtual necessity in professional life. David Boud (and two coauthors) report that it's not a skill that's taught explicitly in most curricular programs. Rather, it's something we assume students pick up on their own and without instruction.
Considerable research has been done on self-assessment (much of it completed by Boud, who has spent much of his career working in this area). He explains that not only do we fail to teach self-assessment skills but also “The capacity to make judgments is not well represented in many current assessment practices. Assessment items are often strongly knowledge-based, with criteria unilaterally set by teachers. The role of students tends to be to offer themselves to be assessed by others.” (p. 942) In general, education encourages students to depend on the judgments of others. They come to believe there's no need to assess their own work. Others will do that for them.
Moreover, good self-assessment skills don't develop quickly with one or two opportunities to try doing it. “We assume that the key feature of the development of judgment, like any other kind of expertise, is that it requires consistent engagement over time.” (p. 943) That makes sense given the complexity of the skills involved. Boud and his coauthors say those skills develop when students consistently make judgments guided by explicit criteria. Then their assessments must be compared with those given by others, either by experts, such as teachers, or by peers. Students must explore the reasons their self-assessments are not the same as those given by others. What are the reasons behind their incorrect assessments? What did they miss that others saw?
Out of these background issues emerged the four questions explored in this study: (1) Do the grades students give themselves agree with teacher assessments? (2) Do the differences between self-assessments and those given by others decrease as students do more self-assessment? (3) Does the overall performance of a student affect his or her ability to self-assess? and (4) Does the ability to accurately self-assess lead to improved performance?
The research team had an interesting opportunity to explore these questions in the context of an undergraduate degree program in design offered at an Australian university. A software program called ReView gave students criteria-based feedback and comments. It included a self-assessment option. Students could use assignment criteria (samples are included in the article) to assess their work (which included group projects, research reports, oral presentations, critical and reflective essays, and individual portfolios). They could then submit their assessment and see the teacher's evaluation of their work. The teacher graded the student's work without seeing how the student assessed it. Use of this self-assessment component was voluntary, but the study looked at its use across individual courses as well as across various courses within the program. For the study, they looked at almost 2,200 self-assessments from 182 students.
In the results, they found significant disagreement between teacher and student self-assessments on first tasks, with students often rating themselves higher than the teacher did. But those differences diminished as students completed more self-assessments. Here's how researchers describe these results. “Although students may initially struggle to accurately self-assess, with time and benchmark scores from their tutor [think teacher], they appear to get more accurate.” (p. 950) The same increase in accuracy was seen within a course and across several of them. However, when students started a new course, their first self-assessments were always less accurate than subsequent ones.
Academic achievement levels (measured by grades received on these assignments) did make a difference. Consistent with previous research, low-achieving students overestimated their performance and high achievers underestimated theirs. But it was the group in the middle that “[was] the most able of the three groups in developing self-assessment skills in this context.” (p. 951) The difficulty of those low achievers persisted when researchers grouped the students by whether they were accurate estimators, underestimators, or overestimators. “Over estimators, who tend to be poor achievers, do not appear to learn how to improve their performance over time.” (p. 951) “It may be that this group is content to merely pass each task and has no desire to invest the effort to do better, or they may not have the capability to improve without additional educational interventions.” (p. 952)
“Notwithstanding that this is an initial study with incomplete data biased towards students enthusiastic in seeking to judge their own performance, there are interesting pointers to phenomenon that if confirmed would have quite substantial pedagogic implications.” (p. 954)
Reference: Boud, D., Lawson, R., and Thompson, D.G. (2013). Does student engagement in self-assessment calibrate their judgment over time? Assessment & Evaluation in Higher Education, 38 (8), 941-956.